Advances in Colloid and Interface Science 161 (2010) 2–9
Contents lists available at ScienceDirect
Advances in Colloid and Interface Science j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c i s
Plasma nanostructuring of porous polymer membranes Marek Bryjak ⁎, Irena Gancarz, Katarzyna Smolinska Department of Polymer and Carbon Materials, Wroclaw University of Technology, 50-370 Wroclaw, Poland
a r t i c l e
i n f o
Available online 1 October 2010 Keywords: Membrane modification Molecular architecture Plasma treatment
a b s t r a c t Several methods for membrane modification have been presented. Chemical modification of a neat polymer followed by membrane formation and modification of just formed membranes have been compared to plasma action. The following plasma modes are discussed in detail: treatment with non-polymerizable gases, treatment with vapors and plasma initiated grafting. Some examples of modified membrane properties are given. Finally, it was concluded that plasma treatment offers the fastest, environment friendly and versatile method that allows tailoring brand new membranes. © 2010 Elsevier B.V. All rights reserved.
Contents 1. 2. 3. 4.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . Polymer derivatization followed by membrane preparation . . . . Polymer blending . . . . . . . . . . . . . . . . . . . . . . . . Surface modification of membrane. . . . . . . . . . . . . . . . 4.1. Chemical modification . . . . . . . . . . . . . . . . . . 4.2. Plasma modification . . . . . . . . . . . . . . . . . . . 5. Plasma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1. Microwave plasma device . . . . . . . . . . . . . . . . 5.2. Dielectric barrier discharge chamber . . . . . . . . . . . 6. Modes of plasma action . . . . . . . . . . . . . . . . . . . . . 7. Plasma of non-polymerizable gases . . . . . . . . . . . . . . . 7.1. Carbon dioxide plasma . . . . . . . . . . . . . . . . . . 7.2. Nitrogen plasma . . . . . . . . . . . . . . . . . . . . . 7.3. Ammonia plasma . . . . . . . . . . . . . . . . . . . . 7.4. Summary for plasma treatment of non-polymerizable gases 8. Plasma of polymerizable molecules . . . . . . . . . . . . . . . 8.1. Allyl alcohol plasma . . . . . . . . . . . . . . . . . . . 8.1.1. Plasma polymerization of allyl alcohol . . . . . . 8.2. Amine plasma . . . . . . . . . . . . . . . . . . . . . . 8.2.1. Butylamine and allylamine plasma polymerization 8.3. Summary for plasma treatment by polymerizable molecules 9. Plasma induced grafting on membrane surface . . . . . . . . . . 9.1. Grafting of acrylic acid . . . . . . . . . . . . . . . . . . 10. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
2 3 3 3 3 3 3 4 4 4 4 4 4 5 5 5 6 6 7 7 7 7 8 8 8 8
1. Introduction
⁎ Corresponding author. E-mail address:
[email protected] (M. Bryjak). 0001-8686/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.cis.2010.09.004
Various polymers can be used for membrane preparation. However, due to limited number of available materials it is not expected to have any significant increase of a variety of offered
M. Bryjak et al. / Advances in Colloid and Interface Science 161 (2010) 2–9
membranes. Ultracki [1] predicted more than fifteen years ago that the world-wide production of polymers should reach the saturation level. The rough estimation shows at 200–250 million of metric tones production per annum as the value that has not changed for last decade. What is more, any large scale production of new polymers has not been commercialized lately and it is not foreseen in the nearest future. These obstacles have forced material scientists for a search of methods that can offer a wide diversity of prepared membranes. In the case of polymer materials the task is not so difficult as organic materials are modified easily. There are two ways to obtain new membranes: i) to modify a polymer and then prepare a membrane from it, and ii) to prepare a membrane first and then modify it. As the first way needs some separate studies for each derivative (including establishing the procedure of membrane formation), the second one seems to be much easier to deal with. There are some ready membranes on the market now and the material scientists' effort can be limited to alteration of their properties. This approach is fast and less complicated and, what is also important, it offers a whole bunch of new membranes based on one starting matrix. For simplicity, the membranes obtained according to the last procedure can be called a nanostructured or composite membrane to point out that their unique properties result from alteration of the surface layer structure. The authors' intention is to show some methods for membrane modification. In the sections below, these methods are described and some examples of membrane properties are presented. 2. Polymer derivatization followed by membrane preparation From the methodological point of view, this method cannot be included to formation of nanostructured membranes. After modification of a polymer, a process of polymer membrane formation is similar to that used for the starting material. Solvent, non-solvent and modifier(s) should be selected by the try-and-error method before someone decides to prepare a series of new membranes and preparation protocol is not changed: the modified polymer is dissolved in a good solvent and the membrane is formed after its coagulation in a precipitation bath. As an example, the procedure for preparation of sulfonated polysulfone membranes can be given [2]. Here, polysulfone was first sulfonated by chlorosulfonic acid in a chloroform solution and the obtained derivative was used to form porous membranes by phase-inversion. When this membrane was used in a micelle-enhanced ultrafiltration process it was possible to reject up to 80% of boron entrapped in Rokwin 80 micelles when the unmodified membranes did it with 10% extent. In the described studies a surfactant was applied in a 10-time higher concentration than its critical micellar concentration. The data show the importance of repulsion forces between membrane and particles that improve the rejection by modified membranes. 3. Polymer blending When the preparation of porous membranes from the derivatized polymer seems to be too expensive, the use of polymer blends offers a much rational method. When the neat polymer and its derivative are dissolved in the same solvent, they precipitate with phase separation in the non-solvent bath with different extent. Hence, the surface layer of the obtained membrane should be enriched with one of the components. Such situation appeared when membranes were prepared from the mixture of polysulfone and sulfonated polysulfone [3]. Surface tension of membranes, calculated by means of advancing and receding contact angles of water, showed that the surface had a patch-like structure with the larger excess of sulfonated polysulfone. It was shown that the protein fouling properties for membranes obtained from sulfonated polysulfone were the same as the properties for membranes prepared from 30:70 and 50:50 derivative:neat polymer blends. Hence, without losing the membrane performance,
3
it is possible to prepare membranes from the blend containing only a small part of the derivative. 4. Surface modification of membrane As was emphasized before, the most interesting case of membrane modification is related to the change of surface character that can be achieved by attaching some functional groups to the membrane surface. Such procedure can be called ‘molecular architecture’ or surface nanostructuring. This operation can be carried out mainly by chemical or plasma modifications, by UV irradiation, by gamma bombardment and so on. 4.1. Chemical modification When a membrane is formed, it is not difficult to alter its surface character where some chemically reactive groups are located. Such approach was used for porous polyethylene membranes obtained from gel-like interpenetrating polymer network membranes designed for electrodialysis [4]. The polystyrene domains of gel membranes were removed by means of the Fe(II)/H2O2 system. During this process, some amounts of unsaturated bonds were formed on the pore surfaces. They were subjected to chorosulfonation (with chlorosulfonic acid in dichloroethane) and hydrolyzed in a NaOH solution. Finally, the surface of the membrane was enriched with sulfonate groups that changed the character of membranes. The surface sulfonated membranes as well as its off-charge analogues were tested in milk filtration. It was found that both membranes were fouled completely after 4 h of the process. However, the alternating wash with 0.1M NaOH and 0.1M HCl allowed to regenerate modified membranes in 95–99%. In the case of off-charge analogues, the membrane was regenerated in the level of 15–18%. What was more, the regeneration ratio of modified membranes was almost the same in the 1st as well as in the 9th filtration cycles. Similar conclusions were withdrawn in the case of hydrolyzed porous polyacrylonitrile membranes. It was proved that prolonged exposure to a 1M NaOH aqueous solution turned the surface nitriles to carboxylic groups [5]. The resultant membranes were less prone for deposition of filtrated protein. 4.2. Plasma modification Plasma treatment of polymer membrane becomes the interesting issue for three reasons: the technique is fast, effective and meets ecological requests for clean technology. Hence, it is understandable why this method has gained an attention for the last decades and why it is still developed successfully for alteration of membrane properties. When plasma acts on the polymer membrane, two competing processes can take place: ablation and deposition. The balance between them depends on the kind of plasma gas used and the applied process parameters. New functional groups are created on the surface layer and they change the character of the membrane. Short review of the plasma methods is presented below. 5. Plasma Plasma is a mixture of electrons, charged ions, and neutral atoms, or molecules, or both. It is formed when the gaseous matter is treated by energy large enough to reorganize the electron structure of atoms/ molecules. There are two methods to deliver energy to the system — by using high temperature or by applying electro-magnetic field. In the first case we deal with high temperature plasma while the second one is called cold plasma. It is commonly used for modification of organic materials. In this case, electron entering the electric field should get energy large enough to remove the next electron from the neutral atom/molecule. As the results the stream of electrons as well
4
M. Bryjak et al. / Advances in Colloid and Interface Science 161 (2010) 2–9
as ionized molecules are obtained. When these species interact to each other plasma is ignited. There are several methods for plasma initiation and the most popular are: low-frequency discharge and electro-magnetic waves of radio and microwave frequencies. 5.1. Microwave plasma device A laboratory microwave plasma device, assembled with a 2.45 GHz frequency generator, a glass reaction chamber and a vacuum line is commonly used for membrane modification. Remote plasma is ignited in a quartz tube and activated molecules are forced to bombard membranes placed in a reaction chamber. Before the modification process starts some parameters should be adjusted. They are as follows: pulse frequency, duty time, plasma power, vacuum and flow rate of applied gas. Fig. 1 shows an example of a microwave plasma device. 5.2. Dielectric barrier discharge chamber Dielectric barrier discharge (DBD) plasma is created by the electrical discharge between two electrodes separated by an insulating barrier. The process uses a high voltage of alternate current with a lower frequency. Usually, DBD devices are made by plate electrodes separated by a dielectric layer of 0.1 mm thickness. As barrier plasma works under atmospheric pressure, the process needs high energy levels to be sustained. The critical point of using DBD plasma is to fit tightly the membrane to electrode and to arrange the smooth flow of gas round it. An example of the barrier plasma device is shown in Fig. 2. 6. Modes of plasma action Plasma treatment of polymer membranes can be carried out in three different modes: a) with non-polymerizable gas molecules. Such plasma shows mostly the etching properties; pores of the membrane are enlarged and surface chemistry is changed. b) with polymerizable vapors. In this case, the plasma polymer is deposited on the membrane surface; surface character is altered and pore diameter is decreased, c) when plasma activates grafting of polymer chains to the membrane surface. For this mode, formation of polymer brushes is expected. Surface changes its character and pores become narrow.
Some exemptions of this simple classification are being observed in practical use of plasma. 7. Plasma of non-polymerizable gases Plasma of non-polymerizable gases changes the nature of the surface by altering chemical groups. However, the most relevant phenomenon is polymer etching resulting in many cases in an increase of the pore diameter. Always, after plasma treatment, the improvement of filtration performance is observed for porous membranes — permeate fluxes get larger value, adsorption of permeant drops down and flux recovery after cleaning is much better in comparison to the untreated membrane. 7.1. Carbon dioxide plasma When microwave plasma is used with oxidizing gases the treated polymer is usually subjected to degradation [6,7]. Plasma of air, oxygen or carbon dioxide is known to etch polymer surfaces to a great extent [8,9]. Polysulfone [10] and poly(ether sulfone) [11] ultrafiltration and microporous membranes [8,12] belong to the most frequently CO2-plasma treated membranes. The strong etching character of CO2 plasma was used to tailor the properties of asymmetric cellulose acetate membranes by reconstruction of chemistry and structure on the top layer [14]. When plasma is used to porous membranes, one may achieve an enlargement of the pore diameter. Hence, the ordinary ultrafiltration polysulfone membrane exposed to CO2 plasma increases the pore dimension with an extent of time for plasma action [13]. Prolonged exposition to the carbon dioxide plasma has resulted in material ablation and increase of the pore diameter (Fig. 3). Plasma also changes the surface character. Some polar groups formed increase the polar component of the surface tension. In this case, the chemical reconstruction of the membrane surface took place at the very beginning of the process and is completed after the first 60 s. The polar contribution to the surface tension reached almost 50% (see Fig. 4) [13]. 7.2. Nitrogen plasma Nitrogen is the plasma medium that can introduce various chemical functionalities onto the surface, making it more hydrophilic and prone for further reactions. After exposition of the membrane to nitrogen plasma, such groups as amine, imine, amide or nitrile could be created [15,16]. In some cases, polymer degradation and etching
Fig. 1. Remote microwave chamber with ignited plasma in a glass tube.
M. Bryjak et al. / Advances in Colloid and Interface Science 161 (2010) 2–9
5
Fig. 2. DBD reactor with ignited plasma.
processes were also observed [17–19]. When one considers the ablation process caused by N2 plasma, one notes its low extent [18]. Pore size distribution for polysulfone membrane exposed to nitrogen plasma is shown in Fig. 5. The weak effect of N2 plasma on the polysulfone membrane is visible by slower changes of the surface tension for modified membranes. If the surface character alters within 1 min for CO2 plasma, N2 plasma needs 5 min to generate the same hydrophilicity on the membrane surface (Fig. 6). According to the preliminary assumptions, the surface of the N2 plasma modified polymer should have a basic nature. However, the results of the surface titration [20] revealed its amphoteric character, with 0.28 μm/m2 basic and 0.06 μm/m2 acidic sites [20]. It seems that this property is related to post reactions when the plasma activated membrane is exposed to the air and some carboxylic groups can be created. 7.3. Ammonia plasma It has been shown that ammonia plasma treatment turned the surface character to a more basic one due to appearance of nitrogen containing moieties [15,20–22]. However, it is difficult to generate plasma using ammonia molecules only. In such case, the presence of argon in the gaseous mixture improves plasma stability and keeps the plasma flame lighted [23]. Hence, in the case of ammonia treatment two kinds of plasma should be evaluated: neat ammonia and ammonia–argon mixture plasma. Fig. 7 shows changes with the pore diameter for both of them. After careful comparison of both figures, one gets a surprising conclusion: NH3 and NH3/Ar plasmas did not cause any degradation. Both plasmas altered the surface character and make it more hydrophilic — they increased the polar component of the surface tension form 0.3 mN/m to 30 mN/m [24]. However, there are some differences in the concentration of functional groups created on the
Fig. 3. Pore size distribution for a) untreated membrane, b) membrane exposed to CO2 plasma for 2 min, and c) membrane exposed for 10 min [from 13].
membrane surface. According to photoelectron spectroscopy [24], the N/C ratio was zero for non-treated polysulfone, 0.113 for ammonia and 0.223 for ammonia/argon plasma treated membranes. The difference appeared also in the forms of functional groups bearing nitrogen atom. When NH3 plasma was applied, the relative fraction of C–N functionalities was 9.2%. After the use of NH3/Ar plasma that value increased to 22.7%. This difference shows that the presence of argon in the mixture of gases is highly profitable — more functional groups are generated on the polymer surface. Besides that argon stabilizes plasma and makes it more homogenous. 7.4. Summary for plasma treatment of non-polymerizable gases Plasma of such simple gases as Ar, N2, O2, CO2, NH3 and their mixtures are profitable for polymer membrane modification. In many cases, the increase of the pore diameter and the widening of pore size distribution point out that polymer ablation is the main process. However, gasses caring oxygen are more effective in polymer degradation. In the case of larger molecules, argon presence in the mixture is needed. It allows to stabilize the process and to make modification more homogenous. The obtained membranes are less prone for fouling and easily pass the regeneration protocol. In all investigated cases, the flux recovery after membrane cleaning was almost 100% while for neat membranes that value varied from 70 to 90%. 8. Plasma of polymerizable molecules Plasma polymerization can be used for modification of polymer surfaces by deposition of the thin polymer film with thickness from
Fig. 4. Alteration of the surface tension for polysulfone membranes treated with CO2 plasma [from 13].
6
M. Bryjak et al. / Advances in Colloid and Interface Science 161 (2010) 2–9
Fig. 5. Effect of the nitrogen plasma action on pore size distribution. a) Untreated membrane, b) membrane after 2 min and c) 10 min of plasma action [from 18].
several nanometers to 1 μm. Such film is usually made of highly crosslinked material and shows good adhesion to the substrate. Plasma polymerization is a complex process that depends on many parameters: reactor geometry, temperature, plasma parameters (power, frequency, and due time), pressure and gas flow rate. Generally speaking, two polymerization processes take place during plasma action: plasma induced polymerization and plasma state polymerization. In the first case, polymerization appears on the surface with easily polymerizable monomers (caring multiple bounds or cyclic structures). In the second case, polymerization processes run in the plasma state when some bonds can break and form radicals or ions that can polymerize and deposit on the membrane surface. Hence, any kind of molecule, even not bearing multiple bonds, can be used in plasma state polymerization. 8.1. Allyl alcohol plasma Plasma polymerization of allyl alcohol was intensively investigated by several researchers [25–29] and a high retention of hydroxyl group (50–70%) was confirmed experimentally. Due to the presence of a double bond in the molecule, its plasma polymerization occurs more readily and faster than deposition of any saturated alcohol. Allyl alcohol however has been rarely used for membrane modification. To
Fig. 7. Pore size distribution of the polysulfone membrane modified with plasma. a) Ammonia plasma, and b) ammonia argon plasma [from 24].
the best knowledge of the authors', allyl alcohol has been applied to a porous ceramic filter for gas separation [30], to poly(vinyl alcohol) membrane used for pervaporation [31] and to polysulfone for enzyme immobilization [32].
Fig. 6. Change of the surface tension for polysulfone membranes treated with N2 plasma [from 18].
8.1.1. Plasma polymerization of allyl alcohol Plasma of allyl alcohol as well as mixture of allyl alcohol with argon was applied for membrane modification [32]. The change of surface energy for membrane samples exposed to both plasmas is shown in Table 1. The scattering of data for neat AllOH plasma proves its instable character. When argon was added to the vapor mixture, plasma became more stable and the changes of the surface character became smoother. The narrow scans on XPS spectra revealed that use of neat AllOH plasma resulted in deposition of the polymer with almost twice more C–OH functional groups than in the case of AllOH/Ar plasma (see Table 2) [32].
M. Bryjak et al. / Advances in Colloid and Interface Science 161 (2010) 2–9 Table 1 Polarity of the surface (part of the polar component in the total surface tension) as a function of plasma action. Time of plasma action [s]
Surface polarity [%] AllOH
AllOH/Ar
0 30 60 90 120 180
2.0 44.4 46.3 35.7 43.4 57.4
2.0 31.4 46.0 54.3 51.2 57.0
Table 2 Relative contents of various functionalities for the polysulfone membrane treated with AllOH and AllOH/Ar plasma [from 33]. Plasma gas
AllOH AllOH/Ar
Relative fraction of functionalities [%] C–C–C
C–OH
C–O–C
C=O
C(O)O
47.7 71.6
34.6 17.1
8.1 11.2
5.9 0
3.8 0
7
ablation can dominate but the most frequently noted process is deposition of the plasma polymer. Depending on the geometry of the plasma chamber, concentration of the polymerizable component, time of deposition and other plasma parameters, different nanostructures with various amounts of retaining monomer structure can be obtained. The spectroscopic studies, XPS and ATR-FTIR, allowed to quantify surface functionalities. The obtained membranes served with good performances for supports to enzyme immobilization, multi-layer NF membranes, dialytic membranes, filtration membranes with improved fouling resistance, membranes for hybrid processes as micellarenhanced ultrafiltration and polymer-enhanced ultrafiltration.
9. Plasma induced grafting on membrane surface Plasma can generate radicals on the polymer surface that are stable in vacuum but can react rapidly when one exposes them to monomer vapor.
8.2. Amine plasma Such amines as allylamine [33–36], vinylpyridine [37,38], ethylene diamine [36,39,40], n-propylamine [36], n-butylamine [33], aniline [41] and N,N-dimethylaniline [42,43] have been used frequently for membrane modification by plasma treatment. These membranes were used in filtration [40], pervaporation [39], reverse osmosis [34,36] or as supports for catalytic membranes [41]. 8.2.1. Butylamine and allylamine plasma polymerization The effect of both kind of plasma on pore size distribution is shown in Fig. 8. It seems that allylamine plasma did not deposit a large amount of polymer on polysulfone support — pore size was not altered significantly, independently with or without Ar. In the case of butylamine plasma, the presence of argon activated the membrane etching process. The neat ButNH2 plasma deposited the polymer layer that made the pore narrow. XPS study allowed a deeper insight into the results of the vapor polymerization process. For butylamine plasma, the N/C ratio was 0.06 for plasma with or without argon, while for allylamine plasma the ratio reached a value of 0.21–0.27. It means, AllNH2 plasma allowed to graft more groups than ButNH2 plasma did [33]. 8.3. Summary for plasma treatment by polymerizable molecules Such vapors as AllOH, AllNH2 or ButNH2 are commonly mixed with Ar when used for membrane modification. In same cases, membrane
However, when the plasma activated membrane is exposed to oxygen or air, some peroxides and hydroperoxides are formed on the surface. They can be employed in graft polymerization of the selected monomer in their solutions.
This process is called plasma induced graft copolymerization or plasma grafting. The grafting density and length of grafted chains can be controlled by the parameters of both: plasma activation and grafting process. The literature dealing with plasma grafting is enormously large. The porous membranes selected according to the used polymer were chosen as a support for plasma modification: poly (vinylidene fluoride) [44,45], polytetrafluoroethylene and polyamide [46,47], polypropylene [48,49], polyacrylonitrile [50,51], polysulfone [52,53], PET [54], poly(phenylene oxide) [55], and so on. The list of monomers being grafted is also very long, from acrylic, methacrylic
Fig. 8. Pore size distributions for ButNH2 (left) and AllNH2 (right) plasma treated membranes. a) Untreated membrane, b) neat vapor plasma, and c) vapor/Ar plasma [33].
8
M. Bryjak et al. / Advances in Colloid and Interface Science 161 (2010) 2–9
Water flux [dm3/m2 h]
350
10. Summary
300 250 200 150 100 50 0 0
2
4
6
Degree of grafting [mmol/g] Fig. 9. Permeate flux of the grafted membrane in relation to the grafting degree.
acid, acrylamide and N-isopropylacrylamide, glycidyl methacrylate and its derivatives, N-vinylpyrrolidone, styrene to derivatives of styrenesulfonic acid and vinylsulfonic acid. Plasma grafting was applied to get cation-exchange [44] or bipolar [45] membrane, to manufacture thermo-responsive membranes [47], to improve cell adhesion [56] or to lower membrane electrical resistance [49]. Thanks to increased wettability, the filtration performance of the membranes is significantly improved — they are less susceptible to protein fouling and easier for cleaning [51–53,55].
Acknowledgement Authors would like to appreciate financial support of the Wrocaw Research Center EIT+ in the frame of project NanoMat — The Application of Nanotechnology in Advanced Materials.
9.1. Grafting of acrylic acid One of the best examples of membrane grafting is grafting of acrylic acid onto a porous polypropylene membrane. Polypropylene is a very hydrophobic polymer with a surface tension about 30 mN m− 1 and very low polar component. Acrylic acid is soluble in water, and a cheap and easy polymerizable monomer that introduces carboxyl functionalities. After plasma grafting, the surface polarity reaches the value of 20% [49]. However, when poly(acrylic acid) is grafted to the microporous polypropylene membrane its molecules plug pores and the modified membrane becomes less permeable. This phenomenon is well seen in Fig. 9. Grafting yield varies with plasma parameters (power, pressure, time, distance from plasma edge in the remote plasma, and sample arrangement) and polymerization parameters (kind of solvent, monomer concentration, and grafting time). It can reach values up to 20 mmol of acrylic acid per 1 g of polypropylene [57]. Celgard 2500 membrane grafted with acrylic acid shows interesting features. First of all, the plugging of the membrane pores by poly(acrylic acid) makes it less permeable for water (see Fig. 9). Membranes with grafting degree higher than 2 mmol/g are practically impermeable. As all prepared membranes were poorly permeable for basic solutions, they showed a much better permeability for acidic solutions. Hence, such membranes can be used as a pH sensitive valve in any kind of self adjusting pH devices. The phenomenon of flux-regulation of polypropylene membranes grafted with acrylic acid is shown in Fig. 10 70
water flux [dm^3/m^2h]
Modification of existing membranes offers preparation of unique separation bodies that can be applied in various processes. It seems that among several methods available today, plasma modification of existing membranes gives the best results. The undeniable benefit of such treatment is its flexibility and environmental friendly character. The process takes place in the gaseous phase with a minimum amount of produced wastes. The very short time of membrane modification is the next profit of the plasma technology. Versatility of the plasma system allows to adapt it to production of new membranes very fast. Finally, the cost of new membrane production should not exceed significantly the cost of membrane matrix preparation. Unfortunately, plasma treatment shows some drawbacks. One of them is its weak repeatability and some problems with scaling-up. Every plasma device has to be optimised separately before its use. The next drawback is related to the use of vacuum in cold-plasma devices. There are some technical problems with carrying out the continuous process in such apparatus. To some extent, the last problem can be bypassed by the use of normal pressure plasmas — like corona or dielectric barrier discharge.
References [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]
60
[30] [31] [32] [33] [34] [35] [36] [37] [38]
50 40 30 20 10 0 2,2
3,1
4
4,1
4,6
7,6
8,8
11,3
pH Fig. 10. Water flux through the Celgard 2500 membrane grafted with 1.6 mmol/g of poly(acrylic acid). Evaluation was performed at 0.01 MPa pressure.
[39] [40] [41] [42] [43]
Ultracki LA. Polym Eng Sci 1995;35:2. Bryjak M, Duraj I, Poźniak G. Environ Geochem Health 2010;32:275. Poźniak G, Bryjak M, Trochimczuk W. Angew Makromol Chem 1995;233:23. Bryjak M, Trochimczuk W. Angew Makromol Chem 1993;207:111. Bryjak M, Hodge H, Dach B. Angew Makromol Chem 1998;260:25. Yasuda H. Plasma polymerization. Orlando: Academic Press; 1985. Pal S, Ghatak SK, De S, DasGupta S. J Membr Sci 2008;323:1. Yu HY, Xie YJ, Hu MX, Wang JL. J Membr Sci 2005;254:219. Dattatray S, Wavhal DS, Fisher ER. Desalination 2005;172:189. Poncin-Epaillard F, Chevet B, Brosse J-C. Makromol Chem 1991;192:1589. Pal S, Glhatak SK, De S, DasGupta S. Appl Surf Sci 2008;255:250. Yu HY, Liu LQ, Tang ZQ, Yan MG, Gu JS, Wei XW. J Membr Sci 2008;310:409. Gancarz I, Bryjak M, Pozniak G. Eur Polym J 1999;35:1419. Riekerink MB, Engbers GHM, Wessling M, Feijen J. J Coll Interf Sci 2002;245:338. Kull KR, Stehen ML, Fisher ER. J Membr Sci 2005;246:203. Wagner AJ, Fairbrother DH, Reniers F. Plasma Polym 2003;8:119. Yu HY, He XC, Liu LQ, Gu JJ, Wei XW. Water Res 2007;41:4703. Gancarz I, Pozniak G, Bryjak M. Eur Polym J 2000;36:1563. Villeger S, Sixou M, Durand J, Ricard A. J Phys D Appl Phys 2006;39:3826. Bryjak M, Gancarz I, Pozniak G. Langmuir 1999;15:6400. Muller M, Oehr C. Surf Coat Technol 1999;116–119:802. Yu HY, Hu MX, Xu ZK, Wang JL, Wang SY. Sep Pur Technol 2005;45:8. Benrejeb S, Tatoulian M, Khonsari F, Durand NF, Martel A, Lawrence JF, Amouroux J, Legoffic F. Anal Chim Acta 1998;376:133. Bryjak M, Gancarz I, Pozniak G, Tylus W. Eur Polym J 2002;38:717. Watkins L, Bismarck A, Lee AF, Wilson D, Wilson K. Appl Surf Sci 2006;252:8203. Ameen AP, Short RD, Ward RJ. Polymer 1994;35:4382. Rinsch CL, Chen X, Panchalingam V, Savage CR, Wang YH, Eberhart RE. Langmuir 1996;12:2995. Yoshimura K, Hozumi K, Tatsuta T, Sawai M, Tsuji O. J Appl Polym Sci 1996;9:1033. Kurosawa S, Hirokawa T, Kashima K, Aizawa H, Han DS, Yoshimi Y. Thin Solid Films 2000;374:262. Ley V, Kruzic AP, Timmons RB. J Membr Sci 2003;226:213. Rafik M, Mas A, Guimon C, Elharfi A, Schue F. Polym Int 2003;52:1222. Gancarz I, Bryjak J, Bryjak M, Pozniak G. Eur Polym J 2003;39:1615. Gancarz I, Poźniak G, Bryjak M, Tylus W. Eur Polym J 2002;38:1937. Tran DT, Mori S, Tsuboi D, Suzuki M. Plasma Process Polymer 2009;6:110. Tran DT, Mori S, Suzuki M. Thin Solid Films 2008;516:4384. Kim I, Kim H, Soo S. J Membr Sci 2006;286:193. Meng JZ, Jiang Y, Jie Z, Yicai S. Surf Rev Let 2009;16:297. Koji M, Satoshi C, Yasutoshi I, Takeshi A, Masao M, Kenji K, Zempachi O. Thin Solid Films 2008;516:3309. Yu CH, Kusumawardhana I, Lai JY, Liu YL. J Coll Interf Sci 2009;336:260. Liu B, Wu VH, Wu ZY, Wang H, Shen GL, Yu RQ. Sens Actuators B Chem 2005;104: 186. Demidova EN, Drachev A, Grigoreva GA. Russ Electrochem 2009;45:533. Ktavets L, Dmitriev S, Gilman A, Drachev A, Dinescu G. J Membr Sci 2005;263:127. Kravets L, Gilman A, Drachev A. High Energy Chem 2005;39:114.
M. Bryjak et al. / Advances in Colloid and Interface Science 161 (2010) 2–9 [44] [45] [46] [47] [48] [49] [50] [51]
Jian C, Jiding L, Cuixian C. Plasma Sci Technol 2009;11:42. Li SD, Wang CC. J Membr Sci 2008;318:429. Chang Y, Cheng TY, Shih YJ, Lee KR, Lai JY. J Membr Sci 2008;323:77. Tu CY, Liu YL, Lee KR, Lai JY. J Membr Sci 2006;274:47. Choi EY, Bae B, Moon SH. J Appl Polym Sci 2007;105:2314. Ciszewski A, Kunicki J, Gancarz I. Electrochim Acta 2007;52:5207. Chen J, Li J, Zhao ZP, Wang D, Chen CX. Surf Coat Technol 2007;201:6789. Zhao ZP, Li J, Wang D, Chen CX. Desalination 2005;184:37.
9
[52] Wavhal DS, Fisher ER. J Membr Sci 2002;209:255. [53] Gancarz I, Poźniak G, Bryjak M, Frankiewicz A. Acta Polym 1999;50:317. [54] Xie R, Zhang SB, Wang HD, Yang M, Li PF, Zhu XL, Chu LY. J Membr Sci 2009;326: 618. [55] Pozniak G, Gancarz I, Tylus W. Desalination 2006;198:215. [56] Lopez-Perez P, Marques AP, da Silva RMP, Pashkuleva I, Reis RLJ. J Mater Chem 2007;17:4064. [57] Gancarz I, Kunicki J, Ciszewski A. Chem Listy 2008;102:1467.
