Novel Biohybrid Polysulfone Membranes with Physically Immobilized

Subscriber access provided by Warwick University Library

Separations

Novel biohybrid polysulfone membranes with physically immobilized gramicidin for ion exchange applications. kamila szalata, Gaurav Pande, and T Gumi Ind. Eng. Chem. Res., Just Accepted Manuscript • DOI: 10.1021/acs.iecr.8b00661 • Publication Date (Web): 23 Apr 2018 Downloaded from http://pubs.acs.org on April 24, 2018

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 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 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.

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 35 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

Industrial & Engineering Chemistry Research

Novel biohybrid polysulfone membranes with physically immobilized gramicidin for ion exchange applications. Kamila Szałata1, Gaurav Pande2, Tania Gumí1* 1

Departament d’Enginyeria Química, Universitat Rovira i Virgili, Av. dels Països Catalans 26,

43007 Tarragona, Spain (+34977559617, +34 977559621, [email protected]) 2

TERI University, Plot No. 10, Institutional Area, Vasant Kunj, New Delhi, Delhi, India 110070

KEYWORDS: polymer membranes, gramicidin, magnetite nanoparticles, ion transport.

ABSTRACT: The goal of this work was to design a system incorporating gramicidin (gA) into a polysulfone (Psf) membrane. The strategy involved immobilization of gA on the surface of two types of magnetite nanoparticles (MNP) dispersed in polymeric matrix. Membranes have been treated with surfactant Triton X100 (TRX) and surfactant/protein micelles (TRX/gA). MNPs morphology was studied using TEM and their chemical structure was confirmed by FTIR. Membranes morphology was observed under ESEM, water interactions were tested using static Contact Angle and Water Uptake, whereas ion transport was evaluated by means of permeability test and current-voltage experiments. Depending of the type of nanoparticles used, immobilization leaded to either higher or lower membranes hydrophilicity, but did not influenced their

ACS Paragon Plus Environment

1

Industrial & Engineering Chemistry Research 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 35

morphology. Protein immobilization had enhanced ion diffusion properties and membranes selectivity. This novel method to fabricate materials has a tremendous potential to be used for ion exchange applications such as fuel cells or nanofiltration.

1. Introduction Ion exchange membranes are important technological tool in many applications like desalination, clean energy production or chemical recovery.1-3 Those materials demand specific properties like: high ion conductivity and selectivity, very good chemical and thermal stability, good mechanical properties and low production cost.4,5 It is commune to use polymeric materials in order to achieve most of those requirements. Polysulfone (Psf), well known for its outstanding chemical and thermal resistance and also mechanical strength, is a perfect base material for creating such membranes.6,7 Moreover, it has been previously reported that Psf is a hydrophobic polymer, compatible with biomolecules such as tissues and blood proteins.8-10 In nature, ion channels are formed mainly by biogenic molecules which are responsible for ionic balance regulation in life cells. Example of such molecule is a linear hydrophobic polypeptide gramicidin (gA), produced by Bacillus brevis. It is composed of 15 hydrophobic amino acids with the following sequence: formyl-X-Gly-Ala-Leu-Ala-Val-Val-Val-Trp-Leu-Trp-Leu-Trp-LeuTrp-ethanolamine, where X is either valine or isoleucine. The 11th amino acid can be either tryptophan, phenylalanine, or tyrosine, determining gramicidin type: A, B, or C, respectively. This membrane protein forms channels that possess the fastest ion transport rate, in the case of proton transport, it reaches values up to 2•10-9 H+/s11, maintaining selectivity to monopositively charged species.12,13 It was reported that gA can maintain its transport properties in a wide range of


2

Page 3 of 35 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


temperatures, even up to 100°C.14,15 Extended working range to elevated temperatures is its strong advantage over other biomolecules. The desire to mimic perfection of natural systems encouraged us to design a bioartificial membrane which would combine chemical, thermal and mechanical resistance of a polymer with the selectivity and high ion conductivity of a protein. In present work, we designed ion transporting polysulfone membrane with incorporated magnetite nanoparticles and immobilized gramicidin (Psf/MNP/gA). The modification involved physical immobilization of gA on MNP surface, which is considered to be a good strategy for noninvasive attachment of biomolecule to the artificial material. Polysulfone enhances MNP´s stability in acidic media and supports the immobilization of the active groups which are afterwards involved in protein immobilization.16 Two types of functional MNPs: with hydroxyl and amine groups were used to immobilize gA on their surface. Nanoparticles synthesized were characterized both by Fourier Transform Infrared (FTIR) and Transmission Electron Microscopy (TEM) to elucidate their chemical structure and their morphology, respectively. To characterize the obtained membranes morphology we used Environmental Scanning Electron Microscopy (ESEM). Contact angle measurements (CA) was used to investigate materials hydrophobicity and determine their wetting properties. Water uptake (WU) was performed to estimate the swelling properties of the membranes. Furthermore, ion permeability experiments were carried out to collect information about concentration driven diffusion transport. Current-voltage measurements allowed to characterize actual ion conductivity properties of membranes. 2. Experimental work 2.1 Materials


3


Page 4 of 35

Polysulfone (Mw 35,000) in transparent pellet form, N,N-Dimethylformamide (DMF, 99 %) used for flat sheet membrane fabrication were purchased from Sigma Aldrich. Gramicidin from bacillus aneurinolyticus (Bacillus brevis) was purchased from Sigma Aldrich. Salts used for PBS (phosphate buffer saline) preparation and for permeability experiments: sodium chloride, magnesium chloride hexahydrate and potassium chloride were purchased from Sigma Aldrich, potassium phosphate dibasic trihydrate, disodium hydrogen phosphate dihydrate and calcium chloride dihydrate were purchased from Panreac. Hydrochloric acid ≥37% was purchased from Sigma Aldrich. Iron (II) chloride tetrahydrate, Iron (III) Chloride hexahydrate, Ammonium hydroxide 25%, Aminopropyltriethoxysilane (APTES) for magnetite nanoparticles (MNP) fabrication were purchased from Sigma Aldrich. Surfactant Triton X-100 (TRX) was purchased from Sigma Aldrich. All solutions were prepared in MiliQ water from Millipore. 2.2 Methods 2.2.1 Immobilization of gramicidin on polysulfone membrane with magnetite nanoparticles 2.2.1.1 MNP preparation (MNP-OH and MNP-NH2) MNPs were prepared by iron (II) and iron (III) salts co-precipitation in alkaline solution according to protocol described by Cruz-Izquierdo et al.17 A volume of 450 ml of 1M ammonia hydroxide solution was added dropwise to the solution containing 0.72 M FeCl3 and 0.36 M FeCl2 salts, at room temperature. The solution was continuously stirred and nitrogen was continuously bubbled during addition of ammonia hydroxide. The obtained black precipitate was separated by centrifugation. The precipitate was washed three times with MilliQ water and twice with Phosphate buffer saline (PBS). The pellets after centrifugation were left to dry in an oven at 60 oC


4

Page 5 of 35 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


overnight. It was reported that obtained Fe3O4 nanoparticles, contained hydroxyl groups (–OH) on surface, when prepared in the aqueous phase.18-21 The next step was the conversion of -OH functional nanoparticles, to -NH2 functional nanoparticles: 1 ml (per 10 mg of MNP-OH) of 2% Aminopropyltriethoxysilane (APTES) solution in dry acetone was used. As reported

19

, APTES immobilizes directly on the MNP-OH surface

through silanization reaction, introducing -NH2 functional groups. The APTES solution with MNP-OH was stirred with orbital shaker for 24 h at the temperature of 70° C. After stirring, formed brown precipitate was washed again as in the case of MNP-OH, three times with MilliQ water and twice with PBS solution, and left to dry overnight in an oven at 60o C. The dried solid was powdered using an agate mortar. 2.2.1.2 MNP membranes preparation (Psf/MNP-OH and Psf/MNP-NH2) With the purpose to obtain membranes containing MNP, firstly, polymeric solutions were prepared by dissolving 15 wt% of Psf pellets in DMF. The mixtures were stirred at 300 rpm using magnetic stirrer during at least 12 h at room temperature. Afterwards, the polymeric solutions were allowed to stabilize for 24 h, without stirring, to remove air bubbles formed. After that, 25wt% of MNP was added to the solution (final weight ratio 1:3 MNP:Psf) and the mixture was sonicated for 10 minutes. Membrane films were prepared using a casting knife with thickness of 200 µm and films were left overnight to dry (in contact with air) to ensure slow solvent evaporation. The excess of MNPs of the membrane surface was removed by washing the membranes with water. 2.2.1.3 TRX treatment (Psf/MNP-OH/TRX and Pfs/MNP-NH2/TRX)


5


Page 6 of 35

The MNP membranes were immersed in an aqueous or in a phosphate buffer saline (PBS) solution. Both stock solutions (water and PBS) were used in order to check the salts effect on immobilization efficiency. Next, surfactant Triton X100 (TRX) was added. Surfactant concentration was 0.58 mM, above the value of critical micelle concentration (0.29 mM CMC).22 Samples were stirred during 72 h at 300 rpm using magnetic stirrer, at room temperature. 2.2.1.4

TRX/gA

micelles

immobilization

(Psf/MNP-OH/TRX/gA

and

Pfs/MNP-

NH2/TRX/gA) Protein was immobilized by immersing the MNP-OH and MNP-NH2 membranes in an aqueous or phosphate buffer saline (PBS) solution, similarly as in the case of TRX treatment. Next, TRX and gA were added. Surfactant concentration was 0.58 mM. Samples were stirred during 72h at room temperature. It is noteworthy to mention that gA have been already successfully introduced to the micellar systems. Salom et al achieved to form reverse micelles with use of sodium bis(2ethylhexyl)sulfosuccinate (AOT)/isooctane/water, mimicking cell membrane environment in order to study protein conformations. This complex system was also successfully separated via HPLC.23 Sychev et al, with the same purpose, used Triton X-100 to study gramicidin structure by means of circular dichroism, fluorescence and Fourier transform infrared.24 This confirms that micelles reflect good biological lipid bilayer environment and form thermodynamically stable structures with gramicidin. 2.2.2 Plain Psf membrane preparation and modification (Psf/TRX and Psf/TRX/gA)


6

Page 7 of 35 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 Psf membranes were prepared by dissolving 15 wt% of Psf pellets in DMF. The mixtures were stirred at 300 rpm using magnetic stirrer during at least 12 h at room temperature. Afterwards, the polymeric solutions were allowed to stabilize for 24 h, without stirring, to avoid air bubbles formation. Membrane films were prepared using a casting knife with thickness of 200 µm and films were left overnight to evaporate the solvent. 2.3 Characterization methods 2.3.1 Statistical analysis Student´s t hypothesis test was performed on all the collected analytical data using XLStat 2016 software. Statistical method was comparison of means and significance level α was 0.05. 2.3.2 FTIR analysis Fourier transform infrared (FTIR) spectra of MNP´s powders were obtained at room temperature with a FTIR spectrophotometer (680 Plus from Jasco) with a resolution of 4 cm−1 and scanning speed of 2 mm s−1, in transmittance mode. 2.3.3 TEM analysis Transmission Electron Microscopy (TEM) was used to determine the morphology and the size of the nanoparticles. TEM images were obtained by JEOL 1011 at 80kV and the nanosize was confirmed by ImageJ software. For size distribution determination, measures of at least 200 particles have been taken. 2.3.4 ESEM analysis


7


Page 8 of 35

Surface and cross-sectional morphologies of membranes with MNP before and after TRX/gA micelles immobilization were observed at 20 kV with high-vacuum ESEM FEI Quanta 600 apparatus, without sputter coating. Membrane cross-sections were prepared by fracturing in liquid nitrogen. The images were analyzed with ImageJ software to observe the effect of protein immobilization on a pore size and membrane thickness. Values of pore sizes and membrane thicknesses are the average of 3 measurements taken from 3 membranes prepared with the same experimental conditions. For the statistical evaluation, the following null hypothesis were made: (i) the pore size of the Psf/MNP-OH and Psf/MNP-NH2 membranes before TRX/gA immobilization is different than the pore size of the membranes after immobilization (ii) the thickness of Psf/MNP-OH and Psf/MNP-NH2membranes before TRX/gA immobilization is different than the thickness of membranes with immobilized TRX/gA. The p-values below 0.05 are in agreement with those hypotheses. 2.3.5 Static Contact Angle measurement Static Contact Angle analysis was performed on a Dataphysics OCA15EC contact angle analyzer with MiliQ water as a testing liquid. The angle was measured immediately after the drop (3 µm) was released on membrane`s surface. Measurements were repeated 5 times on different areas of 3 different membranes prepared under the same experimental conditions. For the statistical evaluation, the following hypothesis was made: (i) the CA of the non-modified Psf/MNP-OH and Psf/MNP-NH2membrane is more than 5 degrees different from the CA of modified membranes (whether after TRXtreatment or TRX/gA immobilization). The p-values below 0.05 correspond to the samples which are in agreement with this assumption. 2.3.6 Water uptake test


8

Page 9 of 35 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


Membranes were weighted before and after 2, 4, 6 and 24 h of water immersion, in order to ensure complete saturation with the solvent. The surface water drops were removed from wet membranes with the use of filter paper and the mass was measured immediately. Then the membranes were dried in the oven (Nabertherm L9/12/P330) at the temperature of 100° C for 24 h, and weighted. The water uptake was calculated as a difference between the wet membrane mass and dry membrane mass and expressed as a mg of water gain per mg of the membrane. Experiment was repeated 3 times for each membrane prepared under the same experimental conditions. The hypothesis for statistical evaluation was: the amount of water absorbed by membranes after TRX treatment or after TRX/gA micelles immobilization is different than amount of water absorbed by non-modified membrane. The p-Values below 0.05 are in agreement with this hypothesis. 2.3.7 Permeability test Ion diffusion experiments were performed in a Teflon test cell with two compartments, containing the feed and stripping solutions in equal volumes of 220 ml, separated by the tested membrane. For all experiments performed, initial feed solution was 0.1 M HCl aqueous solution and stripping solutions tested were 0.1 M aqueous electrolyte solutions: NaCl, KCl, MgCl2, CaCl2 or pure MiliQ water. The ion transport was monitored by pH change of stripping solution. The pH of the stripping solution was measured every 10 s by an Orion Dual Star pH/ISE Multimeter. From the pH change slope, ion diffusion coefficient and permeability were calculated according to the Fick’s First law 25,26 . In these experiments we determined the permeability value for Na+, K+, Ca2+ and Mg2+ using chlorides of subsequent salts as a stripping solutions and hydrochloric acid was a feed. Exact procedure for calculation of permeability is included in Supplementary Information (Procedure S1). Experiments were repeated 3 times for each different membrane. For the statistical evaluation, the following hypothesis was made: (i) the permeability value of membranes after TRX


9


Page 10 of 35

treatment or after TRX/gA micelles immobilization is different than permeability value of nonmodified membrane. The p-values below 0.05 correspond to the samples which are in agreement with this assumption. 2.3.8 Current-Voltage measurements (CV) Current-voltage (CV) measurements were performed using Autolab PGstat100 in potentiostatic mode with current ranging (automatic) from 100 mA to 100 µA, potential range from 0 V to 5 V, with the step of 0.01 V and scan rate 0.01 V/s. The distance between the reference electrodes (Ag/AgCl) and membrane was 1 cm. The solution volume in each compartment was 200 mL. Tested solutions were 0.1 M concentrated HCl, NaCl, KCl, MgCl2 and CaCl2. The measurements were performed at the ambient temperature approx. 24±1°C. The values of voltage for samples were calculated by subtracting the solution resistance without the membrane. Compensation was calculated according to Equation 1 25: 𝑈𝑐𝑜𝑚𝑝 = 𝑈𝑚 𝑅𝑠𝑜𝑙 𝐼𝑚

(1)

where U is the voltage, R is the resistance, I is the current and the subscripts comp, m and sol refer to compensated value, measured value and solution alone, respectively. The ion selectivity was expressed by means of conductance (reverse resistance) of alkali ion value over proton ion value (Yion/YH+).25 For the statistical analysis following hypothesis have been made: for the membranes after treatment with TRX: the resistance of membranes after TRX treatment is lower than the resistance of unmodified membrane; and for the membranes after TRX/gA immobilization: the resistance of membranes after TRX/gA immobilization is different than the resistance of membranes after TRX treatment. The p-values below 0.05 are in agreement with this hypothesis.


10

Page 11 of 35 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


3. Results and discussion The membrane system developed in this case was designed to create a robust material combining the best characteristics of used components. From one side chemical, mechanical and thermal stability of polysulfone, on the other side efficient and selective ion transport of gramicidin. The first step of our strategy consists of MNP´s functionalization with hydrophilic – OH or -NH2 groups, and then introducing them to the Psf membrane. This process increases the membrane surface affinity to the surfactant/protein micelles, which are spontaneously formed in the aqueous solutions over the critical micelle concentration (CMC) of the surfactant.23,24 MNP are one of the nanomaterials commonly used in recent years in applications like biotechnology, sensors and biomedical devices.19,20,27-29 They have good thermal and chemical stability, mechanical hardness and low electrical losses. Iron in oxide form is naturally abundant and easy to extract. Magnetite nanoparticles are simple to prepare, comparing to polymeric nanoparticles, and exhibit good dispersion capability. Apart of their abovementioned advantages they have been successfully employed for various biomolecules immobilization, due to their noteworthy biocompatibility.30-33 Moreover, they can be reusable, easy to separate and functionalizible, what makes them much more ecofriendly than e.g. conductive silver nanoparticles, which apart of its potential toxicity are much more expensive.34 Izquierdo et al. have reported that recovered activity of enzyme immobilized on MNP was maintained for concentration 0.4 mg protein/mg of MNP, because at higher concentration protein tends to precipitate and resulted in protein aggregates that affects immobilization efficiency.17 Thus, the gA quantity to be added was calculated according to the amount of MNP in the membrane, and the proportion was kept always at 0.4 mg protein to 1 mg of MNP. Table 1 shows all membranes prepared in this study.


11


Page 12 of 35

Table 1. List of all the membranes used in this investigation

Membranes prepared by TRX treatment and TRX/gA micelles immobilization method

Stock solution

Base material

Psf membranes:

H2 O

Psf/MNP-NH2 membranes:

Psf/MNP-OH membranes:

Psf membranes:

PBS

Psf/MNP-NH2 membranes:

Psf/MNP-OH membranes:

No.

Membrane

1

Psf

2

Psf/TRX

3

Psf/TRX/gA

4

Psf/MNP-NH2

5

Psf/MNP-NH2/TRX

6

Psf/MNP-NH2/TRX/gA

7

Psf/MNP-OH

8

Psf/MNP-OH/TRX

9

Psf/MNP-OH/TRX/gA

10

Psf

11

Psf/TRX

12

Psf/TRX/gA

13

Psf/MNP-NH2

14

Psf/MNP-NH2/TRX

15

Psf/MNP-NH2/TRX/gA

16

Psf/MNP-OH

17

Psf/MNP-OH/TRX

18

Psf/MNP-OH/TRX/gA

3.1. MNPs chemical structure and morphology characterization FTIR spectra of magnetite Fe3O4 nanoparticles are presented in Figure 1. Both the broad peak at around 3400 cm-1 and the sharper peak at 1630 cm-1 refers to the hydroxyl groups of MNPOH. Also, an intensive sharp peak assigned to Fe-O bonding around 600 cm-1 is visible.


12

Page 13 of 35 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


Silanization lead to obtain -NH2 functionalized MNPs with slightly visible broad peak around 1640 cm-1 characteristic for amide bond stretching and a clearly distinguishable signal between 9801100 cm-1 which probably comes from Si-O-Fe bonds. This confirm the condensation between hydroxyl groups of the magnetite nanoparticles surface and the ethoxysilane molecule. These findings are in agreement with previously reported results.35-39

Figure 1. FTIR spectra of MNPs: MNP-OH (grey line) and MNP-NH2 (black line).

Figure 2 presents TEM images of both types of MNPs. The average diameter determined was 10.8 ± 2.3 nm in the case of MNP-OH and 10.3 ± 2.6 nm in the case of MNP-NH2. As can be seen, average size was not influenced by silanization reaction. However, introducing -NH2 functional groups to the magnetite nanoparticles, results in obtaining narrower size distribution, what has been also reported previously.40-43


13


Page 14 of 35

Figure 2. TEM micrographs of nanoparticles: MNP-OH (a) and MNP-NH2 (b) with their size distribution. 3.2 Membrane morphology ESEM micrographs of representative membranes are presented in Figure 3. As it can be seen from the cross-section images, all membranes present a sponge-like morphology, with a dense surface and internal voids, as expected in the case of membrane prepared by solvent evaporation. MNPs are homogenously distributed and form a part of polymeric structure and internal pores. Detailed results of the morphological analysis, including internal pores size and membranes thickness are collected in Table S1 of Supplementary Information.


14

Page 15 of 35 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 3. ESEM micrographs of cross sections in magnification x500: a) Psf/MNP-NH2, b) Psf/MNP-NH2/TRX/gA, c) Psf/MNP-OH, d) Psf/MNP-OH/TRX/gA membranes, all presenting sponge-like morphology. The final thickness of the films ranged 94-110 µm. Pictures taken after TRX/gA micelles immobilization look similar to the ones taken before (Figure 3). In both types of membranes, before and after TRX/gA micelles immobilization, some agglomerations of MNPs are visible. Nevertheless, after TRX/gA immobilization, in both types of MNP membranes formulations, the


15


Page 16 of 35

pore size and the membrane thickness remain unaffected, what is indicated statistically (difference is not significant according to the p-value, as shown in Table S1 of Supplementary Information). It implies that the membrane structure is indicated by the proportion and interaction between the solvent and non-solvent in the precipitation bath, and the additives such as micelles do not affect it. According to microscopic analysis neither pores obstruction nor collapsing caused by TRX/gA immobilization process have been observed. Clear appearance of the TRX/gA solution used for immobilization suggests that gA is very well dispersed within the micelles and thus, it can be assumed that gA is homogenously distributed within the membrane film. 3.3 Contact angle (CA) and water uptake (WU) It has been previously reported, that although gramicidin structure is determined mainly by hydrophobic interactions, water also plays an important role in the protein assembling and channels formation.44-45 Thus, in order to extract more information about hydrophilic/hydrophobic properties of the membrane systems CA and WU test were performed. Due to the limited sensitivity of the method, for the statistical analysis we presumed a difference in CA of at least 5 degrees to be meaningful. All results of CA are listed in Table S2 in Supporting Information. Psf/MNP-NH2 membranes, after TRX treatment and TRX/gA micelles immobilization, when prepared in water solution do not exhibit significant change of hydrophilicity: 82.4°, 81.1° and 86.9° subsequently for untreated, TRX treated and TRX/gA treated membrane. However, when the modification is performed in PBS buffer environment, Psf/MNP-NH2 membrane became slightly more hydrophilic (from 83.3° and 80.8° for untreated membrane, to 78.4° and 69.9° for TRX and TRX/gA treated, respectively), what is supported also by low p-values (see Table S3 in Supporting Information). That could suggest that membranes stored in water have higher affinity


16

Page 17 of 35 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 interact with non-micellized surfactant (very low increase in hydrophobicity, caused by attaching hydrophilic head of surfactant and leaving on the surface hydrophobic tie), while after storage in PBS, micelles are more attracted to the surface (hydrophilicity increase could be caused by hydrophilic nature of the micelles, which have hydrophobic core and fully hydrophilic shell). In the case of Psf/MNP-OH membranes, in both preparation solutions, after TRX treatment, the membranes hydrophobicity increases from 73.3° to 80.7° (very low p-values, see Table S2 in Supporting Information). Nevertheless, after TRX/gA micelles immobilization, membranes recover their hydrophilicity (75.1°). This can suggest that in aqueous environments, OH groups of MNPs interact stronger with the non-micellized surfactant, but when TRX/gA micelles are present, MNP-OH immobilize them on the surface. In Figure 4 water uptake experiment data are presented. The results present the general trend, that more hydrophilic membranes after TRX treatment or TRX/gA micelles immobilization swell water faster and more than hydrophobic ones. Only for the Psf/MNP-OH membrane modified in water solution, this tendency was not statistically demonstrated, giving a p-value higher than 0.05 (0.051 and 0.071 after TRX treatment and after TRX/gA micelles immobilization, respectively, see the Table S3 of the Supporting Information). Much lower p-values of other membranes states that amount of water swelled by membranes after TRX treatment or after TRX/gA micelles immobilization is significantly different than the water uptake of non-modified membranes. According to morphology analysis performed by ESEM (Section 3.2), immobilization of TRX/gA does not affect membrane porosity. It has been reported that exposure to the surfactant enhance wetting properties of polymeric membranes and can enhance water uptake properties,46-48 what is observed also in the present case.


17


Page 18 of 35

Figure 4. Water uptake experiment results for Psf/MNP membranes: Psf/MNP-OH membranes prepared in H2O (a) and in PBS (b). Psf/MNP-NH2 membranes prepared in H2O (c) and in PBS (d). Columns highlighted with * refers to the samples for which obtained p-value was

Novel Biohybrid Polysulfone Membranes with Physically Immobilized

Recommend Documents