Polyelectrolyte multilayer (PEM) membranes and their use

Abstract

The present invention relates to a polyelectrolyte multilayer (PEM) membrane comprised of at least one bilayer, wherein the bilayer is comprised of a layer of a polymeric polycation and a layer of a polymeric polyanion. Furthermore, present invention relates to methods for the production of these PEM membranes by layer-by-layer deposition and the use of these PEM membranes for the decontamination of liquids, preferably water.

Claims

1. A polyelectrolyte multilayer (PEM) membrane for the filtration of liquids, wherein the membrane is comprised of at least one bilayer, wherein the at least one bilayer is comprised of a layer of a polymeric polycation and a layer of a polymeric polyanion, wherein linked to the backbone of the polycation are functional non-charged groups that do not dissociate in water, wherein hydration of the PEM membrane when contacted with the liquids is reduced through covalent incorporation of functional groups in said polymeric polycation or in said polymeric polyanion, in comparison to a PEM membrane that does not comprise incorporation of functional groups in said polymeric polycation or said polymeric polyanion, wherein the functional groups are selected from the group consisting of sulfone, vinylpyrrolidone, and styrene, wherein the polycation is an amine based polycation selected from the group consisting of poly(diallyldimethylammonium chloride) (PDADMAC), Poly(acrylamide-co-diallyldimethylammonium chloride) (PDADMAC/AM), poly(allylamine hydrochloride) (PAH), poly(ethyleneimine) (PEI), poly(diallyl methyl amine hydrochloride) (PDAMAHC), and a copolymer of 2-propen-1-amine-hydrochloride with N-2-propenyl-2-propen-1-aminehydrochloride (CPPAHC), and wherein the polyanion is selected from the group consisting of poly(styrene sulfonate) (PSS), poly(acrylic acid) (PAA) and poly(vinylsulfonic acid sodium) (PVS).

2. The polyelectrolyte multilayer (PEM) membrane according to claim 1, wherein the polymeric polyanion is comprised of poly(styrene sulfonate) and the polymeric polycation is comprised of poly(diallyldimethylammonium chloride), wherein the poly(diallyldimethylammonium chloride) has covalently incorporated sulfone, resulting in poly(diallyldimethylammonium chloride)-co-sulfone (SPDADMAC).

3. The polyelectrolyte multilayer (PEM) membrane according to claim 1, wherein the membrane multilayer comprises at least two bilayers.

4. The polyelectrolyte multilayer (PEM) membrane according to claim 1, wherein the hydration of the PEM membrane is determined as an increase in membrane thickness of at most 80%, when compared to the PEM membrane that was not contacted with liquid.

5. The polyelectrolyte multilayer (PEM) membrane according to claim 1, wherein the membrane multilayer comprises at least 5 mg/m.sup.2 of polyelectrolyte.

6. The polyelectrolyte multilayer (PEM) membrane according to claim 1, wherein the membrane has a permselectivity of about at least 5 bar.sup.−1.

7. The polyelectrolyte multilayer (PEM) membrane according to claim 1, wherein the membrane has and a permeability of about at least 2 L*m.sup.−2h.sup.−1bar.sup.−1.

8. The polyelectrolyte multilayer (PEM) membrane according to claim 1, wherein the membrane is a nanofiltration, microfiltration, ultrafiltration or reverse osmosis membrane.

9. A filter device comprising at least one polyelectrolyte multilayer (PEM) membrane according to claim 1.

10. A method for the production of a polyelectrolyte multilayer (PEM) membrane of claim 1, the method comprising the step of layer-by-layer deposition of a layer of polymeric polyanion followed by deposition of a layer of polymeric polycation, wherein said polymeric polycation or said polymeric polyanion comprises covalent incorporated functional groups, and wherein at least one bilayer of polymeric polyanion and polymeric polycation are deposited.

11. A method comprising using a polyelectrolyte multilayer (PEM) membrane of claim 1 for removal of contaminants from liquids, comprising the step of filtering liquids by contacting said liquids with said PEM membrane.

12. The method according to claim 11, wherein the contaminants are selected from the group that consist of micro pollutants, salts, chemicals, hormones, pesticides, antibiotics and Endocrine Disrupting Chemicals (EDCs) and mixtures thereof.

13. The polyelectrolyte multilayer (PEM) membrane according to claim 1, wherein the functional groups are sulfone groups.

14. The polyelectrolyte multilayer (PEM) membrane according to claim 1, wherein the functional groups are vinylpyrrolidone groups.

15. The polyelectrolyte multilayer (PEM) membrane according to claim 1, wherein the functional groups are styrene groups.

16. The polyelectrolyte multilayer (PEM) membrane according to claim 1, wherein the hydration of the PEM membrane is determined as an increase in membrane thickness of at most 40% when compared to the PEM membrane that was not contacted with liquid.

17. The polyelectrolyte multilayer (PEM) membrane according to claim 1, wherein the hydration of the PEM membrane is determined as an increase in membrane thickness of at most 25% when compared to the PEM membrane that was not contacted with liquid.

18. The polyelectrolyte multilayer (PEM) membrane according to claim 1, wherein the hydration of the PEM membrane is determined as an increase in membrane thickness of at most 10% when compared to the PEM membrane that was not contacted with liquid.

19. The polyelectrolyte multilayer (PEM) membrane according to claim 1, wherein the membrane is a nanofiltration membrane.

20. The polyelectrolyte multilayer (PEM) membrane according to claim 1, wherein the membrane multilayer comprises at least three bilayers.

Description

(1) The present invention will be further detailed in the following examples and figures wherein:

(2) FIG. 1: Shows the molecular structure of the amine based polycations that can be used in present invention, poly(diallyldimethylammonium chloride) (PDADMAC), poly(allylamine hydrochloride) (PAH), poly(ethyleneimine) (PEI), poly(diallyl methyl amine hydrochloride) (PDAMAHC), and a copolymer of 2-propen-1-amine-hydrochloride with N-2-propenyl-2-propen-1-aminehydrochloride (CPPAHC) and the polycation poly(diallyldimethylammonium chloride) with covalently incorporated non-charged moiety, i.e. a sulfone group resulting in poly(diallyldimethylammonium chloride)-co-sulfone (SPDADMAC).

(3) FIG. 2: Shows the molecular structure of polyanions that can be used in present invention, Poly(styrene sulfonate) (PSS), poly(vinylsulfonic acid sodium) (PVS) and poly(acrylic acid) (PAA).

(4) FIG. 3: Shows a schematic representation of the filtration setup. The feed solution was pumped through eight parallel connected membrane modules. The flow and pressure were adjusted by the valve. Transmembrane pressure and temperature were monitored via meters.

(5) FIG. 4: Shows the multilayer growth of polyelectrolytes PDADMAC/PSS and SPDADMAC/PSS membranes as indicated by the adsorption in moles per square meter. SPDADMAC has a lower molar adsorption than PDADMAC. The lower adsorption of SPDADMAC indicates that per mole PSS, less moles of SPDADMAC compared to PDADMAC are adsorbed. This is also observed when looking at the total amount of moles polyelectrolyte adsorbed.

(6) FIG. 5: Shows the resistance of PDADMAC/PSS and SPDADMAC/PSS multilayers as a function of the number of bilayers. A strong increase in resistance is observed between bilayer 2 and 3, i.e. the permeability is decreasing with increasing amount of bilayers, the membrane pores are closing at this number of bilayers. Furthermore, it is observed that SPDADMAC results in a higher resistance and therefore a lower permeability after 7 bilayers than PDADMAC.

(7) FIG. 6: Shows the PDADMAC/PSS and SPDADMAC/PSS membrane retention performance. Membrane retention experiments were carried out using four different salts; MgSO.sub.4, MgCl.sub.2, Na.sub.2SO.sub.4 and NaCl. The left column bar is the polycation terminated layer (5.5 bilayer) and the right column bar the polyanion terminated layer (6th bilayer). Especially, SPDADMAC/PSS shows a high retention % for both terminating layers (in contrast to PDADMAC/PSS) and so is much less/not dependent on the terminating layer. For SPDADMAC no clear charge based separation mechanism was observed. An extra sulfone group in the polyelectrolyte chain results in a much higher retention, probably caused by a denser configuration of the layers.

(8) FIG. 7: Shows the differences in permeability and permselectivity of SPDADMAC/PSS and PDADMAC/PSS membranes. In the case of PDADMAC and SPDADMAC, a roughly identical permeability results in a much higher permselectivity for SPDADMAC (about 10 bar-1). This effect can be explained by a lower charge per monomer in the case of SPDADMAC and therefore less overcompensation of charge within the layer and resulting in less swelling. This results in an improvement of the membrane performance without giving up on the permeability.

(9) FIG. 8: Shows the potential of membranes for micropollutant removal (retention %). SPDADMAC/PSS and PDADMAC/PSS membranes were tested on retention performance of a selection of four common micropollutants; Atenolol, Bisphenol A, SMX and Naproxen. Filtration measurements are obtained by HPLC. In FIG. 8A, the retention results for PDADMAC/PSS are shown, in FIG. 8B the results for SPDADMAC/PSS.

EXAMPLES

(10) Materials

(11) Poly(allylamine hydrochloride) (PAH, Mw 150,000, 40 wt. %); Poly(diallyldimethyl ammonium chloride)-sulfur dioxide copolymer (SPDADMAC, Mw 4,000, 40 wt. %); Copolymer of 2-propen-1-amine-hydrochloride with N-2-propenyl-2-propen-1-amine-hydrochloride (CPPAHC, Mw 100,000, 40 wt. %) and Poly(diallyl methyl amine hydrochloride) (PDAMAHC, Mw 20,000, 50 wt. %) were purchased from Nittobo Medical Co. Poly(ethersulfone) (PES, Ultrason E 6020 P) and Sulfonated poly(ether sulfone) (SPES, Ultrason GM0559/111) were obtained from BASF. Glycerol (EMSURE, 85% Reag. Ph Eur) was obtained from Merck Millipore.

(12) The following chemicals were purchased from Sigma Aldrich and used without further modification. Poly(diallyldimethylammonium chloride) (PDADMAC, Mw 200,000-350,000, 20 wt. %); Polystyrene sulfonate (PSS, Mw˜200,000, 30 wt. %); Branched polyethyleneimine (PEI, average Mw˜25,000 by LS, average Mn˜10,000 by GPC); Linear polyethyleneimine (PEI(L), average Mn 10,000); Poly(vinylsulfonic acid, sodium salt) solution (PVS, 25 wt. %); Sodium chloride (NaCl, BioXtra, ≥99.5%, (AT)); N-Methylpyrrolidine (NMP, 97%); Glycerine (99.5% HP); Magnesium sulfate heptahydrate (MgSO4, ACS≥99.0%); Magnesium chloride hexahydrate (MgCl2, ACS reagent 99.0-102.0%); Sodium sulfate (Na2SO4, ACS reagent ≥99.0% anhydrous granular); Atenolol (≥98% (TLC)); Bisphenol A; Sulfamethoxazole (VETRANAL); Naproxen.

(13) Polyelectrolyte multilayers were grown on a membrane and its performance was tested by means of permeability, salt retention and micropollutant retention tests on laboratory scale. The systematic approach used in this study gives a detailed insight into the influence of specific polyelectrolytes on multilayer growth and the performance of PEM nanofiltration membranes.

(14) Multilayer Growth of Hollow Fibre NF Membranes Using the LbL Method

(15) Multilayer growth on the hollow fibre membranes was carried out as follows. Membranes were immersed in a polyanion solution (0.1 g.Math.L.sup.−1, 0.05 M NaCl) for 20 minutes. Thereafter three rinsing steps (5, 10 and 5 minutes respectively) in 0.05 M NaCl were conducted. Then, the membranes were immersed in a polycation solution (0.1 g.Math.L.sup.−1, 0.05 M NaCl) for 20 minutes and afterwards again rinsed by the three rinsing steps in 0.05 M NaCl. This sequence was repeated for 7 times in order to obtain 7 bilayers. Each time, after the second rinsing step, three membrane samples were removed for characterization purposes. All membrane samples were stored in a glycerol/water mixture (15/85 wt. %) for 4 hours and then dried overnight.

(16) Single PEM coated membrane fibres were potted in a 6 mm diameter module with a fibre length of approximately 165 mm for filtration experiments. Each fibre has an inner diameter of 0.68 mm resulting in a total membrane area of about 3.5.Math.10−4 m2 per module.

(17) Membrane Characterization and Membrane Performance

(18) The polyelectrolyte systems (polycations/polyanions) studied were: PAH/PSS, PAH/PVS, PDADMAC/PSS, SPDADMAC/PSS, PEI/PSS, CPPAHC/PSS and PDAMAHC/PSS. The pure water flux was measured with DI water in cross-flow mode at a trans-membrane pressure of 2 bar. To measure the salt retention, a cross-flow through the fibres was applied using an in-house build set-up (FIG. 3). The cross-flow velocity of the feed through the fibres was set at 1.0 m.Math.s.sup.−1. This corresponds to a Reynolds number of approximately 678, which is in the laminar regime. The salt concentration was measured with a WTW cond 3310 conductivity meter. The retention was based on the ratio between the feed and per mate concentrations.

(19) The performance of a membrane used in water filtration is determined by two main factors: water permeability and salt retention. Using equations (1), (2) and (3), the permeability and resistance were determined. For each data point, the average of three different membranes was taken unless stated otherwise. Also for calculating the error bars, the standard deviation between these three different membranes was taken.

(20) The water permeability can be determined by measuring the volumetric flow through the membrane per unit of the membrane area (A), time (t) and applied pressure (ΔP) (equation 1). The permeability is a measure for the ability of the membrane to transport a liquid.
Lp=Q/(A*ΔP)  (1)

(21) Here Lp is the permeability (L.Math.m.sup.−2.Math.h.sup.−1.Math.bar.sup.−1), Q is the volumetric flow (L.Math.h.sup.−1), A is the membrane area (m.sup.2) and ΔP the applied pressure (bar). The salt retention can be calculated by using equation (2).
R=(c.sub.f−c.sub.p/c.sub.f)*100%  (2)

(22) Here R is the retention (%), c.sub.f is the feed concentration (mg.Math.L.sup.−1) and c.sub.p is the permeate concentration (mg.Math.L.sup.−1). When the value of R is 100%, complete retention of the solute takes place, whereas with an R-value of 0% both the solute and solvent are passing through the membrane at an equal rate. Besides the retention and permeability, the resistance during filtration is also an important factor characterizing the performance. The resistance can be calculated with equation (3).
R=ΔP/(μ*J)  (3)

(23) Here R is the membrane resistance (m.sup.−1), ΔP the transmembrane pressure (Pa), μ the dynamic viscosity (Pa.Math.s) and J the membrane flux (m.Math.s.sup.−1).

(24) Multilayer Growth of Polyelectrolytes PDADMAC/PSS and SPDADMAC/PSS

(25) It is essential to accurately monitor the growth of the PEM. The hollow fibre geometry has practical limitations to monitor this on the fibre with adequate resolutions. Therefore, the growth of the PEM is studied on model surfaces. Typically these are flat surfaces based on silicon or gold. These surfaces can be primed or chemically pre-treated the unsure a proper growth of the multilayers. On these surfaces, the multilayer growth, thickness, (and thereby also its hydration) can be accurately monitored with techniques like ellipsometry, quart crystal microbalance, atomic force microscopy, or reflectometry. For these studies, reflectometry measurements on silicon wafers were used. These measurements yield the multilayer adsorption in mg/m2.

(26) To be able to compare the two polyelectrolytes in a fair manner, we looked at adsorption in moles instead of masses. For this, the change in mass per adsorption step was calculated and divided by the molar mass of one repeating unit polyelectrolyte to convert to moles. By doing this for each step, alternatingly using the mass for the polycation and polyanion, the increase in moles per layer could be calculated and summing up results in the total adsorbed amount in moles per square meter. The adsorption in moles per square meter for PDADMAC/PSS and SPDADMAC/PSS is shown in FIG. 4 and shows that SPDADMAC has a lower molar adsorption than PDADMAC. The lower adsorption of SPDADMAC indicates that per mole PSS, less moles of SPDADMAC compared to PDADMAC are adsorbed. This is also observed when looking at the total amount of moles polyelectrolyte adsorbed.

(27) PDADMAC is adsorbing 10% more than PSS (measured after 7 layers of polycation respectively polyanion adsorbed), whereas in the case of SPDADMAC/PSS, SPDADMAC is adsorbing 20% less than PSS. This difference could possibly be caused by the lower charge density. From literature it is known that PDADMAC is overcompensating the charge within the layer and therewith after a few bilayers, the charge of the multilayer stays positive, despite the addition of PSS. We conclude that SPDADMAC is not overcompensating as much as PDADMAC is doing and the net charge is less positive, neutral or even negative. This results in a multilayer which is more densely packed because less charges are present and less repulsion takes place, which lead to better retention properties in membrane performance.

(28) Resistance as a Function of Number of Bilayers

(29) After coating the polyelectrolyte multilayers on hollow fibre membranes and potting them into modules, permeability measurements were performed. By measuring the permeate volume over time and applying equation (2), permeability results (in L.Math.m.sup.−2h.sup.−1bar.sup.−1) with increasing amount of bilayers were obtained. It is shown that the permeability is decreasing with increasing amount of bilayers with a sudden drop between 2 and 3 bilayers.

(30) The influence of polyelectrolyte on the membrane performance specified as resistance will be studied. With the permeability, also the resistance could be calculated using equation (3). In FIG. 5 the resistance of PDADMAC/PSS and SPDADMAC/PSS multilayers are shown. A strong increase in resistance is observed between bilayer 2 and 3. This means that for these polyelectrolyte combinations the pores are closing at this number of bilayers. Furthermore, it is observed that SPDADMAC results in a higher resistance and therefore a lower permeability after 7 bilayers than PDADMAC (6.9 versus 9.4 L.Math.m.sup.−2h.sup.−1bar.sup.−1). PDADMAC/PSS shows a clear odd-even effect, which means that the membrane permeability increases when an additional layer of PDADMAC is coated and decreases when a PSS layer is applied. The higher swelling degree of PDADMAC compared to PSS will results in more open layers with a lower resistance and higher permeability. No clear odd-even effect is shown for SPDADMAC, although a little odd-even effect is observed which is oppositely to the one for PDADMAC. This suggests that SPDADMAC is swelling much less than PDADMAC. Comparing the resistance results, we can conclude that the less adsorbed SPDADMAC layer results in a layer with a higher resistance than PDADMAC, meaning that it is more closely packed than PDADMAC.

(31) PDADMAC/PSS and SPDADMAC/PSS Retention

(32) In order to study the retention of the membranes, experiments were carried out with salts. Salt retention gives a impression about the quality of the membrane performance. To be able to study the mechanism of separation for the salts, four different salts were studied: MgSO4, MgCl2, Na2SO4 and NaCl. In FIG. 6 the results for salt retention measurements for all the polyelectrolytes are shown. The left column bar is the polycation terminated layer (5.5 bilayer) and the right column bar the polyanion terminated layer (6th bilayer).

(33) Membrane retention of PDADMAC/PSS and SPDADMAC/PSS were compared. Interesting to see is that PDADMAC/PSS already shows some good retention results (depending on the terminating layer, 96.9% for MgCl2, 90.5% MgSO4, 90.9% for Na2SO4 and 46.7% for NaCl) with a clear charged based, or Donnan exclusion mechanism. However, SPDADMAC even shows much better retention results. Especially, SPDADMAC/PSS does show high retention for both terminating layers and so is not dependent on the terminating layer and also does not show a Donnan exclusion mechanism. The retentions obtained for SPDADMAC/PSS were 97.1% for MgCl2, 97.7% for MgSO4, 90.6% for Na2SO4 and 37.7% for NaCl. In the case of SPDADMAC no clear charge based separation mechanism can be observed. Clearly, an extra sulfone group in the polyelectrolyte chain results in a much higher retention, probably caused by a denser configuration of the layers.

(34) Permeability and Permselectivity of PEM Membranes

(35) In order to compare the different PEM membranes, a parameter should be used which is independent on process conditions. This makes is possible to combine permeability and retention results of PEM membranes independently on the thickness of the layer and find the membrane with the best performance. For this purpose, the permselectivity is introduced, which is described as the water permeability over the salt permeability (L/B). Knowing the retention and water permeability and using equation (4), the salt permeability can be calculated. Thereafter the permselectivity for each polyelectrolyte multilayer membrane can be obtained by dividing the water permeability by the salt permeability. Because the thickness of the multilayer is appearing in both the L and B factor, it is disappearing when dividing by each other and not influencing the performance anymore.
B=L.Math.ΔP((100−R)/R)  (4)
When the permselectivity versus the pure water permeability is plotted (FIG. 7), a classical behaviour is observed; in order to reach a higher selectivity, the water permeability should be lower. With a decrease in water permeability (L), the salt permeability (B) decreases power three times faster and the permselectivity is expected to increase.

(36) In the case of PDADMAC and SPDADMAC, a roughly identical permeability results in a much higher permselectivity for SPDADMAC (about 10 bar.sup.−1). It therefore can be stated that a lower charge per monomer as in the case of SPDADMAC and therewith less overcompensation of charge within the layer and less swelling as observed earlier, results in an improvement of the membrane performance without giving up on the permeability.

(37) Micropollutant Analysis

(38) Next was to investigate the potential of our membranes for micropollutant removal (micropullutant retention). Therefore a selection of membranes was made to test on micropollutant retention. SPDADMAC/PSS and PDADMAC/PSS were tested in order to see if there is a difference in performance between the two membranes. In FIG. 8 the results for the filtration measurements are shown, obtained by HPLC.

(39) For the characterization of the NF membranes on micropollutant removal, a selection of four common micropollutants has been made for retention measurements; Atenolol, Bisphenol A, SMX and Naproxen. The selection is made such that it covers neutral, positive and negative molecules and contains both hydrophilic and hydrophobic molecules (Table 1). In this way the role of the solute charge can be assessed properly. The micropollutant molecular range is between 200 and 300 g.Math.mol−1, which is in the order of the molecular weight cut off of NF membranes.

(40) TABLE-US-00001 TABLE 1 Properties of the selected micropollutants for retention measurements. pKa (−) Log P (−) Mw (g .Math. mol.sup.−1) Charge Atenolol 9.7 0.43 266.33 + Bisphenol A 10.1  4.04 228.29 0 SMX 2.0/7.7 0.79 253.28 0/− Naproxen 4.2 2.99 230.26 −

(41) Solutions containing 10 mg.Math.L.sup.−1 of micropollutant where filtrated through the prepared membranes for 24 hours before collecting the permeate sample. This time was needed in order to ensure steady state retentions. During filtration a cross-flow velocity of 1 m.Math.s.sup.−1 was applied and a trans-membrane pressure of 2.3 bar. The micropollutant retentions were calculated based on the difference between permeate and concentrate concentrations. For this, the concentrations were determined by means of high pressure liquid chromatography (HPLC). A Dionex Ultimate 3000 U-HPLC system equipped with a RS variable wavelength detector was used. Micropollutant separation was performed on an Acclaim RSLC C18 2.2 mm column (Thermo Scientific) at 40° C., while applying a gradient flow from 95 wt. % H.sub.2O+5 wt. % acetonitrile at pH 2 to 5 wt. % H.sub.2O+95 wt. % acetonitrile at 1 mL.Math.min.sup.−1.

(42) In FIG. 8A, the retention results for PDADMAC/PSS are shown, in FIG. 8B the results for SPDADMAC/PSS. In general, atenolol is retained the highest in the membranes, naproxen is the second highest retained, followed by SMX and Bisphenol A. Micropollutant retention of SPDADMAC/PSS PEM membranes represented in FIG. 8B is increased as compared to the PDADMAC/PSS in FIG. 8A. The positive terminating layer (SPDADMAC), results for each micropollutant in a higher retention. This is attributed to the odd-even effect of SPDADMAC in combination with PSS, where in general terminating with SPDADMAC results in a lower permeation and terminating with PSS in a higher permeation. This is a result of the lower swelling degree of SPDADMAC compared to PSS and therewith denser structure which results in a higher retention. For SPDADMAC/PSS high retention to atenolol is observed (98.2%), and average retention for naproxen (59.9%). The neutral compounds are retained in lowest amount, namely 33.1% for SMX and 30.3% for Bisphenol A. Especially the latter, neutral compounds show much higher retention with SPDACMAC/PSS than the retention with PDADMAC/PSS. Therefore it can be said that SPDADMAC/PSS is retaining micropollutants much better than PDADMAC/PSS, which was expected because also the performance as indicated with permselectivity, was higher for PDADMAC than for SPDADMAC.