Challenges for recycling ionic liquids by using

PAPER

www.rsc.org/greenchem | Green Chemistry

Challenges for recycling ionic liquids by using pressure driven membrane processes Kurt Haerens,*a,b Stephanie Van Deuren,a Edward Matthijsb and Bart Van der Bruggena Received 6th August 2010, Accepted 19th October 2010 DOI: 10.1039/c0gc00406e Although the extremely low vapour pressure of ionic liquids prevents their emission to the atmosphere, they are at least partly miscible with water and will inevitably end up in the aqueous environment. One example where this can be expected to occur is the application of ionic liquids for the electrodeposition of metals. During industrial use, ionic liquids will also get mixed with other product streams and will have to be separated and recycled. Economically and ecologically it is important to recycle and re-use ionic liquids as efficiently as possible. The use of pressure driven membrane processes, nanofiltration, reverse osmosis and pervaporation, as a possibility to recycle ionic liquids from water was investigated. Ethaline200 (a deep eutectic formed between choline chloride and ethylene glycol) was used to perform these tests and the results were compared with those found in the literature. The osmotic pressure was determined experimentally to explain the results. High ion retentions (up to 0.95) were obtained, but the retention of the non-charged molecules was too low. For concentrating ionic liquids the osmotic pressure was found to be the limiting factor when using nanofiltration or reverse osmosis. Only a five-fold concentration of the ionic liquid was possible, to a maximum concentration of 20–25 vol% of ionic liquid. Pervaporation was investigated as an alternative. It was found to have limited usability for this application as the water content is too high. For low water contents, pervaporation is applicable although the flux is very low due to the presence of the ionic liquid, which decreases the activity of the water and thus the flux through the membrane. The necessary membrane area would be very high and makes pervaporation rather impractical.

Introduction Ionic liquids are salts, composed of a combination of mostly organic cations and anions or short-lived ion pairs, with a melting point below 100 ◦ C. Ionic liquids are studied for a broad range of different applications including organic synthesis and biological reactions, separation processes such as extraction, and electrochemical applications.1–3 In recent years, both toxicity and ecotoxicity of different ionic liquids have been studied4–7 and attempts have been made to predict them.8–9 Their low volatility was often used to call them green but their toxicity, ecotoxicity, environmental fate and their exposure also have to be taken into account in the qualification of a green solvent. Also these properties depend very much on the choice of anion and cation, as do practically all properties of ionic liquids.10–11 Many ionic liquids are hydrophilic or at least partially miscible with water and will end up in the aquatic environment.7,12 The recyclability of ionic liquids, although a key property for their increasing popularity, has only been scarcely studied.13 Recycling of ionic liquids prevents them from ending up in the aquatic environment, as their low volatility prevents them from release into the atmosphere; at the same time, it is economically a Department of chemical engineering, K.U. Leuven, Willem de Croylaan 46, 3001, Heverlee, Belgium b Department of industrial engineering, KaHo Sint-Lieven, Gebroeders Desmetstraat 1, 9000, Gent, Belgium. E-mail: [email protected]; Fax: +32 9 265 87 24; Tel: +32 9 265 37

2182 | Green Chem., 2010, 12, 2182–2188

beneficial because ionic liquids still remain very expensive in comparison to their alternatives. As far as we know, the recyclability of ionic liquids is predominantly studied when two different phases are formed.14–19 Even in these examples, the ionic liquid can only be recycled a limited number of times due to impurities that cannot be removed.15,19–21 In electrodeposition processes ionic liquids will end up in the rinse water.22 However, during other applications ionic liquids will also have to be separated from water or solvents, for example in the production of zeolite,23 Suzuki coupling24 and cellulose processing.17,25 Depending on the application the aim is to permeate or to retain the ionic liquid. In the case of electrodeposition, the aim is to retain the ionic liquid as the ionic liquid will be the minor compound in the mixture. Recyclability or purification26 of ionic liquids, especially those soluble in water, has proved to be more difficult than expected. For the separation of water from an ionic liquid or vice versa, evaporation is the first one thinks of. However, this is very energy consuming and requires higher temperatures, at which some ionic liquids may decompose. Although generally ionic liquids are denoted to have high decomposition temperatures based on TGA scans, recent studies have indicated that in more real circumstances the decomposition temperature can be significantly lower.27–29 In the case of protic ionic liquids, loss of the conjugated acid or base species may occur through evaporation, leaving hydroxide or proton impurities in solution.26 Compared with other separation techniques, membrane separations generally have a lower energy consumption and allow This journal is © The Royal Society of Chemistry 2010

a solvent-less operation. Nanofiltration could be an interesting technique for the separation of ionic liquids from solvents as they only consist of ions. This is especially the case for the recycling of ionic liquids from non-volatile compounds or for systems not suitable for distillation or if the ionic liquid is the minor component in the mixture.30 Charged and neutral compounds or mono- and divalent ions can be separated by nanofiltration membranes. Charge and size of the ions or molecules and the choice of membrane influence the retention.31 For the separation of ionic liquids and non-volatile products, nanofiltration was tested32 and promising results could be obtained. The ionic liquid and the non-volatile products were dissolved in water first and the ionic liquid could be permeated through the membrane where the product was rejected. Pervaporation is often used as a less energy demanding alternative to vacuum or extractive distillations.33 The separation of compounds during pervaporation is based mainly on the interaction of the feed components with the membrane material combined with the gradient in a chemical potential.34 Pervaporation has been shown to be an efficient technique to remove different organic solvents and water from BMIM PF6 ,35 the ionic liquid could be purified to 99.2%. Pervaporation has also been used to improve the conversion of organic synthesis in ionic liquids by removing the water formed as a side product.36–37 The deep eutectic solvent Ethaline200, used in this study, is very comparable to the first generation of ionic liquids as they consist of a choline cation and a complex anion composed of chloride and two ethylene glycol molecules.38 Ethaline200 is easily prepared by mixing 1 mole of choline chloride with 2 moles of ethylene glycol.39 Deep eutectic solvents can be used under ambient conditions, as they are not air and moisture sensitive, and are furthermore easy to prepare, less toxic and sometimes even biodegradable.40 Therefore they are very suitable as a test case for other ionic liquids and deep eutectic solvents. In this study, the aim was to recycle the deep eutectic solvent from a synthetic waste water stream. For this purpose, nanofiltration, reverse osmosis and pervaporation were selected. For nanofiltration and reverse osmosis, on the basis of a performance test, the retention was evaluated as well as the attainable flux. Furthermore, the osmotic pressure was examined as it determines the feasibility of this technique. Finally, the use of pervaporation was evaluated, both for the removal of water from low and for high concentrated solutions of Ethaline200.

Experimental section General Ethaline200 was prepared as described earlier,41 the water content was determined with a KarlFisher Metrohm volumetric 870 KF titrino plus to be 11.13 weight%. MilliQ water was made with a Synergy ultra pure water system (Millipore). All solutions are expressed in volume% Ethaline in water unless otherwise mentioned and the water content of the used Ethaline200 is calculated in the results. Membranes Three nanofiltration membranes were selected, FilmTec NF90 (Dow), FilmTec NF207 (Dow) and DK (GE Osmonics). The This journal is © The Royal Society of Chemistry 2010

DK polyamide membrane is very dense and has a MWCO of 150–300. Also two reverse osmosis membranes were selected, FilmTec 102326 (Dow) and FilmTec BW30-XLE (Dow). The FilmTec BW30-XLE (XLE = Extra Low Energy) membranes are specially designed to work at lower pressure with the same flux compared to other reverse osmosis membranes. Both are very dense and have a retention of 99% for aqueous solutions. All used membranes were commercially available hydrophilic membranes. All membranes were supplied in a dry form, therefore the membranes were immersed in water for at least 24 h prior to use. To check the stability of the membranes to the ionic liquid used, the membrane sheets were immersed in the ionic liquid. No visual damage was observed for these membranes after two weeks of exposure. For the pervaporation experiments, a PERVAP 1201 (Sulzer Chemtech Gmbh) membrane was selected as it is a hydrophilic PVA-composite membrane and thus highly permeable for water while hindering the permeation of hydrophobic molecules.31 Filtration The experimental setup was described previously.42 The applied transmembrane pressure for the FilmTec nanofiltration membranes was 20 bar whereas for the dense DK nanofiltration membrane and the two reverse osmosis membranes 30 bar was used. The active membrane area of each cell was 14.6 cm2 . A Teflon-coated magnetic stir bar on top of the membrane was used to minimize concentration polarization at the membrane surface. Membrane permeability was determined by measuring the concentrations in the permeate volume collected over 10 min. A 5 vol% Ethaline200 in water solution was used as solvent. All experiments were performed in duplicate. As a first performance indicating test, 100 ml feed was filtered until 30 ml had permeated and the ion concentration was measured. Osmotic pressure The dead-end module was used in combination with the FilmTec BW30-XLE membrane. The flux was determined for different concentrations (ranging from 2 to 30 vol% Ethaline200 in water) and different pressures. For each concentration at least three different pressures were used, the concentration of the retentate was measured to determine the mean feed concentration. The volume of the permeate was collected during 1 h and this value was used to calculate the flux. Through extrapolation of these results, the osmotic pressure DP can be determined using following equation: J w = K w (DP - DP). As when the flux is zero, DP = DP. Pervaporation The experimental setup was described previously.43 The flow rate of the feed was kept constant at 0.7 l min-1 and the circular fleet sheet membrane had an area of 56.74 cm2 . At the vacuum side the pressure was kept constant at 1 mbar. 2 litres of Ethaline200 with 12.11 w% water (determined with KF) was used as the feed. For the other experiments, the feed concentration was changed by replacing 0.5 litre of the Green Chem., 2010, 12, 2182–2188 | 2183

feed by pure water, the water concentration was determined by KarlFisher. The permeate was collected in a glass tube during 1 h and the weight determined to calculate the total flux (J m = mp /Atm with mp being the total mass of collected permeate, A the active membrane area and tm the time (s) of the experiment). The concentrations of choline and ethylene glycol in the permeate were determined. From these results, the selectivity was calculated. All experiments were performed in duplicate. Sample composition analysis Conductivity measurement (Orion model 160 with a 016010 conductivity cell) of the sample was used as a quick determination of the ion concentration. Choline concentration was determined using a Dionex DX-120 ion chromatograph with a CG12A 4 mm guard column, CS12A 4 mm cation exchange column, a CSRS300 4 mm suppressor and a conductivity detector. 9 mM H2 SO4 was used as an eluent with a flow rate of 1 ml min-1 . An external standard method was applied for the analysis of the sample composition. The concentration of ethylene glycol was determined using a Perklin Elmer Autosystem XL gas chromatograph with a plot fused silica column (10 ¥ 0.53 mm) and a flame ionization detector. An external standard method was applied for analysis of the sample composition.

Results and discussion In Fig. 1 the conductivity of a water/Ethaline200 mixture is plotted. The conductivity has a maximum at 35 vol% Ethaline200 and then decreases. The same behaviour had also previously been reported for other ionic liquids.44–46

Fig. 2

Conductivity of Ethaline200/water mixtures from 0 to 5%.

Table 1 Performance test of the selected membranes Membrane

Type

Ion retention (%)

FilmTec NF90 DK FilmTec BW30XLE FilmTec 102326

Nanofiltration Nanofiltration Reverse osmosis Reverse osmosis

20.0 88.0 90.5 91.1

possible recovery and to determine the flux. The retention is also higher than the retention of [BMIM][BF4 ] in water which was only 82% with a Desal DVA 032 membrane and only slightly lower than the 95% retention of [BMIM]2 [SO4 ] with a Desal DVA 00 membrane.32 The last one gave a higher retention due to the presence of divalent ions. The ion retention of Ethaline200 is lower than the retention for NaCl specified by the supplier. The retentions of [C4 MIM]Br and [C4 MIM]BF4 in an aqueous solution were also found to be lower than those of NaCl/water system.30 This is remarkable, since these ions are larger than Na+ and Cl- . The stronger hydration of NaCl compared with the ionic liquids could be a possible explanation. Fig. 3 shows the total flux for the three tested membranes as a function of feed concentration. The flux for the FilmTec 102326 membrane is considerably lower than for the other two membranes. The FilmTec BW30-XLE, has a comparable flux to the nanofiltration membrane DK, although it is a reverse osmosis membrane. This is not unexpected as the FilmTec BW30-XLE membrane is designed to have higher fluxes whereas the DK membrane is a very dense nanofiltration membrane resulting in a moderate flux.47

Fig. 1 Conductivity of Ethaline200/water mixtures from 0 to 100%.

In Fig. 2 it can be observed that the conductivity as a function of the concentration is linear up to a concentration of 5% Ethaline200. Conductivity can thus be used as an easy determination of the concentration. Nanofiltration In Table 1, the results of the performance tests are given. The performance of the NF90 membrane was not satisfactory as the permeate still contained 4 vol% Ethaline200. As the NF270 membrane has even lower salt retention, this membrane was no longer tested. The three other membranes, DK, FilmTec 102326 and FilmTec BW30-XLE showed a retention of 88, 91.1 and 90.5% respectively, which were promising results to investigate 2184 | Green Chem., 2010, 12, 2182–2188

Fig. 3 Flux as a function of normalised feed concentration for the membranes DK (䉬), FilmTec 102326 (䉱) and FilmTec BW30-XLE (¥).

The ion retention of these three membranes is summarised in Fig. 4. The retention for both reverse osmosis membranes This journal is © The Royal Society of Chemistry 2010

Fig. 4 Retention as a function of recovery for the membranes DK (䉬), FilmTec 102326 (䉱) and FilmTec BW30-XLE (¥).

(FilmTec) is slightly higher than for the nanofiltration membrane (DK). The FilmTec 102326 membrane could only be tested up to 60% recovery as the flux was very low at that point. For both other membranes, the retention decreases strongly when the recovery is higher than 70%. This behaviour has been reported and explained by Cornelissen et al.48 for a cross-flow module. The combination of the equations for water flux J w (eqn (1)) with the salt flux J s (eqn (2)), the mass balance (eqn (3) and eqn (4)) and the recovery S (eqn (5)) lead to an equation for the retention R (eqn (6)). It is clear from this equation that if the recovery S increases, the retention R will decrease. J w = K w (DP - DP)

(1)

J s = K s Dc + K c cc J w

(2)

Vf = Vc + Vp

(3)

V f cf = V c cc + V p cp

(4)

S=

R=

Vp Vf

×100

1− K c ⎡ 2−S ⎤ ⎡ S ⎤ Ks ⎥ + K c ×⎢ ⎥ 1+ ×⎢ ⎢ ⎥ ⎢ K w ( DP − DΠ ) ⎣ 2(1 − S ) ⎦ ⎣ 2(1 − S ) ⎥⎦

(5)

(6)

Although the ion retention is high, the retention of ethylene glycol is not sufficient (mean value of 44.3%) dropping to less than 10% for a recovery of 60% and more. The measured osmotic pressure of an Ethaline200 solution is presented in Fig. 5 and compared with calculated values based on the osmotic coefficients found in the literature for choline chloride, choline bromide,49 BMIM bromide50 and BMIM·BF4 .30 The curve of the limiting law is the integration of the activity coefficient for point charges (eqn (7)) substituted in eqn (8), the correlation between the activity coefficient and the osmotic coefficient. This integration results in eqn (9) in which m is the concentration of ions (mol kg-1 ) and Am is a constant. This journal is © The Royal Society of Chemistry 2010

Fig. 5 Osmotic pressure as a function of ion concentration for Ethaline200 (䊉) compared with salts (choline chloride (䉬) and choline bromide ()49 ) and ionic liquids (BMIM bromide (¥)50 and BMIM BF4 (䉱)30 ) at room temperature and the limiting law ().

logg = − z + z − A c I c

f −1 +

2.302585 × m

∫

(7)

m

m d log g c

(8)

0

f = 1 − 2.302585 ×

Am m 3

(9)

The smaller the ions, the better the curve of the osmotic pressure follows the curve of the limiting law. Until 0.6 mol kg-1 all values correspond well with each other. For higher concentrations, the values and the curve differ strongly from the expected values. This can be explained by the accuracy of the experiments: as the flux becomes extremely low for high osmotic pressures, the permeated volume becomes very low and the difference between the different fluxes at different transmembrane pressures negligible. At low concentrations, the osmotic pressure of Ethaline200 is higher compared to the other ionic liquids as both the anion and cation in Ethaline200 are smaller and thus more strongly solvated compared to the BMIM, Br and BF4 ions. Generally, the osmotic pressure of ionic liquids is lower than for conventional salts. However, for high concentrations the osmotic pressure will still be limiting. Due to this high osmotic pressure, it is impossible to purify the ionic liquid to the point where almost no water remains. As the maximal operation pressure for the membranes is 41 bar,51 the maximal concentration that can be reached by reverse osmosis is 1.75 mol kg-1 (30 vol%) for point charges and for the salts and BMIMBr in Fig. 5 the maximum is 1.2 mol kg-1 (22 vol%). In our example, where the starting concentration is 5 vol%, nanofiltration or reverse osmosis can only be used to concentrate the ionic liquid in the water with a factor of 4 to 5. A complete removal of water will thus be impossible with pressure driven membrane processes but it can act as a preconcentration step for other processes which require a certain minimal concentration of ionic liquids. This happens with hydrophilic ionic liquids where a certain minimal concentration is required before separation with supercritical carbon dioxide can take place.52 In our case, not only the ion retention was important but also the retention of ethylene glycol. As can be seen in Table 2, the Green Chem., 2010, 12, 2182–2188 | 2185

Table 2 Retention of ethylene glycol as a function of the recovery for the FilmTec BW30XLE membrane Recovery (%)

Retention

10 20 30 40 50 60 70 80

0.12 0.51 0.31 0.53 0.41 0.46 0.43 0.05

Table 4 Vapour pressure of pure water and ethylene glycol and of mixtures of water/ethylene glycol

Water Ethylene glycol

Concentration (vol%)

pi sat (mbar)

pi (mbar)

20 80

123.28 1.12

50.45 0.61

0

Table 3 Total flux (kg m-2 h-1 ) for 3 conventional solvent/water mixtures and Ethaline200/water mixture as a function of feed water concentration Feed water concentration (w%)

Ethanol

Isopropanol

Methanol

Ethaline200

20 40 60

0.20[x] 0.38[x] 0.60[x]

0.20[x] 0.38[x] 0.58[x]

0.58[x] 0.85[x] 1.15[x]

0.08 0.19 0.49

Fig. 7 Water flux as a function of feed water concentration.

[x] Refers to values from Fig. 1 with a PERVAP 2201 membrane.53

retention of ethylene glycol was very low. This is not unexpected as ethylene glycol is a very small and polar molecule, so it can dissolve easily in the membrane and also transport through the pores of the membrane. For this application, the permeate cannot be re-used, since the ethylene glycol concentration is too high. For ionic liquids where no uncharged molecules are present, the permeate will have a low ion concentration and can thus be re-used as rinse water. Pervaporation In Fig. 6 the results for the total flux of the pervaporation experiments are shown. For low water concentrations, up to 50 w% water, the total flux increases linearly as can be expected.33–34 But for concentrations above 50 w% of water, there is a more than linear increase in flux. For low water concentrations, and thus high concentration of salts, the activity coefficient of the water molecules is lower than for more diluted samples, which leads to lower fluxes. This is because the driving force of pervaporation is the chemical potential which is dependent on the activity coefficient. Below 50 w% water, the total fluxes are considerably lower than the total

Fig. 6 Total flux as a function of feed water concentration for 2 series of pervaporation experiments.

2186 | Green Chem., 2010, 12, 2182–2188

Fig. 8 Ethylene glycol flux as a function of feed water concentration.

fluxes of conventional solvent/water mixtures measured using a PERVAP 2201 membrane,53 which has comparable properties to the PERVAP 1201 membrane (See Table 3). The permeate did not contain any ions, as they have a negligible vapour pressure. Ethylene glycol, however, also permeates through the membrane. The flux for water and ethylene glycol through the membrane for different feed concentrations are presented in Fig. 7 and 8, respectively. Both curves correspond well with the curve of the total flux in Fig. 6. Vapour pressures for the pure compounds water and ethylene glycol and for a 20 and 80 w% can be found in Table 4. These values indicate the preference of water to permeate through the membrane and no ethylene glycol is expected to permeate based on its vapour pressure. Furthermore a hydrophilic membrane is used, which excludes solvation of ethylene glycol. The ethylene glycol is probably transported passively through the membrane together with the water. Another indication for this mechanism is the fact that the ethylene glycol concentration in the permeate increases even though the concentration of ethylene glycol in the feed decreases. So the flux of ethylene glycol is not influenced by its own gradient in chemical potential but by the permeation of the water through the membrane. The same behaviour was reported earlier for ethanol/water mixtures.53 In Fig. 9 the selectivity of the pervaporation membrane used is given. The selectivity decreases with increasing water concentration, this is not unexpected as a higher flux often This journal is © The Royal Society of Chemistry 2010

5 6 7 8 9 10

Fig. 9

Selectivity as a function of feed concentration.

11 12 13

corresponds to a loss in selectivity during pervaporation,54–56 meaning that selectivity is rather low. As previously reported54 pervaporation adds an additional resistance by the introduction of a membrane compared to evaporative methods, the only advantage is the possibility to work under more moderate conditions (temperature and pressure).

17

Conclusion

18 19

Recycling of hydrophilic, water miscible ionic liquids by means of membrane separation technologies was studied. The applicability of nanofiltration and reverse osmosis was found to be limited by the osmotic pressure of the solution of ionic liquids in water. Even for ions which have a comparable behaviour comparable to point charges, the maximal obtainable concentration after separation is 30 vol% of the ionic liquid. Pervaporation requires a large membrane area, with the existing pervaporation membranes, and only the removal of small amounts of water is possible. More specific membranes with a higher water flux could be an improvement for this application. Other separation technologies, such as extraction in supercritical carbon dioxide, and deep vacuum distillation, have to be evaluated for the recycling of ionic liquids from water or new technologies have to be explored. However both nanofiltration and pervaporation can probably improve the recycling and/or purification of hydrophobic or partially water immiscible ionic liquids but this still has to be studied.

Acknowledgements The authors would like to thank the EU under framework programme 6 for funding this work through the IONMET project (www.ionmet.eu) and also the IWT-Flanders (Belgium) for their support of the SBO project MAPIL.

14 15 16

20 21 22 23 24 25 26 27 28 29 30 31

32 33 34 35

Notes and references 1 Ionic liquids in synthesis, 2nd edn, ed. P. Wasserscheid, T. Welton, Wiley-VCH, Weinheim, 2007. 2 S. Werner, M. Haumann and P. Wasserscheid, Annu. Rev. Chem. Biomol. Eng., 2010, 1, 203–230. 3 Electrochemical aspects of ionic liquids, ed. H. Ohno, WileyInterscience, New York, 2005. ¨ 4 B. Jastorff, K. Molter, P. Behrend, U. Bottin-Weber, J. Filser, A. ¨ Heimers, B. Ondruschka, J. Ranke, M. Schaefer, H. Schroder, A. ¨ Stark, P. Stepnowski, F. Stock, R. Stormann, S. Stolte, U. Welz-

This journal is © The Royal Society of Chemistry 2010

36 37 38

39

¨ Biermann, S. Ziegert and J. Thoming, Green Chem., 2005, 7, 362– 372. P. Stepnowski, A. C. Skladanowski, A. Ludwiczak and E. Laczynska, Hum. Exp. Toxicol., 2004, 23, 513–517. K. M. Docherty and C. F. Jr. Kulpa, Green Chem., 2005, 7, 185–189. D. Zhao, Y. Liao and Z. Zhang, Clean, 2007, 35, 42–48. (a) P. Luis, I. Ortiz, R. Aldaco and A. Irabien, Ecotoxicol. Environ. Saf., 2007, 3, 423–438; (b) P. Luis, A. Garea and A. Irabien, J. Mol. Liq., 2010, 125, 28–33. E. J. Maginn, D. J. Couling, R. J. Bernot, K. M. Docherty and J. K. Dixon, Green Chem., 2006, 8, 82–90. N. V. Plechkova and K. R. Seddon, Chem. Soc. Rev., 2008, 37, 123– 150. D. R. MacFarlane and K. R. Seddon, Aust. J. Chem., 2007, 60, 3–5. ¨ J. Ranke, S. Stolte, R. Stormann, J. Aming and B. Jastorff, Chem. Rev., 2007, 107, 2183–2206. B. Wu, W. Liu, Y. Zhang and H. Wang, Chem.–Eur. J., 2009, 15, 1804–1810. M. J. Earle, P. B. McCormac and K. R. Seddon, Green Chem., 1999, 1, 23–25. A. Kamal and G. Chouhan, Adv. Synth. Catal., 2004, 346, 579–582. J. K. Park, P. Sreekanth and B. M. Kim, Adv. Synth. Catal., 2003, 346, 49–52. S. Zhu, Y. Wu, Q. Chen, Z. Yu, C. Wang, S. Jin, Y. Ding and G. Wu, Green Chem., 2006, 8, 325–327. F. Zulfiqar and T. Kitazume, Green Chem., 2000, 2, 296–297. V. Gauchot, W. Kroutil and A. R. Schmitzer, Chem.–Eur. J., 2010, 16, 6748–6751. A. P. Abbott, G. Capper, D. L. Davies, R. K. Rasheed and V. Tambyraja, Green Chem., 2002, 4, 24–26. J. L. Scott, D. R. MacFarlane, C. L. Raston and C. M. Teoh, Green Chem., 2000, 2, 123–126. K. Haerens, E. Matthijs, A. Chmielarz and B. von der Bruggen, J. Environ. Manage., 2009, 90, 3245–3252. E. R. Cooper, C. D. Andrews, P. S. Wheatley, P. B. Webb, P. Wormald and R. E. Morris, Nature, 2004, 430, 1012–1016. P. S. Shinde, S. S. Shinde, A. S. Renge, G. H. Patil, A. B. Rode and R. R. Pawar, Lett. Org. Chem., 2009, 6, 8–10. ¨ G. Bentivoglio, T. Roder, M. Fasching, M. Buchberger, H. Schottenberger and H. Sixta, Lenzinger Ber., 2006, 86, 154–161. B. Clare, A. Sirwardana and D. R. MacFarlane, Top. Curr. Chem., 2010, 209, 1–40. M. Kosmulski, J. Gustafsson and J. B. Rosenholm, Thermochim. Acta, 2004, 412, 47–53. K. J. Baranyai, G. B. Deacon, D. R. MacFarlane, J. M. Pringle and J. L. Scott, Aust. J. Chem., 2004, 57, 145–147. T. J. Wooster, K. M. Johanson, K. J. Fraser, D. R. MacFarlane and J. L. Scott, Green Chem., 2006, 8, 691–696. B. Wu, Y. Wu, Q. Chen, Z. Yu, C. Wang, S. Jin, Y. Ding and G. Wu, Molecules, 2009, 14, 1781–1788. (a) M. H. V. Mulder, E. M. van Voorthuizen and J. M. M. Peeters, in Nanofiltration: Principles and applications, ed. A. I. Schäfer, A. G. Fane and T. D. Waite, Elsevier Ltd., Oxford, 2003, pp. 89–117; (b) T. D. Waite, in Nanofiltration: Principles and applications, ed. A. I. Schäfer, A. G. Fane and T. D. Waite, Elsevier Ltd., Oxford, 2003, pp. 147–168. ¨ J. Krockel and U. Kragl, Chem. Eng. Technol., 2003, 26, 1166– 1168. J. Néel, in Membrane separations technology. Principles and applications, ed. R. D. Noble and S. A. Stern, Elsevier science B.V., Amsterdam, 1995, pp. 143–211. M. Mulder, Basic principles of membrane technology, ed. M. Mulder, Kluwer academic publishers, Dordrecht, 1996. T. Schäfer, C. M. Rodrigues, C. A. M. Afonso and J. G. Crespo, Chem. Commun., 2001, 1622–1623. ´ ´ Green L. Gubicza, N. Nemestothy, T. Fráter and K. Bélafi-Bako, Chem., 2003, 5, 236–239. P. Izak, N. M. M. Mateus, C. A. M. Afonso and J. G. Crespo, Sep. Purif. Technol., 2005, 41, 141–145. A. Abbott and C. O’Connor, Oral presentation at the IONMET/Prosurf seminar Munich, Germany, November 14, 2007, https://www.ionmet.eu/fileadmin/ionmet/training/Productronica_2007/6_ElectrolessAg_Abbott.pdf. A. P. Abbott, G. Capper, K. J. McKenzie and K. S. Ryder, Electrochim. Acta, 2006, 51, 4420–4425.

Green Chem., 2010, 12, 2182–2188 | 2187

¨ 40 A. P. Abbott, K. S. Ryder and U. Konig, Trans. Inst. Met. Finish., 2008, 86, 196–204. 41 K. Haerens, E. Matthijs, K. Binnemans and B. Van der Bruggen, Green Chem., 2009, 11, 1357–1365. 42 S. Darvishmanesh, J. Degrève and B. Van der Bruggen, Chem. Eng. Sci., 2009, 64, 3914–3927. 43 A. Verhoef, A. Figoli, B. Leen, B. Bettens, E. Drioli and B. Van der Bruggen, Sep. Purif. Technol., 2008, 60, 54–63. 44 J. Vila, P. Ginés, E. Rilo, O. Cabeza and L. M. Varela, Fluid Phase Equilib., 2006, 247, 32–39. 45 H. Shekaari and S. S. Mousavi, Fluid Phase Equilib., 2009, 286, 120– 126. 46 A. Jarosik, R. Krajewski, A. Lewandowski and P. Radzimski, J. Mol. Liq., 2006, 123, 43–50. 47 Dow Water & Process solutions: FILMTECTM Reverse osmosis membranes. Technical manual. http://www.dow.com/liquidseps/ lit/techman.htm, 2010. 48 E. R. Cornelissen, J. Verdouw, A. J. Gijsbertsen-Abrahamse and J. A. M. H. Hoffman, Desalination, 2005, 178, 179–192.

2188 | Green Chem., 2010, 12, 2182–2188

49 G. E. Boyd, A. Schwarz and S. Lindenbaum, J. Phys. Chem., 1965, 70, 821–825. 50 H. Shekaari and M. T. Zafarani-Moattar, Fluid Phase Equilib., 2007, 254, 198–203. 51 Dow Sales info FILMTECTM membranes. http://www.dow.com/ published literature/dh_0068/0901b803800686ef.pdf, 2010. 52 (a) S. N. V. K. AkiA. M. Scurto, J. L. Anthony and J. F. Brennecke, Proceedings of the 6th International symposium on supercritical fluids, 2003, 28–30 April, Versailles, France; (b) A. M. Scurto, S. N. V. K. Aki and J. F. Brennecke, Chem. Commun., 2003, 572–573. 53 D. Van Baelen, B. Van der Bruggen, K. Van den Dungen, J. Degrève and C. Vandecasteele, Chem. Eng. Sci., 2005, 60, 1583–1590. 54 T. Schäfer, L. C. Branco, R. Fortunato, P. Izák, C. M. Rodrigues;, C. A. M. Afonso and J. G. Crespo, ACS symp. ser. 902: Ionic liquids: Fundamentals, progress and opportunities, 2005, 97–110. 55 A. Mixa and C. Staudt, Int. J. Chem. Eng., 2008, DOI: 10.1155/2008/319392. 56 T. A. Peters, C. H. S. Poeth, N. E. Benes, H. C. W. M. Buijs, F. F. Vercauteren and J. T. F. Keurentjes, J. Membr. Sci., 2006, 42–50.

This journal is © The Royal Society of Chemistry 2010