A Method for the Efficient Fabrication of Multifunctional Mosaic

Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article

A Method for the Efficient Fabrication of Multifunctional Mosaic Membranes by Inkjet Printing Peng Gao, Aaron J. Hunter, Mark J Summe, and William A. Phillip ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.6b06048 • Publication Date (Web): 13 Jul 2016 Downloaded from http://pubs.acs.org on July 18, 2016

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 28

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

A Method for the Efficient Fabrication of Multifunctional Mosaic Membranes by Inkjet Printing Peng Gao, Aaron Hunter, Mark J. Summe, and William A. Phillip* Department of Chemical and Biomolecular Engineering, University of Notre Dame, Notre Dame, Indiana 46556, United States *To whom correspondence should be addressed: [email protected]. Abstract. Most conventional membrane systems are based on size-selective materials that permeate smaller molecules and retain larger ones. However, membranes that can permeate larger molecules more rapidly than smaller ones could find widespread utilization in multiple arenas of technology. Charge mosaic membranes are one example of such a system. Due to their unique nanostructure, which consists of discrete oppositely-charged domains, charge mosaics are capable of permeating large dissolved salts more rapidly than smaller water molecules. Here, we present a combined inkjet printing and template synthesis technique to prepare charge mosaic membranes in a rapid and straightforward manner, and demonstrate the unique transport properties that result from the mosaic membrane design. Poly(vinyl alcohol) based composite inks containing poly(diallyldimethylammonium chloride) or poly(sodium 4-styrenesulfonate) were used to pattern positively-charged or negatively-charged domains, respectively, on the surface of a polycarbonate track-etched membrane with 30 nm pores. The ability to control the net surface charge of the mosaic membranes through the rationale deposition of the oppositelycharged materials was demonstrated, and confirmed through nanostructural characterization, electrokinetic measurements, and piezodialysis experiments. Namely, mosaic membranes that possessed an overall neutral charge (i.e., membranes that had equal coverage of positivelycharged and negatively-charged domains) were capable of enriching the concentration of potassium chloride in the solution that permeated through the membrane. These membranes could be deployed in the many established and emerging nanoscale technologies that rely on the selective transport and separation of ionic solutes from solution. Furthermore, because of the flexibility provided by the membrane fabrication platform, the efforts reported in this work can be extended to other mosaic designs with myriad other functional components. Keywords. inkjet printing, charge mosaic membranes, ion transport, polyelectrolytes, polymer composites

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 28

Introduction The commercial successes of membrane separations have largely been based on sizeselective materials that allow smaller molecules to permeate while retaining larger molecules.1–5 In several arenas, however, significant advantages exist for chemically-selective membranes that are capable of permeating larger molecules more quickly than smaller molecules.6–10 Charge mosaic membranes, which possess discrete oppositely-charged domains, are an example of a membrane that can permeate large dissolved salts more rapidly than similarly-sized neutral solutes and smaller water molecules.6–9,11–14 However, materials processing challenges have hindered their advancement. Here, we demonstrate the straightforward fabrication of charge mosaics using a combination of inkjet printing and template synthesis. Our results suggest that this combination addresses the processing challenges that have stymied the advancement of chemically-selective mosaic membranes and that the simple operation, facile control over surface structure, and diverse range of materials that can be implemented in this method will enable the ultimate widespread utilization of mosaic membranes for myriad applications, e.g., cell patterned sensors and textured surfaces for anti-fouling applications.15 Charge mosaic membranes (Figure 1) possess arrays of both positively and negatively charged domains. The juxtaposition of the counter-charged domains allows both cations and anions to permeate through the charge-functionalized membrane without violating the macroscopic constraint of electroneutrality, which greatly enhances the overall permeability of electrolytes. The concept of a charge mosaic membrane was first proposed by Sollner in 1932.16 Since then, multiple attempts have been made to develop charge mosaic membranes from several polymeric materials platforms including self-assembled block polymers,8,17 ion exchange resins,6,7 electrospun polymers,11 polymer blends,12 and layer-by-layer deposition.13


1

Page 3 of 28

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 1. Schematic diagram of the inkjet printing process developed in this work to fabricate charge mosaic membranes. Charge mosaic membranes consist of distinct cationic (green) and anionic (purple) domains that traverse the membrane thickness. The cationic domains allow the passage of anions but restrict cations from passing, while the anionic domains allow the passage of cations but restrict anions from permeating. Polymer composite inks that contained polyelectrolytes were printed on a template surface to generate membranes with a charge mosaic structure. Membranes with this unique structure can transport dissolved salts more rapidly than similarly-sized neutral solutes and/or solvents. These past efforts have identified some design criteria for the generation of highperformance charge mosaic membranes. For example, the oppositely-charged domains should be densely packed on the membrane surface and should traverse the membrane thickness. Additionally, the surface charge densities of the positively-charged and negatively-charged domains should be as high as possible. The net surface charge averaged over the whole membrane surface, however, should be neutral. The straightforward fabrication of highlyeffective charge mosaics from prior materials systems has proven difficult due to the need to orient the ionic domains perpendicular to the surface, and the morphological changes induced during the harsh chemical treatments required to introduce charged moieties into some materials. These materials processing challenges have made satisfying the known design criteria difficult. And due to this difficulty in producing charge mosaics, the development of the mosaic membrane platform has lagged. Conversely, inkjet printing has emerged as a promising materials processing technology because it provides the ability to deposit functional materials onto a substrate surface rapidly and


2


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 28

precisely. Due to these advantages, inkjet printing has been used to create multidimensional structures18–22 for a large variety of applications (e.g., electronic circuits,23 microbatteries,24 thin film transistors,25 and tissues).26 However, the use of inkjet printing in the preparation of functional membranes has been limited. Here, we report a combination of inkjet printing and template synthesis27 that addresses the materials processing issues that have hindered the development of charge mosaic membranes and enables the straightforward fabrication of mosaics with well-defined and well-controlled surface patterns from a diversity of materials chemistries. Results and Discussion Preparation of polymer composite inks. The generation of polymeric composite inks with varied functionality was critical to our ability to fabricate charge mosaic membranes using a combination of inkjet printing and template synthesis. The composite inks used in this study contained polyvinyl alcohol (PVA), a charged polyelectrolyte, and a fluorescent dye dissolved in DI water. Each component in the formulation of the inks served a specific purpose. PVA is commonly used for preparing polymeric composites12,28 because it can be easily crosslinked to form a semi-interpenetrating network that entraps a functional component (Figure S1 and Figure S2, Supporting Information).29 The reported method is versatile due to the ability to generate polymer composite inks with an almost arbitrary number of functionalities as long as suitable solvents and templates can be identified. In this case where we fabricated charge mosaic membranes successfully, polyelectrolytes were used as the functional component to impart charge to the membrane. In particular, the polyelectrolytes, poly(diallyldimethylammonium chloride) (PDADMAC) and poly(sodium 4-styrene sulfonate) (PSS) were used as the functional component of the positively-charged ink and negatively-charge ink, respectively, because they


3

Page 5 of 28

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


are strong polyelectrolytes that possess high charge densities over a wide pH range. The fluorescent dye was used to enable visual observation of the printed domains. Two factors affected the formulation of our polymer composite inks. First, a solution with a dynamic viscosity less than 20 mPa s was necessary to ensure smooth jetting of the inks onto the template surface.30 For this reason, a 1% (by weight) solution of PVA in water served as the base of the polymer composite inks (Figure S3, Supporting Information). The second consideration that impacted the formulation of the precursor inks was the density of functional moieties within the final composite material. As displayed in Figure 2a, this variable can be adjusted by incorporating different concentrations of polyelectrolyte into the polymer composite ink. Figure 2a displays how the streaming current of the printed membranes changed as the concentrations of polyelectrolyte in the precursor ink was varied. In these experiments, polymer inks of a single type (i.e., PDADMAC-containing or PSS-containing) were printed onto a polycarbonate track-etched (PCTE) membrane with pores 30 nm in diameter. Subsequently, the streaming current, which is proportional to the surface charge, was measured using a previouslydetailed method.18 Using this method, surfaces with a positive charge generated a negative streaming current while surfaces with a negative charge generated a positive streaming current. The magnitude of the streaming current for both of the membranes increased monotonically for polyelectrolyte concentrations that ranged from 0.004 M to 0.1 M, which indicated an increase in surface charge density. For the PSS-based membranes, the streaming current appears to asymptote above a polyelectrolyte concentration of 0.1 M, suggesting a saturation concentration is reached. A concentration higher than 0.1 M was not implemented for the PDADMAC-based membranes because at polyelectrolyte concentrations greater than 0.1 M the PDADMACcontaining inks were prone to clogging the print head. Inks that contained 0.1 M PDADMAC


4


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 28

and 0.5 M PSS were used in all of the following experimentation due to their suitability for printing; and because domains generated from these inks exhibited relatively large streaming currents that were nearly equal in magnitude but opposite in sign, which is needed to produce high performance charge mosaic membranes.

Figure 2. Streaming current of charge-functionalized membranes prepared using a combination of inkjet printing and template synthesis. The composition of the polymer composite ink and the printing conditions can be used to control the surface charge density and nanostructure of the charge-functionalized membranes. The charge-functionalized membranes were printed while applying a constant vacuum of 12 psig to the substrate. A PCTE membrane with 30 nm pores was used as the substrate in all experiments. a) Streaming current for membranes printed with varying concentrations of polyelectrolyte in the polymer composite ink. Three overprints were used. The polymer composite inks contained 1% (by weight) poly(vinyl alcohol) (PVA) and a polyelectrolyte at the prescribed concentration dissolved in DI water. Positively charged inks contained poly(diallyldimethylammonium chloride) (PDADMAC); Negatively charged inks contained poly(styrene sulfonate) (PSS). b) Streaming current for membranes printed with different values for the number of overprints. The polymer composite inks in these experiments were a solution of 1% (by weight) PVA and 0.1 M PDADMAC in DI water and a solution of 1% (by weight) PVA and 0.5 M PSS in DI water for the positively-charged and negatively-charged ink, respectively. c) The mosaic membrane structure after dissolving the PCTE substrate by immersing the charge mosaic in dichloromethane. A mesh of nanowires form inside the pores of


5

Page 7 of 28

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 PCTE membrane. d) A higher magnification micrograph of the nanowires formed within the pores of the PCTE substrate. Selection of materials deposition conditions. In addition to the intrinsic properties of the polymer composite ink upon formulation, the materials processing conditions affect the surface charge of the printed membrane. Controlling the number of ink droplets jetted at each location of the print head, defined as the number of overprints, is critical to tailoring the surface charge density of the membrane materials. Figure 2b displays how the surface charge of printed membranes varied with the number of overprints when charged inks were printed onto a PCTE template. The PCTE template (zero overprints) generated a positive streaming current due to the negative charge on its surface. The sign of the streaming current for the membrane printed with PDADMAC-containing ink flipped and its magnitude gradually decreased to a more negative value with an increasing number of overprints, which indicated that the surface charge of the membrane became more positive as larger volumes of ink were deposited onto the membrane. The result fits well with the hypothesis that as ink is pulled through the open pores of the PCTE template, the polymeric components are deposited on the pore wall of the template, covering and eventually screening the initially-negatively charged surface. Scanning electron microscopy (SEM) micrographs of the membrane after the PCTE template had been dissolved further support this hypothesis. Figure 2c displays a lower magnification image that shows a mesh of nanowires after the dissolution of the template. The higher magnification micrograph in Figure 2d shows that the diameter of the nanowires in the mesh is around 42 ± 3 nm, which is consistent with the 30 nm pore size reported for the PCTE template. This result is in good agreement with our prior work combining inkjet printing with template synthesis.18 A side-by-side comparison of SEM micrographs of the nanowires formed using PSS-containing and PDADMAC-containing inks demonstrates that the nanowires formed in the negative and positive domains possess


6


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 28

similar nanostructures (Figure S4, Supporting Information). The surface charge of the membrane printed with the PSS-containing ink showed little change as the number of overprints was varied, which suggests that the negative ink covered the pore surface with a similar density of charged moieties as that present on the surface of the PCTE template. Based on the results above, five overprints were chosen for all subsequent experimentation because the positive and negative inks produced similar values of surface charge. When printed with equal areal fractions, these two inks produced a charge mosaic membrane that satisfied the design constraint of an overall neutral membrane surface. Printing charge mosaic membranes. Using inkjet printing allowed for the patterning of the charged domains on the membrane surface to be controlled in a straightforward manner, which, thereby, enabled the formation of charge mosaic membranes. Here, we demonstrate the use of this facile and scalable method for producing a charge mosaic membrane that is capable of enriching (i.e., increasing) the salt concentration in the permeate relative to the feed. A pattern of alternating stripes was used because it allowed the areal fraction of positively-charged domains to be adjusted by modifying the relative width of the stripes. Figure 3a displays fluorescent micrographs of membranes with areal fractions of the positively-charged domain that range from 0% to 100%. In these micrographs, the negatively-charged domains appear purple and the positively-charge domains appear green. Printing only the PSS-containing and the PDADMACcontaining inks on the membranes surface generated 0% and 100% surface coverage, respectively. An intermediate areal fraction corresponding to 29% coverage was generated by printing stripes with widths of 106±7 µm (PDADMAC-containing) and 257±7 µm (PSScontaining); 52% coverage was generated from stripes with widths of 94±5 µm (PDADMAC-


7

Page 9 of 28

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


containing) and 101±7 µm (PSS-containing); and 75% coverage was produced using striped with widths of 294±10 µm (PDADMAC-containing) and 96±9 µm (PSS-containing).

Figure 3. Fluorescent images, streaming current, and salt rejection for charge mosaic membranes printed with different areal fractions of positive and negative charge. The patterning of membranes fabricated using a combination of inkjet printing and template synthesis can be easily adjusted in order to control the surface charge and transport properties of the charge mosaic membrane. A PCTE membrane with a pore diameter of 30 nm was used as a substrate in all experiments. Positive regions were formed by printing a polymer composite ink that contained 1% (by weight) PVA, 0.1 M PDADMAC, and 5 µM FITC-PAH. Negative regions were formed by printing a polymer composite ink that contained 1% (by weight) PVA, 0.5 M PSS, and 5 µM Cy5. a) In the fluorescent micrographs, the positive regions appear green in color and the negative regions appear purple in color. The fraction of the membrane surface covered by the oppositely-charged moieties was controlled by printing stripes of different widths. b) The streaming current of the charge mosaic membranes was measured using a 10 mM potassium chloride (KCl) solution with unadjusted pH. Pressure was applied to the side of the system connected to the positive terminal of the sourcemeter. Error bars represent the standard deviation (n = 3). c) Salt rejection of a 0.1 mM KCl feed solution. Experiments were executed with the charge mosaic membranes mounted in a dead-end filtration cell. An applied pressure of 4 bar was used to drive permeation. Error bars represent multiple tests (n = 4) on a single membrane.


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

Page 10 of 28

Table 1. Hydraulic permeability of charge mosaic membranes printed with different areal fractions of positive and negative charge. Percent of positive coverage 0% 29% 52% 75% 100%

Hydraulic permeability (L m-2 h-1 bar-1) 2.3±0.5 1.7±0.4 1.0±0.3 2.1±0.6 2.8±0.5

Transport characteristic of charge mosaic membranes. The hydraulic permeability of the printed mosaic membranes ranged from 0.6 to 3.0 L m-2 h-1 bar-1 and are listed a function of the areal coverage of positive domains in Table 1. The streaming currents measured for this series of membranes are displayed in Figure 3b. Membranes printed with only the PSS-containing ink displayed the most positive streaming current, which corresponds to the highest density of negatively charged moieties. The streaming current decreased monotonically as the surface coverage of the positive domain increased. Given the streaming current of the positively-charged and negatively-charged membranes, the streaming current for the mosaic membranes could be predicted using a simple weighted arithmetic average as shown by the dashed line in Figure 3b (Eqs. 1 and 2, Supporting Information). This suggests that, as designed, discrete domains of positive charge and negative charge are produced upon printing, and that the method reported here enables control of the relative surface coverage of multiple domains. Examining the morphology of the charge mosaic using SEM also confirms that discrete domains are formed. Figure 4a shows the pattern of alternating stripes printed with 52% areal coverage for the positive domain. From this micrograph, it is clear that the topology of the two domains appear different. Higher magnification micrographs (Figure 4b) demonstrate that the positively-charged domains are smooth while the negatively-charged domains are rough. This surface roughness is


9

Page 11 of 28

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


characteristic of composites that contain PVA and PSS.31 The difference in the appearances of the stripes and the variations in the streaming current further confirm that discrete domains are generated by the combination of inkjet printing and template synthesis. The salt rejecting capabilities of membranes that possess only a single type of charge are fairly well established for simple salts such as sodium chloride (NaCl) and potassium chloride (KCl).32 However, the effects of surface charge on the performance of mosaic membranes are not well established. Therefore, salt rejection experiments for membranes patterned with different areal fractions of positively-charged domains were executed using a 0.1 mM potassium chloride (KCl) feed solution (Figure 3c). The low feed solution concentration was selected to ensure that ion selectivity for the individual domains remained high.33 Membranes printed with only the PSS-containing (0%) or PDADMAC-containing (100%) inks showed the highest salt rejection, which was expected based on the high surface charge measured for these membranes (Figure 3b). As mosaic patterning was incorporated into the membranes (29% and 75% coverage), the salt rejection values remained positive, but their magnitude was reduced from ~65% to ~25% rejection. The lower rejection of dissolved salts is in good agreement with the decreased overall surface charge of the membranes. The most interesting result comes from the membrane printed with equal areal coverage of the positive and negative domains (52%). This membrane, which had a nearly neutral surface charge, produced a negative salt rejection (i.e., it enriched the concentration of salt in the permeate relative to the feed). For single salt systems, this is a characteristic unique to charge mosaic membranes.


10


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 28

Figure 4. SEM micrographs of a charge mosaic membrane. The micrographs depict the distinct nanostructures of the oppositely-charged domains on the surface of the charge mosaic membrane. The mosaic membrane was patterned by printing alternating stripes, ~ 100 μm in width, of positively-charged inks (1% (by weight) PVA/0.1 M PDADMAC/5 µM FITC-PAH in water) and negatively-charged inks (1wt% PVA/0.5 M PSS/5 µM Cy5 in water) onto a PCTE membrane with pores 30 nm in diameter. Five overprints were used and a constant vacuum of 12 psig was applied to the substrate. a) The top surface of the charge mosaic membrane. b) Higher magnification micrographs of the positively-charged (top) and negatively-charged (bottom) regions of the mosaic membrane. Because electrostatic interactions between the membrane and dissolved ions play a significant role in the performance of charge-functionalized membranes, KCl enrichment was measured for feed solution concentrations of 1 mM and 10 mM to study the impact of ionic strength of the performance of charge mosaic membranes. A rejection of -17±5% for the 1 mM feed solution and -2.0±1.6% for the 10 mM feed solution were observed, indicating that the mosaic membrane was able to enrich the salt concentration even for these more concentrated feed solutions. Further inspection of these results indicated that membrane performance was optimal when the Debye length is greater than the pore radius, which is consistent with previous studies on other charge-functionalized membranes.34 The Debye length for a surface in a 0.1 mM and 1 mM KCl feed solution (30.5 nm and 9.6 nm, respectively) is greater than the radius of the pore of the printed membranes estimated from PEO rejection experiments, 6.3 nm.35 However,


11

Page 13 of 28

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 Debye length for the 10 mM feed solution, 3.1 nm, is smaller than the estimated pore size (Equation S7, Supporting Information). This suggests that developing charge mosaics from templates with smaller pores as a straightforward route toward the generation of charge mosaic membranes that perform well in high ionic strength environments. Conclusions These results demonstrate conclusively that by simply changing the width of the stripes of charged inks deposited on the template surface, the surface charge and transport properties of charge mosaic membranes fabricated using a combination of inkjet printing and template synthesis can be easily adjusted. This unique ability will enable further fundamental studies on charge mosaic membranes that can be deployed in the many established and emerging technologies where the selective transport of ionic solutes is of critical import. Furthermore, the membrane fabrication platform demonstrated here, which relies on easily-tailored composite inks, can be extended to a wide range of matrix materials and functional components, and as such will enable the design and development of other mosaic membranes with patterned surface chemistries and structures that have yet to have been realized or invented.

Methods. Materials: Polycarbonate track-etched (PCTE) membranes (pore diameter: 30 nm) were purchased from Whatman. Poly(vinyl alcohol) (PVA) powder (98-99% hydrolyzed), poly(diallyldimethylammonium chloride) (PDADMAC, Mw < 100,000), poly(sodium 4styrenesulfonate)

(PSS,

70

kDa)

fluorescein

isothiocyanate-labeled

poly(allylamine

hydrochloride) (FITC-PAH), 37% (by volume) hydrochloric acid, 25% (by weight) glutaraldehyde, potassium chloride were purchased from Sigma Aldrich and used as received.


12


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 28

Sulfo-Cyanine5 (Cy5) was purchased from Lumiprobe. The acrodisc 25 mm syringe filter fitted with a 1 µm glass fiber membrane was purchased from Pall corporation. The water used in all experiments was obtained from a Millipore water purification system. The general procedure to print and characterize the charge mosaic membranes involves the following steps: 1. The polymer composite inks are prepared by dissolving polyvinyl alcohol, a charged polyelectrolyte, and a fluorescent dye in DI water. 2. Charge mosaic membranes were prepared by printing predesigned patterns of the polymer composite inks onto a template substrate and then chemically crosslinking the composite. 3. Charge mosaic membranes were characterized using a series of techniques including streaming current measurements, fluorescent microscopy, scanning electron microscopy, and transport tests. Preparation of polymer composite inks: The polymer composite inks contained PVA, a charged polyelectrolyte, and a fluorescent dye dissolved in DI water. The viscosity of the ink is a significant consideration when formulating the polymer composite ink. Specifically, the dynamic viscosity should be less than 20 mPa s to avoid clogging of the printer head.30 It is known that the concentration of PVA dissolved in DI water affects the solution viscosity.36 Therefore, a 1% (by weight) solution of PVA in water, which has a viscosity of 1.35 mPa s, was chosen for all experiments to ensure a smooth ink jetting. The 1% (by weight) PVA solution was prepared by dissolving PVA powder in water at 80 °C for 24 hours. It was then filtered through an Acrodisc 25 mm syringe filter fitted with a 1 µm glass fiber membrane. The filtration removes any suspended PVA particles that would clog the printer head. The polyelectrolyte PDADMAC was added to the PVA solution to render a positivelycharged composite ink. The negatively-charged ink was prepared by adding PSS to the 1% (by weight) PVA solution. The concentration of polyelectrolyte incorporated into a polymeric


13

Page 15 of 28

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


composite was previously reported to affect the overall charge of the material.12 As such, a series of polymer composite inks with varying polyelectrolyte concentrations were prepared. For clogfree considerations, the concentrations of PDADMAC and PSS incorporated in the composite inks used to fabricate charge mosaics were 0.1 M (3.1 mPa s) and 0.5 M (6.12 mPa s), respectively. Fluorescent dyes were added to the composite inks for direct observation of the printed domains using fluorescent microscopy. 5 µM of FITC-PAH was mixed into the positively charged PVA/PDADMAC ink. This dye appears green in color in the fluorescent micrographs. 5 µM of Cy5 was added to the negatively charged PVA/PSS ink. This dye appears purple in color in the fluorescent micrographs. The concentrations of the dyes are adequate for imaging purposes but low enough not to affect the overall charge of the composite materials (Table S1). The compositions of the polymer composite inks used for printing charge mosaic membranes were 1% (by weight) PVA/0.1 M PDADAMC/5 µM FITC-PAH and 1% (by weight) PVA/0.5 M PSS/5 µM Cy5. Printing procedure: Predesigned patterns were written in scripts and printed using a Jetlab® 4 xlA system (MicroFab Technologies), which uses piezoelectric actuation technology to eject the ink droplets. Two fluid channels with 50-µm-diameter orifice were used to inkjet the polymer composite inks. The number of droplets ejected at the same location (defined as number of overprints) was controlled through the preprogramed scripts. Due to their well-defined pore structure and prior experience with these membranes,18 PCTE membranes (pore diameter: 30 nm; membrane thickness: 10 µm; porosity: ~3x108 pores cm-2) were used as structural templates. Prior to printing, the PCTE was fixed onto an in-house vacuum device and a constant vacuum of


14


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 28

12 psig was applied to the PCTE membrane during printing for all experiments. The vacuum device is detailed in a prior publication.18 Membranes functionalized with a single charge type (i.e., negative or positive charge) were fabricated by printing a charged polymeric composite ink of a single type onto the PCTE template. Charge mosaic membranes were formed by printing alternating stripes of positivelycharged and negatively-charged inks. The width of the positively-charged and negativelycharged stripes were varied independently to control the areal fraction of the positively-charged regions on the membrane surface. The minimum value of for the stripe width was ~100 µm. Charge mosaic membranes with 29%, 52%, and 75% of positively-charged regions were printed from written scripts with 100 µm PDADMAC/300 µm PSS, 100 µm PDADMAC/100 µm PSS and 300 µm PDADMAC/100 µm PSS, respectively. Characterizing surface charge of the charge-functionalized membranes: Streaming current measurements were used to determine the sign and magnitude of the charge imparted to the PCTE template by the polymer composite inks. It was also used to determine the overall average surface charge of the charge mosaic membranes. The procedure for measuring the streaming current was described in the prior paper.18 A membrane square (1.5 cm × 1.5 cm) was prepared to fit in a custom built U-tube cell device that measures streaming current. A more detailed description of the device is available in the prior literature.18 Three overprints of either the positively-charged or negatively-charged ink was printed on the PCTE membranes when the effects of polyelectrolyte concentration on surface charge was investigated. The results of the streaming current measurements can be used to calculate the surface charge density of the membranes as demonstrated by Equations S3-S6 and Table S2. These calculations, however, rely on several assumptions regarding the nanostructure of the membrane and the magnitude of the


15

Page 17 of 28

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


surface charge, which is why we chose to report the experimentally measured streaming current values. Fluorescent and electron micrographs of the charge mosaic membranes: The printed mosaic membranes were visualized using a fluorescent microscope (EVOS FL Auto, Thermo Fisher Scientific) equipped with the GFP and Cy5 light cubes. The morphology of the charge mosaic membranes at the nanoscale was characterized using a field emission scanning electron microscope (SEM) (Magellan 400, FEI).18 2.5 nm of Iridium was sputtered on the membrane by a sputter coater (208 HR, Cressington) to prevent sample charging during imaging. Chemical crosslinking of the charge mosaic membranes: A glass chamber containing a beaker of 37% (by volume) hydrochloric acid in water and a beaker of 25% (by weight) glutaraldehyde in water was used as the reactor for vapor-phase crosslinking of the PVA matrix. The glass chamber was covered with a glass plate and the printed membranes were taped onto the top surface of the glass lid. The crosslinking reaction was conducted at 45 °C for 24 hours. Subsequently, the membranes were removed from the glass lid, rinsed in DI water for 1 h, and dried in air. Fourier transform infrared spectroscopy (FTIR): FTIR spectra were acquired using a FT/IR6300 spectrophotometer (Jasco). Membranes of printed PVA mixtures were prepared with and without chemically cross-linking the PVA that was described in the prior section. FTIR was collected on these membrane samples in the range 4000-695 cm–1 with resolution of every 1 cm – 1

and the average of 56 scans was used.

Transport tests: The detailed procedure for measuring the hydraulic permeability and ion rejection of charge-functionalized membranes was detailed in a prior paper.18 Briefly, membranes were put in a stirred cell (model 8003, Amicon), which was filled with water. A


16


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 28

pressure of 4 bar was applied to drive permeation through the membrane. After 2 h, the throughput stabilized, and the solution that permeated through the membrane was collected in a vial that rests on a balance. The mass of the permeate was recorded using Labview software (National Instruments). This data was used to determine the hydraulic permeability of the membrane. In ion rejection measurements, a 0.1 mM solution of potassium chloride was used as the feed solution. A pressure of 4 bar was applied to drive the solution to permeate through the membrane, and the permeate solution was collected in a vial. During filtration experiments, the stirred cell was placed on a stir plate set at 300 rpm to keep the feed solution well-mixed and minimize the influence of concentration polarization. Subsequently, ion chromatography (ICS5000, Dionex) was used to analyze the concentration of potassium ions in the feed (cf) and permeate solutions (cp). These measured values were used to calculate the percent rejection, R, according to Equation 1 !

𝑅 % = 1 − !! ×100 !

(1)

In poly(ethylene oxide) (PEO) rejection experiments, a similar procedure as the ion rejection measurements was used. The results of these experiments could be used to estimate the pore size of the printed membrane. A solution with 10 kg mol-1 PEO dissolved in 1 g L-1 was used as the feed solution. The concentration of PEO in the permeate solution was measured with a Shimadzu TOC-TN Organic Carbon Analyzer. The percent rejection was calculated by Equation 1.

ASSOCIATED CONTENT Supporting Information. Additional figures and tables (PDF). The Supporting Information is available free of charge on the ACS Publications website.


17

Page 19 of 28

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


AUTHOR INFORMATION Corresponding Author * William A. Phillip, Email: [email protected] Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes The authors declare no competing financial interest. ACKNOWLEDGMENT This project was supported by the Indiana Clinical and Translational Sciences Institute, funded in part by grant #UL1 TR001108 from the National Institutes of Health, National Center for Advancing Translational Sciences, Clinical and Translational Sciences Award. P.G. thankfully acknowledges support from the ND Energy Postdoctoral Fellowship Program at the University of Notre Dame. W.A.P. gratefully acknowledges support from the DuPont Young Faculty Award. We would like to thank Prof. Joan Brennecke for use of her viscometer and Prof. Jeremiah Zartman for use of his fluorescent microscope. We thank the Notre Dame Integrated Imaging Facility (NDIIF) and the Center for Environmental Science and Technology (CEST) at Notre Dame; portions of this research were performed with instruments at these facilities. ABBREVIATIONS PCTE,

polycarbonate

track-etched;

PVA,

poly(vinyl

alcohol);

PDADMAC,

poly(diallyldimethylammonium chloride); PSS, poly(sodium 4-styrenesulfonate); FITC-PAH, fluorescein isothiocyanate-labeled poly(allylamine hydrochloride); Cy5, sulfo-cyanine5; PEO, poly(ethylene oxide).


18


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 28

REFERENCES (1)

Werber, J. R.; Osuji, C. O.; Elimelech, M. Materials for Next-generation Desalination and Water Purification Membranes. Nat. Rev. Mater. 2016, 16018.

(2)

Elimelech, M.; Phillip, W. A. The Future of Seawater Desalination: Energy, Technology, and the Environment. Science 2011, 333, 712–717.

(3)

Shannon, M. A.; Bohn, P. W.; Elimelech, M.; Georgiadis, J. G.; Mariñas, B. J.; Mayes, A. M. Science and Technology for Water Purification in the Coming Decades. Nature 2008, 452, 301–310.

(4)

Geise, G. M.; Lee, H.-S.; Miller, D. J.; Freeman, B. D.; Mcgrath, J. E.; Paul, D. R. Water Purification by Membranes: The Role of Polymer Science. J. Polym. Sci. Part B: Polym. Phys. 2010, 48, 1685–1718.

(5)

Geise, G. M.; Paul, D. R.; Freeman, B. D. Fundamental Water and Salt Transport Properties of Polymeric Materials. Prog. Polym. Sci. 2014, 39, 1–24.

(6)

Weinstein, J. N.; Caplan, S. R. Charge-Mosaic Membranes: Enhanced Permeability and Negative Osmosis with a Symmetrical Salt. Science 1968, 161, 70–72.

(7)

Weinstein, J. N.; Caplan, S. R. Charge-Mosaic Membranes: Dialytic Separation of Electrolytes from Nonelectrolytes and Amino Acids. Science 1970, 169, 296–298.

(8)

Fujimoto, T.; Ohkoshi, K.; Miyaki, Y.; Nagasawa, M. A New Charge-mosaic Membrane from a Multiblock Copolymer. Science 1984, 224, 74–76.

(9)

Ishizu, K.; Amemiya, M. Transport of Electrolytes through Charge Mosaic Composite Membranes. J. Membr. Sci. 1992, 65, 129–140.

(10)

Vannucci, C.; Taniguchi, I.; Asatekin, A. Nanoconfinement and Chemical Structure Effects on Permeation Selectivity of Self-assembling Graft Copolymers. ACS Macro Lett. 2015, 4, 872–878.

(11)

Chen, Y.; Cui, Y.; Jia, Y.; Zhan, K.; Zhang, H.; Chen, G.; Yang, Y.; Wu, M.; Ni, H. Preparation of Charged Mosaic Membrane of Sodium Polystyrene Sulfonate and Poly(4Vinyl Pyridine) by Conjugate Electrospinning. J. Appl. Polym. Sci. 2014, 131, 40716– 40724.

(12)

Higa, M.; Masuda, D.; Kobayashi, E.; Nishimura, M.; Sugio, Y.; Kusudou, T.; Fujiwara, N. Charge Mosaic Membranes Prepared from Laminated Structures of PVA-based Charged Layers. 1. Preparation and Transport Properties of Charged Mosaic Membranes. J. Membr. Sci. 2008, 310, 466–473.

(13)

Rajesh, S.; Yan, Y.; Chang, H.; Gao, H.; Phillip, W. A. Mixed Mosaic Membranes Prepared by Layer-by-layer Assembly for Ionic Separations. ACS Nano 2014, 8, 12338– 12345.

(14)

Weinstein, J. N.; Bunow, B. J.; Caplan, S. R. Transport Properties of Charged-mosaic Membranes I. Theoretical Models. Desalination 1972, 11, 341–377.

(15)

Wang, Z.; Elimelech, M.; Lin, S. Environmental Applications of Interfacial Materials with Special Wettability. Environ. Sci. Technol. 2016, 50, 2132–2150.


19

Page 21 of 28

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


(16)

Sollner, K. Mosaic Membranes. Biochem. Z. 1932, 244, 370–381.

(17)

Zhang, Y. Z.; Sargent, J. L.; Boudouris, B. W.; Phillip, W. A. Nanoporous Membranes Generated from Self-assembled Block Polymer Precursors: Quo Vadis? J. Appl. Polym. Sci. 2015, 132, 41683–41699.

(18)

Gao, P.; Hunter, A.; Benavides, S.; Summe, M. J.; Gao, F.; Phillip, W. A. Template Synthesis of Nanostructured Polymeric Membranes by Inkjet Printing. ACS Appl. Mater. Interfaces 2016, 8, 3386–3395.

(19)

Gladman, A. S.; Matsumoto, E. A.; Nuzzo, R. G.; Mahadevan, L.; Lewis, J. A. Biomimetic 4D Printing. Nat. Mater. 2016, 15, 413–419.

(20)

Andres, C. M.; Kotov, N. A. Inkjet Deposition of Layer-by-layer Assembled Films. J. Am. Chem. Soc. 2010, 132, 14496–14502.

(21)

Badalov, S.; Oren, Y.; Arnusch, C. J. Ink-jet Printing Assisted Fabrication of Patterned Thin Film Composite Membranes. J. Membr. Sci. 2015, 493, 508–514.

(22)

Zhang, L.; Liu, H.; Zhao, Y.; Sun, X.; Wen, Y.; Guo, Y.; Gao, X.; Di, C. A.; Yu, G.; Liu, Y. Inkjet Printing High-resolution, Large-area Graphene Patterns by Coffee-ring Lithography. Adv. Mater. 2012, 24, 436–440.

(23)

Chen, S.; Su, M.; Zhang, C.; Gao, M.; Bao, B.; Yang, Q.; Su, B.; Song, Y. Fabrication of Nanoscale Circuits on Inkjet-printing Patterned Substrates. Adv. Mater. 2015, 27, 3928– 3933.

(24)

Sun, K.; Wei, T. S.; Ahn, B. Y.; Seo, J. Y.; Dillon, S. J.; Lewis, J. A. 3D Printing of Interdigitated Li-ion Microbattery Architectures. Adv. Mater. 2013, 25, 4539–4543.

(25)

Sirringhaus, H.; Kawase, T.; Friend, R. H.; Shimoda, T.; Inbasekaran, M.; Wu, W.; Woo, E. P. High-resolution Inkjet Printing of All-polymer Transistor Circuits. Science 2000, 290, 2123–2126.

(26)

Murphy, S. V; Atala, A. 3D Bioprinting of Tissues and Organs. Nat. Biotechnol. 2014, 32, 773–785.

(27)

Martin, C. R. Nanomaterials: A Membrane-based Synthetic Approach. Science 1994, 266, 1961–1966.

(28)

Shimotori, T.; Nuxoll, E. E.; Cussler, E. L.; Arnold, W. A. A Polymer Membrane Containing Fe0 as a Contaminant Barrier. Environ. Sci. Technol. 2004, 38, 2264–2270.

(29)

Lebrun, L.; Da Silva, E.; Metayer, M. Elaboration of Ion-exchange Membranes with Semi-interpenetrating Polymer Networks Containing Poly(Vinyl Alcohol) as Polymer Matrix. J. Appl. Polym. Sci. 2002, 84, 1572–1580.

(30)

Gans, B. J.; Duineveld, P. C.; Schubert, U. S. Inkjet Printing of Polymers: State of the Art and Future Developments. Adv. Mater. 2004, 16, 203–213.

(31)

Xu, N.; Zhou, D.; Li, L.; He, J.; Chen, W.; Wan, F.; Xue, G. Physical Properties of PVA/PSSNa Blends. J. Appl. Polym. Sci. 2003, 88, 79–87.

(32)

Qu, S.; Dilenschneider, T.; Phillip, W. A. Preparation of Chemically-tailored Copolymer


20


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 28

Membranes with Tunable Ion Transport Properties. ACS Appl. Mater. Interfaces 2015, 7, 19746–19754. (33)

Yan, Y.; Wang, L.; Xue, J.; Chang, H. C. Ion Current Rectification Inversion in Conic Nanopores: Nonequilibrium Ion Transport Biased by Ion Selectivity and Spatial Asymmetry. J. Chem. Phys. 2013, 138, 044706.

(34)

Miyaki, Y.; Nagamatsu, H.; Iwata, M.; Ohkoshi, K.; Se, K.; Fujimoto, T. Artificial Membranes from Multiblock Copolymers. 3. Preparation and Characterization of Chargemosaic Membranes. Macromolecules 1984, 17, 2231–2236.

(35)

Phillip, W. A.; Mika Dorin, R.; Werner, J.; Hoek, E. M. V; Wiesner, U.; Elimelech, M.; Mika, D. R.; Werner, J.; Hoek, E. M. V; Wiesner, U.; et al. Tuning Structure and Properties of Graded Triblock Terpolymer-based Mesoporous and Hybrid Films. Nano Lett. 2011, 11, 2892–2900.

(36)

Park, J. C.; Ito, T.; Kim, K.-O.; Kim, K.-W.; Kim, B.-S.; Khil, M.-S.; Kim, H.-Y.; Kim, I.S. Electrospun Poly(Vinyl Alcohol) Nanofibers: Effects of Degree of Hydrolysis and Enhanced Water Stability. Polym. J. 2010, 42, 273–276.


21

Page 23 of 28

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


For Table of Contents Use Only. A Method for the Efficient Fabrication of Multifunctional Mosaic Membranes by Inkjet Printing Peng Gao, Aaron Hunter, Mark J. Summe, and William A. Phillip*


22

ACS Applied Materials & Interfaces Page 24 of 28

1 2 3 4 5 6 7


Page 25 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22




1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21


Page 26 of 28

Page 27 of 28

1 2 3 4 5 6 7 8 9 10 11 12



ACS Applied Materials & Interfaces Page 28 of 28

1 2 3 4


A Method for the Efficient Fabrication of Multifunctional Mosaic

Recommend Documents