Polycationic Polymer-Regulated Assembling of 2D MOF Nanosheets

Subscriber access provided by University of Birmingham

Article

Polycationic Polymer-Regulated Assembling of 2D MOF Nanosheets for High-Performance Nanofiltration Huixiang Ang, and Liang Hong ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b08383 • Publication Date (Web): 28 Jul 2017 Downloaded from http://pubs.acs.org on July 28, 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 34

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

Polycationic Polymer-Regulated Assembling of 2D MOF Nanosheets for High-Performance Nanofiltration Huixiang Ang,† Liang Hong*,† †

Department of Chemical and Biomolecular Engineering, National University of Singapore,

Singapore 117585, Singapore. ABSTRACT: Herein, two-dimensional metal-organic framework (2D MOF) made of iron porphyrin complex (TCP(Fe)) interconnected with divalent metal ion (M = Zn, Co, Cu) is used to construct a selective layer, which is explored as an ultrafast and energy-saving nanofiltration (NF) membrane for removing organic dyes from water. Among the layered 2D M-TCP(Fe) membranes, Zn-TCP(Fe) membranes display the highest water permeance, which is 3 times higher than graphene-based membranes with similar rejection. To further improve the separation performances, we utilize polycations to anchor the periphery carboxylic groups of nanosheets, regulating the assembly of 2D Zn-TCP(Fe) nanosheets to produce a new class of crack-free selective layer possessing ultrathin and highly ordered nanochannels for efficient NF. Benefiting from these structural features, our polycationregulated 2D Zn-TCP(Fe) membranes could offer ultra-high permeance of 4243 L m-2 h-1 bar1

(2-fold higher than its pristine) and excellent rejection rates (over 90%) for organic dye with

size larger than 0.8 × 1.1 nm. This permeance value is about 2 orders of magnitude higher

1 ACS Paragon Plus Environment


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

Page 2 of 34

than the commercial polymeric NF membrane. Additionally, the membranes demonstrate 20−40% salt rejection. KEYWORDS: two-dimensional

·

metal-organic-framework

·

polycation

·

assembly

·

nanofiltration.

INTRODUCTION Over the last few decades, the membrane-based separation using NF technique has demonstrated increasingly essential to substitute the conventional thermal separation techniques (i.e. distillation and evaporation) due to their high energy-efficiency and costeffectiveness.1-2 However, there are several limitations such as the high operating pressure, poor stability, and low water permeability. For example, the commercial NF membrane made of polyethersulfone suffers from slow water permeance (ca. 10 L m-2 h-1 bar-1) because partly it operates at high pressure (up to 40 bars), leading to large energy consumption and subsequently high operational cost.3-4 Typically, NF is governed by: (1) physical adsorption5 or chemical interaction (i.e. cation-π interaction)6 between solute molecules and the surfaces of mass transport channels; (2) molecular sieving controlled by size exclusion;7 and (3) Gibbs-Donnan effect (i.e. electrostatic repulsion)8 between charged solutes and charged groups anchored to the membrane. These mechanisms are interrelated and the last two modes are more significant to NF processes.9-10 The use of 2D nanosheets as elementary block to construct membranes is emerging as a new-generation material for water purification by NF because their unique intrinsic structures (i.e. in-plane pores/defects and interlayer spacing) could render effective nanosieving, enabling selective transport of molecules to permeate through the nanochannels of the scaffold.11-13 Previous studies on 2D membranes for NF are primarily focused on the graphene family,14-20 while less attention has been paid on the exfoliated dichalcogenides,21-23


Page 3 of 34

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


MXenes,24-25 and metal hydroxides.26 The application of these 2D materials for NF is often limited by low-density perforation and uneven distribution of interplane distance. While many attempts have been made in developing chemical or physical approaches to control the interlayer spacing distance of 2D nanosheets to improve the separation performances.17, 27-28 Nevertheless, those methods are strenuous to obtain narrow-sized interplane distance and nano-sized thick film for both excellent separation and high permeability. For instance, the intrinsic electrostatic repulsive force between the anionic oxygenated group of graphene oxide (GO) nanosheets leads to low stacking density, leaving large interlayer gaps for probe solutes to passage through the selective layer.29-30 As a result, more and more 2D nanosheets are added onto this existing selective layer to compensate the rejection rates at the cost of water permeance. Additionally, 2D MOF material is proposed to be an alternative candidate for making efficient NF membrane because it possesses plentiful and regular in-pane nanopores, which cannot be found in the-state-of-the-art, for example, graphene-based membrane. For now, 2D MOF membranes are mainly used for gas separation,31-34 but reports on water separation are truly rare, for the reason that most of the MOF materials are waterunstable. Recent studies have demonstrated that MOF-like membrane constructed from zeolitic imidazolate framework (i.e. ZIF-8) are water-stable and thus they are intensively scrutinized for water purification via NF. However, the formation of these ZIF-8 crystallites are often randomly oriented such that a thick selective layer of about 1 µm is required to achieve desirable rejection rates.35-36 In this study, we select TCP(Fe) complex to construct nano-sized thick 2D MTCP(Fe) (M = Zn, Co, and Cu) selective layers for NF of aqueous solution because it possesses superior structural stability over most of the metalloporphyrin complexes, and the resulting M-TCP(Fe) nanosheets exhibit strong stacking ability.37-38 Our preliminary results show that various metal nodes in the 2D M-TCP(Fe) selective layers have different effects on



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

Page 4 of 34

the water permeance and the one that contains Zn node exhibits the highest water permeability. Further enhancement of separation performances can be achieved when polycations, such as polyethylenimine (PEI) and poly(diallyldimethylammonium chloride) (PDDA) are used to crosslink the periphery carboxylic groups of 2D Zn-TCP(Fe) nanosheets, which regulate the assembly of 2D MOF nanosheets, resulting in the formation of ultrathin (thickness: ca. 48 nm), highly ordered, and continuous film. The as-prepared polycationregulated 2D Zn-TCP(Fe) membrane presents an ultrafast water permeance up to 4243 L m-2 h-1 bar-1 (2-fold higher than its pristine) with a rejection rate over 90% for small organic dye (i.e. 0.8 × 1.1 nm) and 20−40% salt rejection at a low operating pressure of 0.01 bar. For the first time, such superior nanofiltration performances at such a low operating pressure are obtained as compared to the literature (Table S1, Supporting Information).

RESULTS AND DISCUSSION The fabrication of polycation-regulated 2D Zn-TCP(Fe) membrane is illustrated in Figure 1. To illustrate our synthesis procedure, the Zn-TCP(Fe) MOF, is used as an example. First, 2D Zn-TCP(Fe) nanosheets were prepared by using the soft-template assisted methods.37 The ultrathin 2D Zn-TCP(Fe) nanosheet is made up of TCP(Fe) complex interconnected with the Zn2+ ions, forming periodic porous framework with anionic carboxylic pendant group at the periphery of nanosheets. Subsequently, the 2D Zn-TCP(Fe) nanosheets suspension was added into the polycationic polymer solutions, where the concentrations of PEI and PDDA (0.1 wt.%) are well below the reported overlap concentrations of polymer coils of these two polymers in aqueous medium,39-40 and hence both polymer molecules are presented as independent coils that have the radius of gyration (Rg) for molecular mass of 105: PEI = 16 nm and PDDA = 24 nm. Besides, 2D Zn-TCP(Fe) sheet has roughly an average lateral length of 1.2 µm. As this is substantially larger than Rg of either polycationic polymer, stacking


Page 5 of 34

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


interactions drive regulated assembly of 2D nanosheets must be the sole manner. In this polycation-regulated assembly of nanosheets, the peripheral edge of each nanosheet, bearing free carboxylic groups, is ionically crosslinked by polycation coils (i.e. PEI or PDDA). We propose that the assembly of 2D Zn-TCP(Fe) nanosheets is driven by a combination of molecular attractive forces (i.e. polycation interacts with anionic 2D MOF nanosheets) and compressive force (i.e. pressure drop via vacuum filtration), as depicted in Figure S1 (Supporting Information). As a result, the polycation-regulated 2D Zn-TCP(Fe) assemblies were deposited on ultrafiltration support (i.e. nylon) to form a selective layer for molecular separation. For comparison purpose, similar 2D M-TCP(Fe) (M = Cu and Co) nanosheets were synthesized by replacing the Zn node with Cu and Co nodes, respectively (see the Experimental Section for details). The obtained 2D M-TCP(Fe) (M = Zn, Cu, and Co) MOFs were characterized by X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), field emission scanning electron microscopy (FESEM), atomic force microscopy (AFM), transmission electron microscopy (TEM) image, and energy-dispersive X-ray spectroscopy (EDX). The XRD pattern (Figure 2a) recorded from the as-prepared Zn-TCP(Fe) nanosheets resemble to that of the previously reported by Zhang et al.,37 indicating that we have successfully synthesized Zn-TCP(Fe) MOF. The chemical structure of the Zn-TCP(Fe) nanosheets was further examined by XPS analysis (Figure 2b-f). The XPS C 1s spectrum shows the peaks at binding energies of 284.3, 285.3, 286.3, and 287.7 eV, which are attributed to the C=C, C−C/C−H, C−N/C−O, and C=N/C=O bonds, respectively. This series of carbon species matches the carboxylic and pyrrolic functional groups in Zn-TCP(Fe) nanosheets (Figure 2b). The XPS spectra of the N 1s and Fe 2p regions reveal that the signals of the Fe−N (399.6 eV), pyrrolic (400.3 eV), Fe0 2p3/2 (707.0 eV), Fe-N 2p3/2 (711.9 eV), Fe0 2p1/2 (720.1 eV), and Fe−N 2p1/2 (724.3 eV) are the main features ascribed to the TCP(Fe) complex (Figure 2c-d,). The XPS spectra of O 1s and Zn 2p core-levels reveal the signals of



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

Page 6 of 34

the Zn−O (530.1 eV), Zn 2p3/2 (1021.8 eV), and Zn 2p1/2 (1044.8 eV) corresponding to Zn2(COO)4 nodes (Figure 2e-f). In addition, the signals of the C=O (531.1 eV) is larger than that of the C−O/O−H signal (532.0 eV), indicating the successful coordination of Zn2+ ion with the carboxylic group on TCP(Fe) complex to build Zn-TCP(Fe) MOF (Figure 2e). The FESEM, AFM and TEM images prove the obtained Zn-TCP(Fe) nanosheets possess a lateral size in micrometer scale (Figure 3a-c). The thickness of Zn-TCP(Fe) nanosheets was examined by AFM, giving an average thickness of 1.9 ± 0.1 nm (Figure 3b and Figure S2, Supporting Information). The TEM image (Figure 3c) shows that the as-synthesized ZnTCP(Fe) MOF is composed of sheet-like structure. The selected area electron diffraction (SAED) patterns (Figure 3d) collected along the [001] axis gives diffraction spots, which are attributed to (110) and (100) planes of Zn-TCP(Fe) nanosheets. After replacing the Zn node with Cu and Co nodes, the XRD patterns (Figure S3, Supporting Information) and XPS spectra (Figure S4 and S5, Supporting Information) of the Cu-TCP(Fe) and Co-TCP(Fe) nanosheets show the same crystal characteristic with the as-synthesized Zn-TCP(Fe) nanosheets. In addition, the SEM and AFM images reveal that the sheet-like Cu-TCP(Fe) and Co-TCP(Fe) samples with lateral dimensions in micrometer scale with average thicknesses of 2.2 ± 0.2, and 2.0 ± 0.1 nm, respectively (Figure S6, Supporting Information). Moreover, the XPS measurements (Figure S7, Supporting Information) and inductively-coupled plasma optical emission spectroscopy (ICP-OES) confirmed that the molar ratios of M (M = Zn, Cu, and Co) to Fe in the respective M-TCP(Fe) MOFs are comparable and they are estimated to be 2:1. All the above results suggest that we have successfully synthesized M-TCP(Fe) MOFs (M= Zn, Cu, and Co) nanosheets. Based on the nonlocal density functional theory (NLDFT), the cumulative pore volume plots (Figure 4a and Figure S8a-b, Supporting Information) indicate the presence of microporosity in 2D M-TCP(Fe) nanosheets and the pore size distribution plots (Figure 4b and Figure S8c-d, Supporting Information) show a dominant


Page 7 of 34

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


pore size range of 1.20 to 1.24 nm, which is consistent with the value of 1.18 nm based on crystallographic data (Figure S9a, Supporting Information). The 2D MOF membranes comprises of M-TCP(Fe) selective layers prepared on porous nylon support with pore size ca. 0.2 µm (Figure S10, Supporting Information). Briefly, 0.4 mg mL-1 of the diluted 2D M-TCP(Fe) dispersion was used to form a selective layer on nylon support through vacuum filtration. The SEM image of Zn-TCP(Fe) selective layer shows that the surface morphology of the selective layer exhibits a few wrinkles around the surface (indicated by white arrows in Figure 5a), while the photographs (Figure S11, Supporting Information) show that the resultant Zn-TCP(Fe) layer (ca. 3.5 cm in diameter) is evenly coated on the nylon support (4.7 cm in diameter). The average thickness of the ZnTCP(Fe) selective layer with mass loading of 80 mg m-2 was measured to be about 70 nm based on the cross-sectional SEM image (Figure 5b). The EDX mapping images of the ZnTCP(Fe) selective layer show a uniform compositional distribution of Zn, Fe, N, O, and C elements on the selective layer (Figure 5c). The AFM analysis (Figure 5d-e) shows that the thickness of the Zn-TCP(Fe) layer is 67 ± 2.5 nm, which is in reasonably good agreement with the cross-sectional SEM image (Figure 5b). For comparison, the surface morphologies and elemental distributions of Cu-TCP(Fe) and Co-TCP(Fe) selective layers of the membranes were also examined using SEM and EDX analyses (Figure S12a-b, Supporting Information). The separation performances of the pristine Zn-TCP(Fe), Cu-TCP(Fe), and CoTCP(Fe) membranes were then evaluated, respectively, by using a stirred vacuum filtration setup (for low-pressure < 1 bar) and a dead-end filtration device (for high-pressure ≥ 1 bar), as shown in Figure S13 (Supporting Information). All the separation experiments were stirred at 500 rpm to minimize the concentration polarization effect during filtration. The permeances and the rejection rates were recorded after a steady flux was reached at ca. 1 to 2 7 ACS Paragon Plus Environment


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

Page 8 of 34

h for every experiment, while an operating pressure of 0.01 bar was applied unless otherwise stated. For comparison, the rejection percentage of the methyl red (MR, an electroneutral molecule) through nylon support was examined to be near zero, indicating negligible amount of rejection rate contributed by the ultrafiltration support (Figure S14, Supporting Information). The separation performances of pristine 2D M-TCP(Fe) membranes as a function of their membrane thickness were first evaluated by measuring the water permeances and rejection percentages of the MR under 0.01 bar (Figure 6a-b). In general, the water permeances of the pristine 2D M-TCP(Fe) membranes decline and the rejection rates enhance with increasing membrane thickness. This is expected because the stacking of more 2D nanosheets slowdowns the stream flux but improves the ability to reject smaller solute molecules. The optimum selective layer thickness required to achieve a high water permeance and a rejection cut-off at 90% is identified to be ca. 65−70 nm for all three pristine 2D M-TCP(Fe) (M = Zn, Cu, and Co) membranes. The pristine Zn-TCP(Fe), Cu-TCP(Fe), and Co-TCP(Fe) membranes with optimized thickness present water permeances of 2120, 1679, and 1819 L m-2 h-1 bar-1 and corresponding rejection rates of 98%, 93%, and 94% for MR solute, respectively. The pristine Zn-TCP(Fe) membrane shows the highest water permeance with similar rejection rate as compared with the rest of the M-TCP(Fe) membranes. Subsequently, the impact of pressure on permeation was investigated. The slope of the plot (Figure 6c) is inversely proportional to the viscosity of the fluid (i.e. water) flow according to the Hagen−Poiseuille equation.14 The pristine Zn-TCP(Fe) membrane with the steepest slope indicates that the water filtration through the pristine Zn-TCP(Fe) membrane is the least viscous as compared to the other two pristine M-TCP(Fe) (M = Cu and Co) membranes. In addition, this test manifests that the as-fabricated membranes could withstand pressure at least up to 5 bar.


Page 9 of 34

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


To further understand the separation efficacy of the pristine M-TCP(Fe) membranes, four organic dyes with different charges and molecular sizes were used to probe the separation performances of the membranes. They are methyl red (MR), methylene blue (MnB), methyl orange (MO), and brilliant blue G (BB) and their chemical properties (i.e. dimensions, surface charge distribution, molecular weight and formal charge) of the molecular solutes are summarized in Table S2 (Supporting Information). Typically, MR has zero formal charges with dimensions of 0.8 × 1.1 nm; MnB is positively charged with dimensions of 1.4 × 0.9 nm; MO is negatively charged with dimensions of 1.0 × 1.2 nm; BB is negatively charged with dimensions of 1.6 × 1.9 nm. Figure 6d-f show that the three pristine 2D M-TCP(Fe) membranes could achieve rejection rates greater than 90% for MR, MO, and BB dyes. In addition, the pristine Zn-TCP(Fe) membrane exhibits apparently higher water permeances for the four aqueous dye solutions as compared to the other two membranes. Additionally, by subtracting the amount of a solute in the retentate and permeate from that in the feed on a fixed volume basis, less than 5% of the solute molecules, regardless of their traits, are found to stay in the membranes. Furthermore, the absorbance of a dye in the accumulative retentate solution is higher than the absorbance of feed solution further verifies that most dye molecules is blocked besides minor amount entrapped in the membranes (Figure S15-S17, Supporting Information). Notably, the pristine M-TCP(Fe) (M = Zn, Cu, and Co) membranes of the same cohort display MnB rejection rates below 50%. Based on the zeta-potential analysis (Table S3, Supporting Information), the M-TCP(Fe) selective layers bear negative surface charge, which induces opposite charged attractive forces, driving the MnB solute to slip through the nanochannels of the membrane. To overcome this problem, polycations (i.e. PEI or PDDA) were introduced into pristine Zn-TCP(Fe) nanosheets dispersion with an approximately equivalent molar ratio during the assembly process via vacuum filtration. Subsequently, the resultant polycation-



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

Page 10 of 34

regulated 2D Zn-TCP(Fe) membranes (i.e. PEI/Zn-TCP(Fe) and PDDA/Zn-TCP(Fe)) were washed several times with pure water and dried under vacuum before further characterizations. The polycation-regulated 2D Zn-TCP(Fe) selective layers were first examined by XPS measurements. The XPS spectra of N 1s core-level show that the PEI/ZnTCP(Fe) selective layer contains amine functional group at binding energy of 398.5 eV (Figure 7a) while the PDDA/Zn-TCP(Fe) selective layer possesses quaternary nitrogen species at binding energy of 402.0 eV (Figure 7b). These results validate that we have successfully crosslinked the polycationic polymers with anionic Zn-TCP(Fe) nanosheets. The FTIR spectra (Figure S18, Supporting Information) further enrich the characterization and demonstrate that PEI and PDDA are combined with periphery carboxylic groups of 2D MOF but not block the pores of 2D MOF. Additionally, the XRD patterns (Figure 7c-d) show that the (002) peak of pristine Zn-TCP(Fe) selective layer locates at 9.1o, whereas both the PEI/Zn-TCP(Fe) and PDDA/Zn-TCP(Fe) selective layers are at 9.5o. This suggests that the average interlayer spacing distance (d) is slightly narrowed from 0.97 to 0.93 nm, which is attributed to the crosslinking effect of polycation with anionic 2D Zn-TCP(Fe) nanosheets. These interplane distances are close to the crystallographic data (Figure S9b, Supporting Information) which agrees with the literature value.37 In addition, the full width at half maximum (fwhm) values of polycation-regulated Zn-TCP(Fe) selective layers are lower than its pristine, indicating the former is more orderly packed. To determine the optimum thickness for efficient separation processes, we first evaluate the rejection rates of MR solute (neutral charge) and water permeances against various selective layer thickness for PEI/Zn-TCP(Fe), PDDA/Zn-TCP(Fe), and pristine ZnTCP(Fe) membranes (shown in Figure 8a-b). The polycation-regulated 2D Zn-TCP(Fe) membranes (optimum thickness: ca. 48 nm) display 2-fold enhancement in water permeance of pristine Zn-TCP(Fe) membrane (optimum thickness: ca. 67 nm) with rejection cut-off at


Page 11 of 34

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


90%. These results suggest that the optimum thickness of the selective layer can be reduced by ca. 30% with the use of polycation-regulated 2D Zn-TCP(Fe) selective layers. Based on the optimum thickness of the selective layers, we compare the separation performances of polycation-regulated 2D Zn-TCP(Fe) membranes with pristine Zn-TCP(Fe) membrane using various charges and sizes solutes (i.e. MnB, MR, MO, and BB in Figure 8c). It turns out that there is a significant improvement in rejection rate, especially for MnB solute, when polycation-regulated 2D Zn-TCP(Fe) membranes are used. In addition, these polycationregulated membranes demonstrate up to ~2 times higher water permeance than that of its pristine with similar rejection rates for MR, MO and BB solutes. Notably, the polycationregulated Zn-TCP(Fe) membranes also show a significant improvement in rejection rates for monovalent and divalent salts (i.e. MgCl2, Na2SO4, NaCl, and MgSO4) as compared to its pristine (Figure S19, Supporting Information), indicating that these membranes are suitable to be used for nanofiltration. These results suggest that not only physical size sieving plays a part in separation process, but the electrostatic effect is also a crucial factor for blocking charged solutes. Additionally, higher absorbance in the retentate solution suggests that most of the solutes are blocked instead of being absorbed by the PEI/Zn-TC(Fe) and PDDA/ZnTCP(Fe) membranes (Figure S20-S21, Supporting Information). As a demonstration of practical application, the durability of the PEI/Zn-TCP(Fe) and PDDA/Zn-TCP(Fe) membranes was evaluated by long-term filtration tests (under 0.01 bar) using BB dye as a probe solute and bovine serum albumin (BSA) as a fouling agent. Figure 9(a-b) show that the permeance BB feed levels off after 1.5 h, indicating the stability of the PEI/Zn-TCP(Fe) and PDDA/Zn-TCP(Fe) membranes. In addition, only a slight drop of rejection rate by less than 1% was found after 8 h of prolonged nanofiltration, for which a clear filtrate can still be observed in the end (see Figure 9a-b, inset). It is, therefore, suggested that the polycations-regulated 2D Zn-TCP(Fe) assemblies are performance reliable to



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

Page 12 of 34

construct a membrane for the separation of this type of dye from industrial effluent. To further evaluate the robustness of the PEI/Zn-TCP(Fe) and PDDA/Zn-TCP(Fe) membranes, a standard fouling reagent (BSA) was used to examine the antifouling property of the membrane. As known, BSA is a well-recognized fouling agent, which has been used for the study of the antifouling property of membranes.16 In this antifouling evaluation, the filtration test was first carried out by passing pure water through the polycation-regulated Zn-TCP(Fe) membranes followed by the standard aqueous solution of BSA foulant. After fouling, the membrane was washed with chemical solvents (i.e. ethanol, acetone, and water) to remove the unadsorbed foulant on the membrane before carrying out filtration test using pure water again to complete one cycle of the antifouling test. Such cycle was repeated two more times, as depicted in Figure 9(c-d). Such a simple cleaning procedure using chemical solvent could recover more than 90% of the flux after every cycle of antifouling test. It is worth noting that the additional of polycation into the Zn-TCP(Fe) membrane further increases the hydrophilicity of the membrane (see Table S3, Supporting Information) and thus decreases the interaction with the organic BSA solute, resulting in the enhancement of flux recovery percentage by 3% when compare to its pristine membrane (Figure S22, Supporting Information). After three cycles of fouling test, the SEM images and EDX mapping images show that the morphology and elemental distribution of the PEI/Zn-TCP(Fe) and PDDA/ZnTCP(Fe) selective layers are well retained after long-term antifouling test for 10 h (Figure S23a-b, Supporting Information). The permeance versus operating pressure plot of the MTCP(Fe)-based membranes is compared with the commercial nanofiltration membrane (NFM) and some advanced membrane materials (i.e. carbon/polymer, ceramic/inorganic, and ZIF-8-based MOF), as shown in Figure 9(e-f). Since the volumes of the permeate reported in previous studies were inconsistent, a practical volume of 1000 L was used for demonstration to calculate energy consumption using the energy density relation: E (kJ L-1) =

∆P ×VP , VP 12

ACS Paragon Plus Environment

Page 13 of 34

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


where ∆P is the pressure drop across the membrane and VP is the volume of the permeate.3 All the data collected are based on a rejection cut-off at 90%, as summarized in Table S1 (Supporting Information). At a low operating pressure of 10-2 bar (which is equivalent to 10-3 kJ L-1), the M-TCP(Fe)-based membranes exhibit water permeances as high as 1679−4243 L m-2 h-1 bar-1 (Figure 9e). Such membranes could provide much higher permeances than that of the existing carbon/polymer membranes (17−720 L m-2 h-1 bar-1), ceramic/inorganic membranes (3−1084 L m-2 h-1 bar-1), ZIF-8-based MOF membranes (2−38 L m-2 h-1 bar-1), and commercial NFM (10 L m-2 h-1 bar-1), while the operating pressure is 3 orders of magnitude lower than that of the commercial NFM (40 bar). These capabilities propose the 2D M-TCP(Fe)-based nanosheets to be a promising type of membrane material for nanofiltration device with both high-energy and high separation efficiencies. To comprehend the mechanisms resulting in the excellent water permeability, the contact angles (i.e. hydrophilicity) of the 2D M-TCP(Fe) (M = Zn, Cu, Co), PEI/Zn-TCP(Fe), and PDDA/Zn-TCP(Fe) membranes were characterized and summarized in Table S3 (Supporting Information). As known, the M-TCP(Fe) MOFs bear free carboxylic pendant groups. The survey scan of XPS spectra (Figure S7, Supporting Information) show that the Zn-TCP(Fe) MOF contains approximately 3 times higher O/C ratios as compared to CuTCP(Fe) and Co-TCP(Fe), implying that the Zn-TCP(Fe) MOF possesses more oxygencontaining groups (i.e. COOH) along the periphery of nanosheets. Among the M-TCP(Fe) membranes, the Zn-TCP(Fe) membrane has a smaller contact angle than the other two MTCP(Fe) (M = Cu and Co) membranes. It is rational that a higher density of the edge carboxylic groups carried by the Zn-TCP(Fe) nanosheets is responsible for this high hydrophilic ability. This good wetting ability of the membrane provides multiple hydrophilic sites for water to access, and thus enhanced the water transportation.18 After polycationregulated coupling of the Zn-TCP(Fe) nanosheets with polycations (i.e. PEI or PDDA), the



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

Page 14 of 34

as-obtained membranes can be completely wetted, indicating the as-obtained membrane is superhydrophilic. In addition, XRD measurements were employed to elucidate the packing efficiencies of the polycations-regulated 2D Zn-TCP(Fe) selective layers versus its pristine one. The XRD patterns (Figure 7c-d) reveal that when we increase the molar ratio of PEI:ZnTCP(Fe) from 0:1 to 1:1, the average interlayer spacing distance decreases by 4%. This result indicates that the nanosheets become more closely packed when crosslinked by polycations, and hence the resulting 2D MOF selective layer forms a continuous and crack-free film, as depicted in Figure S24 (Supporting Information), leading to the enhancement of permeances and rejection rates (Figure 8c). This hypothesis can be evidently supported by SEM images (Figure S25, Supporting Information), showing that cracks can be found on the pristine ZnTCP(Fe) membrane but not on the PEI/Zn-TCP(Fe) membrane owing to similar selective layer thickness. This observation also supports the fact that coils of polycation of each nanosheets assembly physically entangle together due to their soft nature, which helps attenuate stress imposed from changes of pressure. On the other hand, when we increase the molar ratio of PEI:Zn-TCP(Fe) from 1:1 to 4:1, the average interplane distance increases by 5% (Figure S26a, Supporting Information), which might attributed to the occupying of interlayer spacing by the extra amount of polycation. As a result, more Zn-TCP(Fe) nanosheets are required to achieve high rejection rate (i.e. cut-off at 90%), but the permeance retards as the thickness of the film increases (Figure S26b-c, Supporting Information). These findings reveal that indeed polycation plays an important role in the assembly of anionic 2D sheets to regulate the interlayer spacing of the nanosheets, which opens prospect for excellent separation applications. To understand the effect of periodic pore system over aperiodic pore system in NF processes, we have fabricated reduced graphene oxide (rGO) for comparison purpose. Based on our experimental study, for selective layers with thicknesses in the range of 60−70 nm, the minimum pressure required for water permeation through Zn-TCP(Fe)


Page 15 of 34

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


membrane is 0.01 bar, which is lower than that of the rGO membrane at 0.56 bar (Table S4, Supporting Information). This could be explained by the mass transport pathway as illustrated in Figure S27 (Supporting Information). The periodic pore system of the asobtained Zn-TCP(Fe) membrane provides similar path distances for water permeation, resulting in less energy loss and thus lowered the operating pressure for superior separation performances (Figure 9e-f).

CONCLUSIONS In summary, we have utilized polycationic polymers (i.e. PEI or PDDA) to regulate the assembling of 2D Zn-TCP(Fe) nanosheets through the crosslinking between polycation coils and the periphery carboxylic groups of nanosheets. This combination produces a novel type of polycation-regulated 2D MOF selective layers, owing ultrathin (thickness: ca. 48 nm), crack-free, and well-aligned channel for efficient water purification by NF. This is the first lamellar nanofiltration membrane constructed by polycation-regulated 2D MOF assemblies, which manifests excellent separation performances with high water permeance of 4243 L m-2 h-1 bar-1 and excellent rejection rate for organic dyes (i.e. 0.8 × 1.1 nm) and salt rejection at operating pressure of 0.01 bar. To the best of our knowledge, the superior separation performance at such a low operating pressure of 0.01 bar (which is equivalent to low energy consumption) is the best record so far. Moreover, the Zn-TCP(Fe)-based membranes exhibit very stable water permeance with a recovery up to 90% through a prolonged fouling operation. This work might provide a showcase for the design of industrially viable membrane with robust and periodic porous framework for water purification, and inspire the development of next-generation membrane for other solvent filtration processes.



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

Page 16 of 34

EXPERIMENTAL SECTION Materials. Zinc nitrate hexahydrate (Zn(NO3)2⋅6H2O, 98%), cobalt nitrate hexahydrate (Co(NO3)2⋅3H2O, 99.9%), copper nitrate trihydrate (Cu(NO3)2⋅3H2O, 99%), polyvinylpyrrolidone (PVP, average mol wt 40,000), pyrazine (C4H4N2, 99%), trifluoroacetic acid

(CF3COOH,

99%),

poly(ethyleneimine)

poly(diallyldimethylammonium

chloride)

(PDDA,

(PEI, Mw

Mw =

=

2.0−3.5

7.5 ×

×

105),

105),

N,N-

dimethylformamide (DMF, 99.8%), methyl orange (MO, ACS reagent grade), methyl red (MR, ACS reagent grade), methylene blue (MnB, 97%) and brilliant blue G (BB, ≥90%), bovine serum albumin (BSA, 98%), magnesium chloride (MgCl2, 98%), sodium sulfate (Na2SO4, ≥99%), sodium chloride (NaCl, ≥99%), magnesium sulfate (MgSO4, ≥99.99%), potassium bromide (KBr, 99.99%), and potassium chloride (KCl, 99.9%) were purchased from Sigma-Aldrich. Fe(III) meso-Tetra(4-carboxyphenyl) porphine chloride (TCP(Fe), 97%) was purchased from Frontier Scientific. Nylon (47 mm in diameter, 0.2 µm pore size) and anodisc circle with support ring (25 mm in diameter, 0.1 µm pore size) microfiltration supports were purchased from Whatman. Ethanol was purchase from Merck. The deionized water was obtained from the Milli-Q System. Synthesis of 2D MOF Nanosheets. All the starting materials were of analytical pure grade and used as received. 2D MOF nanosheets were prepared accordingly to previously reported soft-template assisted method.37 For the synthesis of Zn-TCP(Fe) nanosheets, 0.02 mmol of Zn(NO3)2⋅6H2O, 0.02 mmol of pyrazine and 40 mg of PVP were added into the 20 mL glass vial, containing 12 mL of the solvent mixture (3:1 volume ratio of DMF and ethanol). Then 0.01 mmol of TCP(Fe) dissolved in 4 mL of the solvent mixture (3:1 volume ratio of DMF and ethanol) were added to the above solution followed by sonicating the solution for 5 min. After that, the glass vial was placed in the oven and heated to 80 oC for 24 h. The as-synthesized Zn-TCP(Fe) sample was then washed several times with deionized 16 ACS Paragon Plus Environment

Page 17 of 34

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


water and ethanol by centrifuging at 10,000 rpm for 5 min. The final product was dispersed in deionized water for further characterizations and nanofiltration testing. The syntheses of CoTCP(Fe) and Cu-TCP(Fe) were similar, except that Co(NO3)2⋅3H2O and Cu(NO3)2⋅3H2O were used, respectively. Additionally, 20 μL of CF3COOH (1.0 M) was added into the CuTCP(Fe) growth solution. Fabrication of 2D MOF-Based Membranes. The 2D MOF-based membranes were prepared through filtering a diluted 2D MOF-based dispersion (ca. 0.40 mg mL-1) on the nylon support with pore size ca. 0.2 μm through vacuum filtration at 0.5 bar. For PEI/ZnTCP(Fe) and PDDA/Zn-TCP(Fe) membranes preparation, the as-obtained Zn-TCP(Fe) dispersion was stirred in 0.1 wt.% of PEI (Mw = 7.5 × 105) and PDDA (Mw = 2.0−3.5 × 105) aqueous solutions for 1 h, respectively, followed by vacuum filtration on nylon support. In this preparation, molar ratio of 1:1 for PEI or PDDA to Zn-TCP(Fe) was used, where the average molar mass of Zn-TCP(Fe) is approximately to be 945 g mol-1. All the membranes were washed several times with pure water followed by tested solutions before measurements. Characterization. X-ray diffraction (XRD) was performed on the Shimadzu thin film diffractometer with Cu-Kα irradiation (λ = 1.5406 Å). The patterns were collected from 5o to 45o at a scanning rate of 2o min-1, with a step size of 0.01o. The morphology and structure of the materials were characterized using transmission electron microscopy (TEM, JEOL2100F) operating at 200 kV and field emission scanning electron microscopy (FESEM, JEOL-6700F) equipped with energy-dispersive X-ray spectroscopy (EDX) analyzer (Oxford INCA). The elemental mapping was done by EDX spectroscopy (Oxford instruments, model 7426). The chemical bonding states and elemental compositions of the samples were studied by X-ray photoelectron spectroscopy (XPS) using a conventional monochromated Al Kα Xray source (AXIS-Hsi, Kratos Analytical, hѵ = 1486.71 eV). Inductively coupled plasma



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

Page 18 of 34

optical emission spectroscopy (ICP-OES, Optima 8300 ICP-OES spectrometer, PerkinElmer) was used to determine the molar ratios of M (M = Zn, Cu, and Co) to Fe in the M-TCP(Fe) samples. Atomic force microscopy (AFM, Bruker Multimode 8) was used to determine the thickness of the nanosheets. Fourier transform infrared spectroscopy (FTIR, Bruker) was used to obtain the functional group information of the samples using KBr pellet technique. Argon adsorption/desorption isotherms were conducted at 87.3 K using automatic manometric sorption analyzer (Quantachrome Instruments AutosorbiQ MP). The contact angle measurement was recorded using DataPhysics Instrument (OCA 30). The zeta potentials of the 2D MOF-based samples were analyzed by a SurPASS electrokinetic analyzer (Anton Paar GmbH, Austria) in 1 × 10-3 KCl solution. The concentrations of the dyes were measured using a UV-Vis spectrometer (Biochrom Libra S32). The concentrations of the salts were measured by an ion conductivity meter (ECTESTER11+, Cole-Parmer). Separation Performances. Nanofiltration measurements were performed using stirred vacuum filtration setup for low-pressure (< 1 bar) filtration tests while the stainless steel dead-end filtration device was used for high-pressure (≥ 1 bar) filtration tests. The latter device was pressurized by nitrogen gas. All the as-prepared membranes were supported on nylon and the effective area of the vacuum filtration setup and dead-end filtration device was estimated to be 9.1 cm2 and 1.3 cm2, respectively. All the filtration experiments were stirred at 500 rpm and three samples for each membrane were evaluated to obtain the error bars of permeances and rejection percentages. The rejection tests and long-term filtration tests were carried out at an operating pressure of 0.01 bar using vacuum filtration method. 10 mg L-1 of the dye probes (i.e. MR, MnB, MO, and BB); 1 g L-1 of BSA foulant; and 1 g L-1 of salts (i.e. MgCl2, Na2SO4, NaCl, and MgSO4) in deionized water were used as feed for the filtration experiments while the concentration of the feed, permeate, and retentate solutions were analyzed by using UV-Vis spectrophotometer for dyes and ion conductivity meter for salts.


Page 19 of 34

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 permeance (F) of the device was tabulated using Equation (1): F=

VP A × t × ∆P

(1)

where VP is the permeate volume (L), A is the effective area of the membrane (m2), t is the permeation time, and ∆P is the operating pressure. Therefore, the unit of F is L m-2 h-1 bar-1. The rejection percentage (R%) of the probes was tabulated using Equation (2): R=

CF − CP ×100% CF

(2)

where CF is the concentration of the feed solution and CP is the concentration of the permeate. The adsorption percentage (Ads%) of the probes on the membrane was tabulated using Equation (3): Ads =

VFCF − (VRCR + VPCP) ×100% VFCF

(3)

where CF, CP, CR are the concentrations of the feed, permeate, and retentate solutions, respectively. And VF, VP, and VR are the volumes of the feed, permeate, and retentate solutions, which are corresponding to 150 mL, 30 mL, and 120 mL, respectively. Antifouling property of the Zn-TCP(Fe)-based membrane was evaluated by the flux recovery percentage (FRP %), which can be tabulated by the following equation (4): FRP (%) =

FW , n FW , 1

(4)

where, FW,1 is the initial permeance of the pure water before the first cycle and FW,n is the membrane permeance after cleaning the fouled membrane by chemical solvents (i.e. ethanol, acetone and pure water) for the nth cycle. Computational Methods. The probe molecules were first generated using ChemDraw Ultra version 13.0 followed by performing energy minimization using the MM2 method. Then the dimensions of the solute molecules were measured using the same software in combination with Nano Measurer version 1.2. In addition, the surface charge distribution 19 ACS Paragon Plus Environment


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

Page 20 of 34

of the solutes was simulated using Avogadro version 1.2 software. The crystal structure of the 2D MOF was based on Choe et al’s model.41 In this context, the simulated pore size and interplane distance of 2D MOF were generated using VESTA version 3.4.0 software. The micropore distributions of the samples were calculated from the argon isotherms based on the nonlocal density functional theory (NLDFT) using SAIEUS.42

ASSOCIATED CONTENT Supporting Information Separation performances of various membranes (Table S1), chemical and physical characterizations (XRD patterns, XPS spectra, FTIR spectra, SEM images, EDX images, TEM images, AFM images and NLDFT plots) for M-TCP(Fe)-based nanosheets, and additional information: chemical properties of various solutes (Table S2), zeta potential and contact angle of 2D MOF-based membranes (Table S3), and comparison of permeances and pressure drop of Zn-TCP(Fe) versus rGO membranes (Table S4) together with proposed schematic diagram of permeation mechanism,

photographs of the filtration setup and

fabricated membrane, and crystal structure of 2D Zn-TCP(Fe) MOF with simulated pore size and interplane distance. UV-vis spectra of the feed, permeate, and retentate of various solutes passing through the M-TCP(Fe)-based membranes (PDF)

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected]. Tel: +65 6516 5029. Fax: +65 6779 1936. ORCID


Page 21 of 34

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


Liang Hong: 0000-0001-8955-6694 Notes The authors declare no competing financial interest.

ACKNOWLEDGEMENTS The authors acknowledge funding support from the Department of Chemical and Biomolecular Engineering, National University of Singapore, Singapore 117585, Singapore, research grant R-279-000-467-281.

REFERENCES 1.

Vandezande, P.; Gevers, L. E. M.; Vankelecom, I. F. J., Solvent Resistant

Nanofiltration: Separating on a Molecular Level. Chem. Soc. Rev. 2008, 37, 365−405. 2.

Elimelech, M.; Phillip, W. A., The Future of Seawater Desalination: Energy,

Technology, and the Environment. Science 2011, 333, 712−717. 3.

Marchetti, P.; Jimenez Solomon, M. F.; Szekely, G.; Livingston, A. G., Molecular

Separation with Organic Solvent Nanofiltration: A Critical Review. Chem. Rev. 2014, 114, 10735−10806. 4.

Xuan Nguyen, H.; Van der Bruggen, B., Nanofiltration of Synthetic and Industrial

Dye Baths: Influence of Temperature on Rejection and Membrane Fouling. J. Membr. Sci. Res. 2015, 1 , 34−40.

5.

Howarth, A. J.; Katz, M. J.; Wang, T. C.; Platero-Prats, A. E.; Chapman, K. W.; Hupp,

J. T.; Farha, O. K., High Efficiency Adsorption and Removal of Selenate and Selenite from Water Using Metal–Organic Frameworks. J. Am. Chem. Soc. 2015, 137, 7488−7494. 6.

Sun, P.; Zheng, F.; Zhu, M.; Song, Z.; Wang, K.; Zhong, M.; Wu, D.; Little, R. B.;

Xu, Z.; Zhu, H., Selective Trans-Membrane Transport of Alkali and Alkaline Earth Cations



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

Page 22 of 34

through Graphene Oxide Membranes Based on Cation−π Interactions. ACS Nano 2014, 8, 850−859. 7.

Chang, E. E.; Liang, C.-H.; Huang, C.-P.; Chiang, P.-C., A Simplified Method for

Elucidating the Effect of Size Exclusion on Nanofiltration Membranes. Sep. Purif. Technol. 2012, 85, 1−7. 8.

Coronell, O.; Mi, B.; Mariñas, B. J.; Cahill, D. G., Modeling the Effect of Charge

Density in the Active Layers of Reverse Osmosis and Nanofiltration Membranes on the Rejection of Arsenic(III) and Potassium Iodide. Environ. Sci. Technol. 2013, 47, 420−428. 9.

Liu, G.; Jin, W.; Xu, N., Graphene–Based Membranes. Chem. Soc. Rev. 2015, 44,

5016−5030. 10.

Zheng, Z.; Grünker, R.; Feng, X., Synthetic Two–Dimensional Materials: A New

Paradigm of Membranes for Ultimate Separation. Adv. Mater. 2016, 28, 6529−6545. 11.

Zhang, H.; Xiao, Q.; Guo, X.; Li, N.; Kumar, P.; Rangnekar, N.; Jeon, M. Y.; Al-

Thabaiti, S.; Narasimharao, K.; Basahel, S. N.; Topuz, B.; Onorato, F. J.; Macosko, C. W.; Mkhoyan, K. A.; Tsapatsis, M., Open–Pore Two–Dimensional MFI Zeolite Nanosheets for the Fabrication of Hydrocarbon–Isomer–Selective Membranes on Porous Polymer Supports. Angew. Chem. Int. Ed. 2016, 55, 7184−7187.

12.

Sun, P.; Wang, K.; Zhu, H., Recent Developments in Graphene–Based Membranes:

Structure, Mass-Transport Mechanism and Potential Applications. Adv. Mater. 2016, 28, 2287−2310. 13.

Rollings, R. C.; Kuan, A. T.; Golovchenko, J. A., Ion Selectivity of Graphene

Nanopores. Nat. Commun. 2016, 7, 11408 DOI: 10.1038/ncomms11408. 14.

Han, Y.; Xu, Z.; Gao, C., Ultrathin Graphene Nanofiltration Membrane for Water

Purification. Adv. Funct. Mater. 2013, 23, 3693−3700.


Page 23 of 34

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


15.

Huang, L.; Chen, J.; Gao, T.; Zhang, M.; Li, Y.; Dai, L.; Qu, L.; Shi, G., Reduced

Graphene Oxide Membranes for Ultrafast Organic Solvent Nanofiltration. Adv. Mater. 2016, 28, 8669−8674.

16.

Akbari, A.; Sheath, P.; Martin, S. T.; Shinde, D. B.; Shaibani, M.; Banerjee, P. C.;

Tkacz, R.; Bhattacharyya, D.; Majumder, M., Large-area Graphene–based Nanofiltration Membranes by Shear Alignment of Discotic Nematic Liquid Crystals of Graphene Oxide. Nat. Commun. 2016, 7, 10891 DOI: 10.1038/ncomms10891.

17.

Huang, H.; Song, Z.; Wei, N.; Shi, L.; Mao, Y.; Ying, Y.; Sun, L.; Xu, Z.; Peng, X.,

Ultrafast Viscous Water Flow through Nanostrand-channelled Graphene Oxide Membranes. Nat. Commun. 2013, 4, 2979 DOI: 10.1038/ncomms3979.

18.

Goh, K.; Jiang, W.; Karahan, H. E.; Zhai, S.; Wei, L.; Yu, D.; Fane, A. G.; Wang, R.;

Chen, Y., All−Carbon Nanoarchitectures as High−Performance Separation Membranes with Superior Stability. Adv. Funct. Mater. 2015, 25, 7348−7359. 19.

Quan, Q.; Lin, X.; Zhang, N.; Xu, Y.-J., Graphene and its Derivatives as Versatile

Templates for Materials Synthesis and Functional Applications. Nanoscale 2017, 9, 2398−2416. 20.

Han, C.; Zhang, N.; Xu, Y.-J., Structural Diversity of Graphene Materials and Their

Multifarious Roles in Heterogeneous Photocatalysis. Nano Today 2016, 11, 351−372. 21.

Sun, L.; Ying, Y.; Huang, H.; Song, Z.; Mao, Y.; Xu, Z.; Peng, X., Ultrafast Molecule

Separation through Layered WS2 Nanosheet Membranes. ACS Nano 2014, 8, 6304−6311. 22.

Heiranian, M.; Farimani, A. B.; Aluru, N. R., Water Desalination with a Single–layer

MoS2 Nanopore. Nat. Commun. 2015, 6, 8616 DOI: 10.1038/ncomms9616. 23.

Li, W.; Yang, Y.; Weber, J. K.; Zhang, G.; Zhou, R., Tunable, Strain–Controlled

Nanoporous MoS2 Filter for Water Desalination. ACS Nano 2016, 10, 1829−1835.



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

24.

Page 24 of 34

Ding, L.; Wei, Y.; Wang, Y.; Chen, H.; Caro, J.; Wang, H., A Two-Dimensional

Lamellar Membrane: MXene Nanosheet Stacks. Angew. Chem. Int. Ed. 2017, 56, 1825−1829. 25.

Ren, C. E.; Hatzell, K. B.; Alhabeb, M.; Ling, Z.; Mahmoud, K. A.; Gogotsi, Y.,

Charge- and Size-Selective Ion Sieving Through Ti3C2Tx MXene Membranes. J. Phys. Chem. Lett. 2015, 6, 4026−4031.

26.

Qu, Y.; Zhang, Q. G.; Soyekwo, F.; Gao, R. S.; Lv, R. X.; Lin, C. X.; Chen, M. M.;

Zhu, A. M.; Liu, Q. L., Nickel Hydroxide Nanosheet Membranes with Fast Water and Organics Transport for Molecular Separation. Nanoscale 2016, 8, 18428−18435. 27.

Su, Y.; Kravets, V. G.; Wong, S. L.; Waters, J.; Geim, A. K.; Nair, R. R.,

Impermeable Barrier Films and Protective Coatings Based on Reduced Graphene Oxide. Nat. Commun. 2014, 5, 4843 DOI: 10.1038/ncomms5843.

28.

Senyuk, B.; Behabtu, N.; Martinez, A.; Lee, T.; Tsentalovich, D. E.; Ceriotti, G.; Tour,

J. M.; Pasquali, M.; Smalyukh, I. I., Three-dimensional Patterning of Solid Microstructures through Laser Reduction of Colloidal Graphene Oxide in Liquid-Crystalline Dispersions. Nat. Commun. 2015, 6, 7157 DOI: 10.1038/ncomms8157.

29.

Dikin, D. A.; Stankovich, S.; Zimney, E. J.; Piner, R. D.; Dommett, G. H. B.;

Evmenenko, G.; Nguyen, S. T.; Ruoff, R. S., Preparation and Characterization of Graphene Oxide Paper. Nature 2007, 448, 457−460. 30.

Gao, S. J.; Qin, H.; Liu, P.; Jin, J., SWCNT-Intercalated GO Ultrathin Films for

Ultrafast Separation of Molecules. J. Mater. Chem. A. 2015, 3, 6649−6654. 31.

Wang, X.; Chi, C.; Zhang, K.; Qian, Y.; Gupta, K. M.; Kang, Z.; Jiang, J.; Zhao, D.,

Reversed Thermo-Switchable Molecular Sieving Membranes Composed of Two– Dimensional Metal–Organic Nanosheets for Gas Separation. Nat. Commun. 2017, 8, 14460 DOI: 10.1038/ncomms14460.


Page 25 of 34

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


32.

Peng, Y.; Li, Y.; Ban, Y.; Jin, H.; Jiao, W.; Liu, X.; Yang, W., Metal–Organic

Framework Nanosheets as Building Blocks for Molecular Sieving Membranes. Science 2014, 346, 1356−1359.

33.

Varoon, K.; Zhang, X.; Elyassi, B.; Brewer, D. D.; Gettel, M.; Kumar, S.; Lee, J. A.;

Maheshwari, S.; Mittal, A.; Sung, C.-Y.; Cococcioni, M.; Francis, L. F.; McCormick, A. V.; Mkhoyan, K. A.; Tsapatsis, M., Dispersible Exfoliated Zeolite Nanosheets and Their Application as a Selective Membrane. Science 2011, 334, 72−75. 34.

Rodenas, T.; Luz, I.; Prieto, G.; Seoane, B.; Miro, H.; Corma, A.; Kapteijn, F.;

Llabrés i Xamena, F. X.; Gascon, J., Metal–Organic Framework Nanosheets in Polymer Composite Materials for Gas Separation. Nat. Mater. 2015, 14, 48–55. 35.

Zhang, R.; Ji, S.; Wang, N.; Wang, L.; Zhang, G.; Li, J.-R., Coordination-Driven In

Situ Self-Assembly Strategy for the Preparation of Metal–Organic Framework Hybrid Membranes. Angew. Chem. Int. Ed. 2014, 53, 9775–9779. 36.

Wang, L.; Fang, M.; Liu, J.; He, J.; Li, J.; Lei, J., Layer-by-Layer Fabrication of

High-Performance

Polyamide/ZIF-8

Nanocomposite

Membrane

for

Nanofiltration

Applications. ACS Appl. Mater. Interfaces 2015, 7, 24082−24093. 37.

Wang, Y.; Zhao, M.; Ping, J.; Chen, B.; Cao, X.; Huang, Y.; Tan, C.; Ma, Q.; Wu, S.;

Yu, Y.; Lu, Q.; Chen, J.; Zhao, W.; Ying, Y.; Zhang, H., Bioinspired Design of Ultrathin 2D Bimetallic Metal–Organic−Framework Nanosheets Used as Biomimetic Enzymes. Adv. Mater. 2016, 28 (21), 4149−4155.

38.

Wang, K.; Feng, D.; Liu, T.-F.; Su, J.; Yuan, S.; Chen, Y.-P.; Bosch, M.; Zou, X.;

Zhou, H.-C., A Series of Highly Stable Mesoporous Metalloporphyrin Fe−MOFs. J. Am. Chem. Soc. 2014, 136, 13983−13986.

39.

Bellettini, I. C.; Nandi, L. G.; Eising, R.; Domingos, J. B.; Machado, V. G.; Minatti,

E., Properties of Aqueous Solutions of Hydrophobically Modified Polyethylene Imines in the



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

Page 26 of 34

Absence and Presence of Sodium Dodecylsulfate. J. Colloid Interface Sci. 2012, 370, 94−101. 40.

Dautzenberg, H.; Görnitz, E.; Jaeger, W., Synthesis and Characterization of

Poly(diallyldimethylammonium chloride) in a Broad Range of Molecular Weight. Macromol. Chem. Phys. 1998, 199, 1561−1571. 41.

Choi, E.-Y.; Wray, C. A.; Hu, C.; Choe, W., Highly Tunable Metal−Organic

Frameworks with Open Metal Centers. CrystEngComm 2009, 11, 553−555. 42.

Serrano, D. P.; Aguado, J.; Morales, G.; Rodríguez, J. M.; Peral, A.; Thommes, M.;

Epping, J. D.; Chmelka, B. F., Molecular and Meso− and Macroscopic Properties of Hierarchical Nanocrystalline ZSM-5 Zeolite Prepared by Seed Silanization. Chem. Mater. 2009, 21, 641−654.

Figure 1. Schematic illustration for the fabrication of polycation-regulated 2D Zn-TCP(Fe) membrane. 26 ACS Paragon Plus Environment

Page 27 of 34

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


Figure 2. a) XRD pattern; b) XPS C 1s spectrum; c) XPS N 1s spectrum; d) XPS Fe 2p spectrum; e) XPS O 1s spectrum; f) XPS Zn 2p spectrum of the 2D Zn-TCP(Fe) nanosheets.



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

Page 28 of 34

Figure 3. a) SEM image; b) AFM image; c) enlarged TEM image and its corresponding d) SAED pattern of the 2D Zn-TCP(Fe) nanosheets.

Figure 4. a) Argon (87.3 K) NLDFT cumulative pore volume plot versus the pore width and b) pore size distribution plot of the Zn-TCP(Fe) nanosheets.


Page 29 of 34

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


Figure 5. a) SEM image of the 2D Zn-TCP(Fe) selective layer on nylon support prepared by vacuum filtration method (wrinkles are denoted by the white arrow); b) cross-sectional SEM image of Zn-TCP(Fe) selective layer; c) SEM image (scale bar, 100 μm) of Zn-TCP(Fe) selective layer and their corresponding EDX mapping images; d) AFM image of the ZnTCP(Fe) selective layer on silicon wafer with its corresponding e) height profiles.



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

Page 30 of 34

Figure 6. Measurements of a) permeance and b) rejection percentage of methyl red (inset: molecular structure of methyl red) as a function of various selective layer thickness. c) Water flux versus operating pressure plot for M-TCP(Fe) (M = Zn, Cu, and Co) membranes. (d-f) Water permeances (represents by the blue line), rejection rates (represents by green columns) and adsorption rates (represents by red columns) of the 2D M-TCP(Fe) membranes.


Page 31 of 34

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


Figure 7. XPS N 1s spectra of (a) PEI/Zn-TCP(Fe) and (b) PDDA/Zn-TCP(Fe) selective layers. (c) XRD patterns of the PEI/Zn-TCP(Fe), PDDA/Zn-TCP(Fe), and pristine ZnTCP(Fe) selective layers referenced to crystalline Si (JCPDS No. 35-1158); and (d) the analysis for interplane distance (denoted as “d”) and full width at half maximum (denoted as “fwhm”) of the respective selective layers in the zoom-in region between 6o and 13o.



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

Page 32 of 34

Figure 8. Separation performances of pristine Zn-TCP(Fe) membrane versus polycationregulated 2D Zn-TCP(Fe) membranes. a) Rejection percentage of methyl red solute versus thickness of selective layer; b) permeance as a function of various selective layer thickness for PEI/Zn-TCP(Fe), PDDA/Zn-TCP(Fe), and pristine Zn-TCP(Fe) membranes. For rejection cut-off at 90%, the optimum thickness of the selective layer can be reduced by ca. 30% when polycation-regulated Zn-TCP(Fe) membranes are used, hence its permeance can be boosted by ca. 2 times as compared to its pristine. c) Permeances (denoted by green columns) and rejection rates (denoted by colored lines) of polycation-regulated 2D Zn-TCP(Fe) selective layer (optimum thickness: ca. 48 nm) versus pristine Zn-TCP(Fe) selective layer (optimum thickness: ca. 67 nm) on nylon support for various probe solutes.


Page 33 of 34

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


Figure 9. Long-term separation test on the (a) PEI/Zn-TCP(Fe) and (b) PDDA/Zn-TCP(Fe) membranes using BB solute for 8 h. Inset shows the photographs of feed and permeate solutions after the durability test. Long-term antifouling test on (c) PEI/Zn-TCP(Fe) and (d) PDDA/Zn-TCP(Fe) membranes using BSA solute for 10 h. (e) Permeance as a function of the energy consumption for the 2D M-TCP(Fe)-based membranes, in comparison with some advanced membranes; and the (f) enlarged version in the region between 0 to 0.2 kJ L-1 (see details in the Table S1, Supporting Information).



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

Page 34 of 34

Table of Contents Polycationic Polymer-Regulated Assembling of 2D MOF Nanosheets for High-Performance Nanofiltration


Polycationic Polymer-Regulated Assembling of 2D MOF Nanosheets

Recommend Documents