Performance study of isoporous membranes with tailored pore sizes

Author’s Accepted Manuscript Performance study of isoporous membranes with tailored pore sizes Juliana I. Clodt, Barbara Bajer, Kristian Buhr, Janina Hahn, Volkan Filiz, Volker Abetz www.elsevier.com/locate/memsci

PII: DOI: Reference:

S0376-7388(15)30069-7 http://dx.doi.org/10.1016/j.memsci.2015.07.041 MEMSCI13860

To appear in: Journal of Membrane Science Received date: 28 May 2015 Revised date: 15 July 2015 Accepted date: 18 July 2015 Cite this article as: Juliana I. Clodt, Barbara Bajer, Kristian Buhr, Janina Hahn, Volkan Filiz and Volker Abetz, Performance study of isoporous membranes with tailored pore sizes, Journal of Membrane Science, http://dx.doi.org/10.1016/j.memsci.2015.07.041 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Performance study of isoporous membranes with tailored pore sizes Authors: Juliana I. Clodta*, Barbara Bajera, Kristian Buhra, Janina Hahna, Volkan Filiza*, Volker Abetza,b a

Helmholtz-Zentrum Geesthacht

Institute of Polymer Research Max-Planck-Str.1 21502 Geesthacht, Germany b

University of Hamburg

Institute of Physical Chemistry Grindelallee 117, 20146 Hamburg, Germany

Corresponding author at: Helmholtz-Zentrum Geesthacht Institute of Polymer Research Max-Planck-Str. 1, 21502 Geesthacht, Germany Tel.: +49 4152 872472, +49 4152 872425; fax: +49 4152 872499 [email protected], [email protected] Abstract This performance study deals with isoporous ultrafiltration membranes made through a combination of self-assembly of amphiphilic block copolymers and the non-solvent induced phase separation process (SNIPS). Ten different polystyrene-b-poly(4-vinylpyridine) (PS-bP4VP) diblock copolymers were used to prepare membranes with pore sizes increasing with the molecular weight of the polymers. The pore diameters of the membranes vary from 17 to 86 nm. Pure water permeances were studied with respect to pore sizes, P4VP content,

thickness of the membranes and flux recovery after protein adsorption. Suitable working conditions were identified and rejection of poly(ethylene glycol) (PEG) molecules with molecular weights between 100 and 1000 kDa were carried out. The characteristics of PS-bP4VP diblock copolymer membranes were compared with a commercially available polyethersulfone ultrafiltration membrane. Keywords Block copolymers, isoporous membranes, tailored pore size, performances, ultrafiltration Graphical Abstract

1. Introduction Ultrafiltration (UF) membranes are common in the field of size selective biomolecule separation[1, 2], water treatment[3-5] and controlled drug release[6]. Conventional production processes for commercially available UF membranes include casting, phase inversion, tracketching, anodizing, sintering and film-stretching. Most types of commercially available UF membranes are limited in their separation performances[7] due to huge deviations in pore size, their tendency to adsorb proteins [8-11], low surface porosity, high production costs or membrane fragility[7, 12]. During the last couple of years a new method for the formation of UF membranes became available utilizing the combination of self-assembly of amphiphilic block copolymers (S) and the non-solvent induced phase separation process (NIPS) called

SNIPS. With this method integral-asymmetric membranes with highly ordered, hexagonally arranged pores can be prepared in a fast one-step process.[13] An alternative method to prepare isoporous membranes from block copolymers includes solvent annealing and selective swelling of one block followed by UV-cross-linking and if desired a secondary swelling.[14, 15] Isoporous membranes made from block copolymers can be seen as a new type of membranes promising enhanced performances, namely both high permeability and selectivity, due to their thin selective layer on top consisting of hexagonally arranged cylindrical pores merging in a sponge-like substructure underneath. In comparison common commercial UF membranes are either limited in pore size distribution leading to low selectivities when made through phase separation techniques[16] or bear uniform size selective pores with low permeability like track-etched membranes. Zydney et al calculated a critical upper bond for the selectivity-permeability relation analogous to the Robeson Plot[17].[18] When experimental data for the rejection of bovine serum albumin (BSA) and calculated solvent flow and solute rejection is fitted to Zydney´s Plot, SNIPS made membranes are superior to commercial ones.[19] Sharper molecular weight cut offs (MWCO) than for commercial membranes made by phase-inversion can be found depending on the block copolymer used for SNIPS made membranes.[20, 21] One of the most used and understood diblock copolymers for the formation of UF membranes through the SNIPS technique is polystyrene-block-poly(4-vinylpyridine) (PS-b-P4VP). [2228] Recent results show that the pore sizes of integral asymmetric membranes made from block copolymers are tunable.[24, 29] As PS-b-P4VP diblock copolymers can be synthesized by sequential anionic polymerization, it is possible to use such diblock copolymers with defined amounts of poly(4-vinylpyridine) (P4VP) and defined molecular weights for the membrane formations. The pores of the membranes downsizes by either decreasing the molecular weight of PS-b-P4VP or the P4VP content as described before. Therefor it is possible to generate pores in the range of 20 to 70 nm with this method.[24] In another

approach recently published by Radjabian et al. miscible blends of PS-b-P4VP with different compositions were used to tailor the pore sizes of the membranes.[30] This important result shows that membranes with defined pore sizes can be produced even if an exactly matching block copolymer for a given desired pore size is not available.. First results concerning the performances of such membranes indicate on the one hand high selectivity through their isoporous surfaces. On the other hand a high permeance is expected due to their high porosity and spongy substructure leading to a low resistance as depicted in Figure 1.

Figure 1: Example and expectations for a membrane made from PS-b-P4VP diblock copolymers: SEM of the isoporous surface (left), selective layer and the spongy substructure (right).

In this report we describe the performances of PS-b-P4VP membranes with different pore sizes. Pure water permeance was studied with respect to pore sizes, P4VP content, thickness of the membranes and theoretical expectations. In order to verify suitable working conditions of the membranes water permeance of selected diblock copolymer membranes was measured in dependence of the transmembrane pressure and temperature. Rejection of poly(ethylene glycol) (PEG) molecules with molecular weights between 100 and 1000 kDa was measured and MWCO for the membranes were evaluated. In order to study their fouling behavior the flux recovery after BSA adsorption was tested for selected membranes with different pore sizes. Additionally, the performance of PS-b-P4VP membranes is compared with a commercially available polyethersulfone membrane.

2. Experimental 2.1. Membranes used in this work Diblock copolymer membranes used in this work were prepared according to a procedure recently published by Rangou et al.[24] Therefore PS-b-P4VP diblock copolymers differing in molecular weight and P4VP content were dissolved in a solvent mixture of THF and DMF with concentrations between19 to 35 wt%. The solvent composition was likewise varied according to the total molecular weight and the weight percentage of the P4VP block in the block copolymer. After stirring for 2 days the solution was directly cast by a home-made casting machine using adoctor blade with a 200 um gap height on a polyester nonwoven support. The solutions were cast on a polyester nonwoven support using a home-made casting machine. The films were left for a certain time on air before immersing them in water. The membranes were dried at 60 °C under reduced pressure before use. The GBL-Membranes were prepared from a casting solution with 10 wt% of GBL. A commercial polyethersulfone membrane (hereafter labeled as PES100) used in this work was purchased from Pall Life Science, supplier number Supor 100 membrane 60310, pore sizes 100 nm, typical water flow rate 5 mL/(cm2min) at 0.7 bar (~4300 Lm-2h-1bar-1) as stated by supplier. 2.2. Characterization Scanning electron microscopy (SEM) of the membranes was carried out on a LEO Gemini 1550 VP at a voltage of 3 kV or 5 kV. The samples were coated with 2.0 nm platinum. Crosssections of the membranes were prepared while dipping the membranes in iso-propanol, freezing in liquid nitrogen and cracked. Average pore size values were determined using the software analySIS (Olympus) on basis of the SEM results. ImageJ 1.46 (Wayne Rasband,

National Institute of Health, USA) was used to determine the average porosity of our membranes. 2.3. Clean water flux measurements (Dead-End) Water flux measurements were performed in dead-end mode using a home-made automatic testing device at transmembrane pressures of 2.0 bar to 2.1 bar at room temperature. The volume was measured gravimetrically every 15 minutes for 7 h. The actual pressure was recorded. The effective membrane area was 1.77 cm2. These studies were conducted employing demineralized water with an electrical conductivity of ≈ 0.055 μS×cm-1. The normalized permeance (P) was calculated by normalizing the flux by the transmembrane pressure, later on simply labelled water flux in this work.

(1)

where

V is the volume of water collected between two mass measurements, A is the

membrane surface area,

t is the time between two mass measurements, and

p is the

transmembrane pressure. Measurements were made on a minimum of 5 samples. For the pressure dependent permeances transmembrane pressures were varied stepwise from 0.2 bar to 3 bar. Each step was measured for at minimum 10 minutes starting from the lowest pressure. The permeance (J) was calculated.

(2)

Temperature-dependent water flux was carried out between 5 and 50 °C at a transmembrane pressure of 2 bars. Therefor the membrane cell was placed in a water bath which was heated up within six hours.

2.4. Fouling Tests Fouling tests were carried out through the measurement of the water flux recovery after protein adsorption. Clean water fluxes were measured for 2 hours at a transmembrane pressure of 2.0 bar. Subsequently the membranes were immersed in a vial with 2 ml of protein solution (1g/L). The vials were closed and shaken at 90 rpm for 24 h at 25 °C to reach equilibrium. The membranes were rinsed two times with 2 ml of clean water for 10 minutes. Clean water fluxes were measured again for 2 hours at a transmembrane pressure of 2.0 bar and the flux recovery was calculated as follows.

(3)

where FRR is the flux recovery ratio in %, P0 is the water flux before protein adsorption and P1 is the water flux after protein adsorption. 2.5. Retention measurements In order to provide a closer look to the separation performance of our membranes, retention measurements with poly(ethylene glycol) (PEG) were carried out. Therefore membranes with different pore size were selected and the retention of different PEGs (Mw ~100 kDa, 200 kDa, 300 kDa, 400 kDa, 600 kDa, and 1000 kDa) were measured as follows. Initially pure water flux measurements were carried out for 3 h at 2 bar transmembrane pressure. Pure water was removed and 50 ml of 0.01 wt.% PEG solution was added. The solution was stirred for 5 minutes and a feed sample was collected before starting the measurement. 20 ml of the PEG solution passed the membrane at 2 bar to let the system reach equilibrium before 4 ml was collected as permeate. Afterwards the PEG solution was removed and clean water flux measurements were carried out again. The retention was calculated by the following equation:

(

)

(4)

where c(p) and c(f) represent the concentration of PEG in the permeate and feed, respectively. Concentration for the feed and permeate PEG solutions were measured by gel permeation chromatography (GPC). GPC measurements were performed at 35 °C in bidistilled water with 0.5 g/L sodium azide using PSS acrylate copolymer SUPREMA Pre, 100 Å and 3000 Å columns, particle size 10 µm at a flow rate of 0.5 mL/min (VWR-Hitachi 2130 pump). A Waters 410 refractive-index detector with a PEG/PEO calibration was used. 3. Results and Discussion 3.1. Membranes used in this work In order to obtain membranes with pore sizes from 17 to 86 nm different PS-b-P4VP diblock copolymers were used in this work. The molecular characteristics of the polymers and resulting geometrical feature of the membranes are listed in Table 1 and the membrane surfaces measured by SEM are shown in Figure S1. For the sake of convenience and clarity the membranes are numbered systematically. The letter M is used as a shortcut of the prepared isoporous block copolymer membrane. M is followed by the precise characterization of the membranes. The first value stands for the average pore diameter and the second value for the average membrane thickness excluding the non-woven support; meaning M17/68: pore diameter 17 nm, thickness 68 µm. Table 1: PS-b-P4VP diblock copolymers and the commercial PES membrane used in this work, used polymer concentration of the casting solution, average pore diameter, porosity, average thickness, and numbering of the membranes. PS-b-P4VP

Mn

PS

P4VP

Wt.%

Ø Pore

(Mem)

diameter

Porosity

Ø Thick

No.

** [kg/

[wt.%]

ness [nm]

[%]

[µm]

mol] PS83P4VP1776

76

83

17

38

~17

13.6

50

M17/50

PS75P4VP25100

100

75

25

27

~25

20.1

47

M25/47

PS83.7P4VP16.3113

113

83.7

16.3

31

~23

18.6

54

M23/54

PS83.7P4VP16.3113

113

83.7

16.3

31

~23

21.1

33

M23/33

PS83.7P4VP16.3113

113

83.7

16.3

31

~23

20.3

24

M23/24

PS86.5P4VP13.5150

150

86.5

13.5

31

~24

14.3

56

M24/56

PS86.5P4VP13.5150

150

86.5

13.5

31

~24

19.7

27

M24/27

PS86.5P4VP13.5150

150

86.5

13.5

31

~24

18.4

11

M24/11

PS86.5P4VP13.5150

150

86.5

13.5

27.5

~24

18.8

18

M24/18 GBL

PS81P4VP19160

160

81

19

26

~34

26.7

48

M34/48

PS74P4VP26162

162

74

26

24

~38

23.2

55

M38/55

PS78P4VP22191

191

78

22

22.5

~42

28.1

41

M42/41

PS80P4VP20252

252

80

20

21

~53

31.5

44

M53/44

PS78P4VP22264

264

78

22

19

~68

38.1

31

M68/31

PS76P4VP24330

330

76

24

15.5

~86

41.6

21

M86/21

PES Membrane

100*

PES100

*data from supplier **concentration of the polymer used for the membrane casting solution

3.2. Clean water flux measurement Clean water flux measurements were carried out for all membranes with respect to the following relations: Influence of pore size Figure 2 displays the average pure water fluxes measured by dead-end filtration of all membranes under investigation to represent the performances of a whole membrane sheet

(approximately 600 cm2) the average fluxes of five small samples (d=2 cm, effective membrane area 1.77cm2 due to O-ring) were measured. The error bars express minimum and maximum deviation and reflect the general varieties of the membrane structure. In general, as expected, the fluxes increase with the pore sizes. Some effects will be discussed in detail including the influence of the P4VP content, the influence of an additive used for the membrane formation and the influence of the membrane thickness. For comparison a commercial polyethersulfone membrane PES100 was measured. The commercial membrane has approximately 10 % higher flux than the diblock copolymer membrane M86/21, but the pores are a bit bigger as well. It can be stated that the characteristics of M86/21 with respect to clean water fluxes is on a competitive basis.

Figure 2: Clean water fluxes of the membranes after one hour. Values display the average fluxes from 5 measurements, the error bar shows minimum and maximum error.

In order to gain a deeper insight into the influence of each membrane property on the water flux, theoretical fluxes were compared with experimental findings. Theoretical volume flow rates of membranes may be calculated by the equation of HagenPoiseuille for a laminar flow in a simple straight cylinder: (4) with r is the radius of the open pore, Δ is the transmembrane pressure, η is the viscosity of water (8.94×10-4 Pa×s at 24 °C ) and L the length of the cylindrical pore. The lengths of the

pores were estimated from SEM cross-sections. The corresponding cross-sections are depicted in Figure S2. The calculation is based on the assumption that the spongy substructure of the membrane is negligible and the normalized permeance is basically hindered by the cylindrical top layer of the membrane. Theoretical water fluxes were calculated as followed taken the porosities of the membranes into account:

(5) with A is the area of the pore corresponding π

2

and the average porosity values used as listed

in Table 1. Figure 3 displays the theoretical and experimental water fluxes of selected membranes (M17/50, M25/47, M23/54, M24/56, M34/48, M38/55, M42/41, M53/44 and M68/31) as a function of pore sizes. Theoretical fluxes are approximately ten times higher than experimental ones. A polynomial fit of the theroretical fluxed (pore lengths L and pore radii r are taken from the SEM images) results in an almost quadratic relationship, as given by equation 5. Experimental fluxes show a weaker dependence with mu her lower deviations of the data from a polynomial fit. Therefor we have to conclude that the influence of each membrane structure to the water flux is more complex as originally expected. The influence of the membrane substructure and its swelling seems to be surprising. On the other hand the pore length does not show a significant effect to the water fluxes taking into account that the cylindrical pore layer on top of the membrane is in general smaller than 1 % of the total membrane thickness.

Figure 3: Comparison of experimental (black) and theoretical (grey) water fluxes in dependence of the pore sizes of selected membranes. Theoretical fluxes were calculated using Hagen-Poiseuilles law considering pore radii and cylindrical pore lengths estimated from SEM images.

Influence of P4VP content Figure 4 depicts clean water flux measurements of membranes made from PS-b-P4VP with a molecular weight of 150-160 kg/mol (right), respectively around 100 kg/mol (left). In all cases the flux increases with the P4VP content of the polymers used for the membrane preparation. The pore sizes and membrane thickness of M25/47 and M23/54 are comparable. Nevertheless M25/47 has an almost four times higher flux compared to M23/54 which could be caused by a bigger P4VP content of the polymer used. Since P4VP swells in water leading to decreased pore sizes an opposite behavior was expected. But, on the other hand the P4VP chains make the inner surface of the membranes more hydrophilic which could result in an increased water flux. As reported before the polymer concentration of the membrane casting solution is higher for block copolymers with a low P4VP content.[24] Herein M25/47 was made from a 27 wt.% polymer solution in contrast to 31 wt.% for M23/54. A higher polymer concentration will lead to a denser spongy substructure of the membrane resulting in higher tortuosity and resistance of the flow. This indication is confirmed by the water flux measurements of M38/55, M34/48 and M24/56 where the polymer solution concentration differs (24, 26, 31 wt.%) and the water flux increases with decreasing polymer concentration.

However a bigger P4VP content will lead to an increased pore size as well and therefor to an increased water flux as it was described before.[24] This aspect plays a more important role for higher molecular weights like Mn ~ 150-160 kg/mol. In this case the P4VP chains are longer, too.

Figure 4: Clean water flux measurement of PS-b-P4VP membranes with different P4VP content. Left: membranes made from diblock copolymers with a molecular weight of 150-160 kg/mol and 13.5, 19 and 25 % P4VP; right: membranes made from diblock copolymers with a molecular weight of around 100 kg/mol, 16 and 25 % P4VP.

Influence of membrane thickness and density of the substructure The influence of membrane thickness to the water flux was studied for membranes made from two different PS-b-P4VP diblock copolymers, namely PS83.7P4VP16.3113 and PS86.5P4VP13.5150 as shown in Figure 5. Every membrane was made from polymer solutions using the same amount of each polymer and solvent system. The thickness of the membranes was successfully set by changing the blade gap height of the membrane casting machine. As it can be seen in Figure 5 the water flux increases with decreasing membrane thickness. However this influence is stronger pronounced in case of the membranes made from PS83.7P4VP16.3113. This result confirms the assumption that the substructure of PS-b-P4VP membranes leads to high flux resistance and thus plays an important role for the membrane performance.

Figure 5: Clean water flux measurements of PS-b-P4VP membranes with different membrane thicknesses using the same diblock copolymer.

In order to produce membranes with more open substructures that may lead to lower resistance and higher fluxes, γ-butyrolactone (GBL) was used as an additive in the production process. Using GBL it was possible to decrease the polymer concentration during membrane formation and simultaneously preserve the isoporous surface of the block copolymer membrane. Figure 6 depicts clean water flux measurements for membranes made from PS86.5P4VP13.5150 using different solvent systems, with (M24/18GBL) and without GBL (M24/11). The clean water flux for M24/18GBL is almost four times higher than for M24/11 although the latter one has the lower membrane thickness. Since M24/18GBL is made from a 27.5 wt.% polymer solution and M24/11 from a 31 wt.% polymer solution the membrane substructure will be more dense-graded for the latter one.

Figure 6: Clean water flux measurements of membranes made from PS86.5 P4VP13.5150 using different polymer concentrations and solvent systems for the membrane production in order to change the porosity of the substructure.

Temperature-dependent water flux and pressure-dependent water permeance of PS-bP4VP membranes In order to test feasible operating conditions for block copolymer membranes temperatureand pressure-dependent water flux measurements were carried out for selected PS-b-P4VP membranes. The water flux increases linearly with rising temperature, compare Figure 7 left. Therefor working conditions between 3 and 50 °C are assumed to be suitable for isoporous PS-b-P4VP membranes. The test was not carried out above 50 °C since it is not possible to heat our home made temperature-dependent water flux measurement device above this temperature.

Figure 7: Temperature-(left) and pressure-(right) dependent measurements for selected PS-b-P4VP membranes.

Figure 7, right depicts the pressure-dependent permeance of M34/48. The permeance of M34/48 increases linear with the pressure from 0.2 to 3 bar. Therefor working conditions for PS-b-P4VP membranes are at minimum within this region. Higher pressures were not tested due to the limitation of our water flux measurement device. It is assumed that the results for the temperature- and pressure dependent water flux measurements can be transferred to all PS-b-P4VP membranes used in this work. Retention measurements In order to test the ability of PS-b-P4VP membranes to reject dissolved molecules, aqueous solutions of PEGs ranging from 100 to 600 kDa were used for retention tests. Figure 8 (left) depicts the results of the retention measurements for selected membranes with different pore sizes. For comparison the commercial PES membrane was tested. All PS-b-P4VP membranes have a sharper MWCO than the PES membrane. The supplier of the PES-membrane specifies gives an average pore diameter of 100 nm. However, based on SEM images we calculated approximately 170 nm for the pore diameter resulting in a higher MWCO compared to the PS-b-P4VP membranes. Figure 8 right shows a magnification of the retention for the PS-bP4VP membranes around the 90 % MWCO. Unfortunately all membranes have a 90 % cut off

between 280-300 kDa, as listed in Table 2 independent of their pore sizes. Therefor the PEG molecules must be rejected by the membrane substructures and not only by the pores. If they were rejected by the pores ranging from 17 to 86 nm in pore diameter much higher differences for the cut-off values would be expected.

Figure 8: Retention of PEGs for selected PS-b-P4VP membranes and a commercial PES membrane (left) and magnification of the diagram around the 90% cut off for PS-b-P4VP membranes (right).

Table 2: Performance and structural characteristics of PS-b-P4VP diblock copolymer membranes and a PES membrane. Membrane

dpore, SEM[a]

MWCO > 90

numbering

[nm]

[kDa]

M17/50

~17

285

M23/54

~23

300

M42/41

~42

290

M68/31

~68

290

M86/21

~86

300

PES100

~172

600

[a] Average pore diameter as determined by SEM image analysis

Fouling behaviour (Fouling Test)

In order to examine the membranes concerning their fouling behaviour flux recovery was tested after adsorption of bovine serum albumin (BSA). As it can be seen in Figure 9 the flux recovery ratio varies from 84 to 91 % (average 88.5 %) for the selected PS-b-P4VP membranes independent on their pore size or membranes thickness. Deviations in the flux recovery ratio from the average value can be caused by the error of the measurement itself or by fouling that can occur during the water flux measurements.[31] The PES membrane has flux recovery ratio of 97 %. As it is stated by the supplier PES membranes are known for their low adsorption of proteins. In this connection they are slightly superior to the PS-b-P4VP membranes that already show defined adsorption of different proteins like hemoglobin and catalase in previous work.[31]

Figure 9: Flux recovery after BSA adsorption of selected PS-b-P4VP membranes and a commercial PES membrane.

Conclusion This work comprises a detailed examination of the characteristics of isoporous block copolymer membranes made from PS-b-P4VP. It is possible to tune the pore sizes of such membranes using specific molecular weights of the block copolymer and P4VP content as we described before.[24] The results concerning the performances of such membranes can be summarized as follows:



Water fluxes increases with rising pore sizes and is competitive to a commercial polyether sulfone membrane.



Water flux increases with growing P4VP content probably caused by the lower concentration of the casting solution used for the membrane formation and by the hydrophilicity of P4VP.



The membrane thickness has influences to water fluxes which decrease with increasing thickness.



Working conditions for these BCP membranes are at least between 3 to 50 °C and 0.2 to 3 bar transmembrane pressure.



Retention of PEG molecules is independent of the pore size of the membranes used in this work meaning that the separation is mostly carried out in the substructure.



Sharper MWCO can be found for PS-b-P4VP membranes compared to a PES membrane.



Flux recovery after BSA fouling is not influenced by the pore size of the BCP membrane and is slightly higher than for a PES membrane.

For future work the development of isoporous block copolymer membranes should focus on their spongy substructure. It is expected that thinner membranes or those with less dense substructures could improve the separation behavior. For this purpose alternative techniques to solution casting could be implemented for the membrane fabrication in order to receive ultrathin films on support membranes. The reduction of the polymer concentration due to the usage of additives or blends could allow more open substructures or thinner membranes. Concerning the separation characteristics retention measurements and fouling tests should be carried out in cross-flow mode heading for conditions used in application. In the future the

characterization of the porosity could be determined by BET or HG-intrusion porosity measurement beside image analysis. Acknowledgement The authors thank Brigitte Lademann for the synthesis of the polymers, Thomas Bucher for the support with the temperature-dependent measurements, Damla Keskin for her help with the membrane fabrication, Maren Brinkmann for the GPC measurements, Anne Schroeder, Sofia Dami and Clarissa Abetz for the discussion and measurements of the SEMs, Jan Pohlmann and Carsten Scholles for the automatic water flux device. References [1] A. Saxena, B.P. Tripathi, M. Kumar, V.K. Shahi, Membrane-based techniques for the separation and purification of proteins: an overview, Advances in colloid and interface science, 145 (2009) 1-22. [2] R. van Reis, A. Zydney, Membrane separations in biotechnology, Current Opinion in Biotechnology, 12 (2001) 208-211. [3] G.M. Geise, H.-S. Lee, D.J. Miller, B.D. Freeman, J.E. McGrath, D.R. Paul, Water purification by membranes: The role of polymer science, Journal of Polymer Science Part B: Polymer Physics, 48 (2010) 1685-1718. [4] N. Savage, M. Diallo, Nanomaterials and Water Purification: Opportunities and Challenges, J Nanopart Res, 7 (2005) 331-342. [5] B. Van Der Bruggen, C. Vandecasteele, T. Van Gestel, W. Doyen, R. Leysen, A review of pressuredriven membrane processes in wastewater treatment and drinking water production, Environmental Progress, 22 (2003) 46-56. [6] G. Jeon, S.Y. Yang, J.K. Kim, Functional nanoporous membranes for drug delivery, Journal of Materials Chemistry, 22 (2012) 14814. [7] P. Stroeve, N. Ileri, Biotechnical and other applications of nanoporous membranes, Trends in Biotechnology, 29 (2011) 259-266. [8] W.M. Clark, A. Bansal, M. Sontakke, Y.H. Ma, Protein adsorption and fouling in ceramic ultrafiltration membranes, Journal of Membrane Science, 55 (1991) 21-38. [9] W.S. Dai, T.A. Barbari, Gel-impregnated pore membranes with mesh-size asymmetry for biohybrid artificial organs, Biomaterials, 21 (2000) 1363-1371. [10] C. Mattisson, P. Roger, B. Jönsson, A. Axelsson, G. Zacchi, Diffusion of lysozyme in gels and liquids: A general approach for the determination of diffusion coefficients using holographic laser interferometry, Journal of Chromatography B: Biomedical Sciences and Applications, 743 (2000) 151167. [11] B.C. Robertson, A.L. Zydney, Protein adsorption in asymmetric ultrafiltration membranes with highly constricted pores, Journal of colloid and interface science, 134 (1990) 563-575. [12] N. Ileri, R. Faller, A. Palazoglu, S.E. Letant, J.W. Tringe, P. Stroeve, Molecular transport of proteins through nanoporous membranes fabricated by interferometric lithography, Physical Chemistry Chemical Physics, 15 (2013) 965-971. [13] K.V. Peinemann, V. Abetz, P.F. Simon, Asymmetric superstructure formed in a block copolymer via phase separation, Nat Mater, 6 (2007) 992-996.

[14] Z. Wang, L. Guo, Y. Wang, Isoporous membranes with gradient porosity by selective swelling of UV-crosslinked block copolymers, Journal of Membrane Science, 476 (2015) 449-456. [15] W. Sun, Z. Wang, X. Yao, L. Guo, X. Chen, Y. Wang, Surface-active isoporous membranes nondestructively derived from perpendicularly aligned block copolymers for size-selective separation, Journal of Membrane Science, 466 (2014) 229-237. [16] A.L. Zydney, P. Aimar, M. Meireles, J.M. Pimbley, G. Belfort, Use of the log-normal probability density function to analyze membrane pore size distributions: functional forms and discrepancies, Journal of Membrane Science, 91 (1994) 293-298. [17] L.M. Robeson, Correlation of separation factor versus permeability for polymeric membranes, Journal of Membrane Science, 62 (1991) 165-185. [18] A. Mehta, A.L. Zydney, Permeability and selectivity analysis for ultrafiltration membranes, Journal of Membrane Science, 249 (2005) 245-249. [19] Y. Zhang, J.L. Sargent, B.W. Boudouris, W.A. Phillip, Nanoporous membranes generated from self-assembled block polymer precursors:Quo Vadis?, Journal of Applied Polymer Science, (2014) n/an/a. [20] R.M. Dorin, W.A. Phillip, H. Sai, J. Werner, M. Elimelech, U. Wiesner, Designing block copolymer architectures for targeted membrane performance, Polymer, (2013). [21] W.A. Phillip, M. Amendt, B. O'Neill, L. Chen, M.A. Hillmyer, E.L. Cussler, Diffusion and flow across nanoporous polydicyclopentadiene-based membranes, ACS applied materials & interfaces, 1 (2009) 472-480. [22] V. Abetz, Isoporous block copolymer membranes, Macromolecular rapid communications, 36 (2015) 10-22. [23] J.I. Clodt, S. Rangou, A. Schroder, K. Buhr, J. Hahn, A. Jung, V. Filiz, V. Abetz, Carbohydrates as additives for the formation of isoporous PS-b-P4VP diblock copolymer membranes, Macromolecular rapid communications, 34 (2013) 190-194. [24] S. Rangou, K. Buhr, V. Filiz, J.I. Clodt, B. Lademann, J. Hahn, A. Jung, V. Abetz, Self-organized isoporous membranes with tailored pore sizes, Journal of Membrane Science, 451 (2014) 266-275. [25] P. Madhavan, K.V. Peinemann, S.P. Nunes, Complexation-tailored morphology of asymmetric block copolymer membranes, ACS applied materials & interfaces, 5 (2013) 7152-7159. [26] L. Oss-Ronen, J. Schmidt, V. Abetz, A. Radulescu, Y. Cohen, Y. Talmon, Characterization of Block Copolymer Self-Assembly: From Solution to Nanoporous Membranes, Macromolecules, 45 (2012) 9631-9642. [27] M. Radjabian, J. Koll, K. Buhr, U.A. Handge, V. Abetz, Hollow Fibre Spinning of Block Copolymers: Influence of Spinning Conditions on Morphological Properties, Polymer, (2013). [28] H. Yu, X. Qiu, S.P. Nunes, K.V. Peinemann, Self-assembled isoporous block copolymer membranes with tuned pore sizes, Angew Chem Int Ed Engl, 53 (2014) 10072-10076. [29] R.A. Mulvenna, J.L. Weidman, B.X. Jing, J.A. Pople, Y.X. Zhu, B.W. Boudouris, W.A. Phillip, Tunable nanoporous membranes with chemically-tailored pore walls from triblock polymer templates, Journal of Membrane Science, 470 (2014) 246-256. [30] M. Radjabian, V. Abetz, Tailored pore sizes in integral asymmetric membranes formed by blends of block copolymers, Adv Mater, 27 (2015) 352-355. [31] J. Hahn, J.I. Clodt, V. Filiz, V. Abetz, Protein separation performance of self-assembled block copolymer membranes, RSC Advances, 4 (2014) 10252.

Highlights



Isoporous block copolymer membranes with tailored pore sizes



Detailed water permeance study



Performance study concerning retention and fouling