Accepted Manuscript

Title: A review of water recovery by vapour permeation through membranes Authors: Brian Bolto, Manh Hoang, Zongli Xie

PII: DOI: Reference: To appear in:

S0043-1354(11)00650-6 10.1016/j.watres.2011.10.052 WR 8896 Water Research

Received Date: 26 May 2011 Revised Date: 21 October 2011 Accepted Date: 24 October 2011

Please cite this article as: Bolto, B., Hoang, M., Xie, Z. A review of water recovery by vapour permeation through membranes, Water Research (2011), doi: 10.1016/j.watres.2011.10.052 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 proof before it is published in its final 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.

*Manuscript Click here to download Manuscript: VP WR text no tracking.docx

Click here to view linked References

ACCEPTED MANUSCRIPT
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

A review of water recovery by vapour permeation through membranes

Brian Bolto*, Manh Hoang, Zongli Xie
CSIRO Materials Science and Engineering, Private Bag 33, Clayton South MDC, Victoria 3169, Australia

ABSTRACT

Keywords: Vapour permeation; pervaporation; gas drying; solvent dehydration; steam recovery Contents

1. Introduction 2. Water permeabilities of membrane systems 2.1 Dehydration of natural gas 2.2 Drying of compressed air 2.3 Flue gas dehydration 2.4 Dehydration of ethanol 2.5 Dehydration of isopropanol 2.6 Dehydration of acetonitrile 2.7 Steam recovery 2.8 Miscellaneous water removal applications 3. Membranes used in vapour permeation processes 4. Comparison of vapour permeation and pervaporation References

EP

TE D

In vapour permeation the feed is a vapour, not a liquid as in pervaporation. The process employs a polymeric membrane as a semi-permeable barrier between the feed side under high pressure and the permeate side under low pressure. Separation is achieved by the different degrees to which components are dissolved in and diffuse through the membrane; the system working according to a solution-diffusion mechanism. The materials used in the membrane depend upon the types of compounds being separated, so water transport is favoured by hydrophilic material, whether organic or inorganic. The process is used for the dehydration of natural gas and various organic solvents, notably alcohol as biofuel, as well as the removal of water from air and its recovery from waste steam. Waste steam can be found in almost every plant/factory where steam is used. It is frequently contaminated and cannot be reused. Discharging the spent steam to the atmosphere is a serious energy loss and environmental issue. Recycling the steam can significantly improve the overall energy efficiency of an industry, which is responsible for massive CO2 emissions. Steam separation at high fluxes and temperatures has been accomplished with a composite poly(vinyl alcohol) membrane containing silica nanoparticles, and also, less efficiently, with an inorganic zeolite membrane.

M AN U
1

SC

RI PT

2 2 3 4 4 4 5 5 5 6 6 7 8

AC C

ACCEPTED MANUSCRIPT
49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 *Corresponding author. Tel.: +61 3 9545 2037, fax: +61 3 9545 1128. E-mail address: brian.bolto@csiro.au 1. Introduction Vapour permeation employs a polymeric membrane as a semi-permeable barrier between a feed side under high pressure and a permeate side under low pressure (Brinkmann et al., 2003). Unlike pervaporation, the feed is a vapour and not a liquid, so there is no phase change or significant temperature difference across the membrane. Separation is achieved by the different degrees to which components are dissolved in and diffuse through the polymer. The polymers used in the membrane will hence depend on the types of compounds being separated (Leemann et al., 1996). The driving force can be approximated to the difference in partial pressures of the components in the feed. Compared to pervaporation, vapor permeation effectively increases the permselectivity of water (Fan et al., 2002) and is capable of higher fluxes (Sander and Janssen, 1991). Another advantage is that the membrane area required is less, and there is a lower likelihood of membrane damage by impurities in the feed. Although vapour permeation operates with the same type of membranes as pervaporation, it has taken longer for its technical realisation. The first industrial vapour permeation plant was built in Germany in 1989 for dehydrating 30 kL/d of ethanol (Sander and Janssen, 1991). More than 100 membrane vapour-gas separation systems have been installed worldwide for recovering high value solvents, liquefied petroleum gas, refrigerant gases, monomers such as ethylene, propylene and vinyl chloride, and for removing the acid gases H2S and CO2 from natural gas and hydrocarbon vapours from air streams in the petrochemical industry (Baker et al., 1998; Jonquires et al., 2002). The transport of water vapour and inert gases through polymeric membranes was reviewed in the literature some time ago (Metz et al., 2005). The authors noted that the use of membranes for this purpose is of major industrial importance, with applications in areas such as the drying of natural gas and compressed air, protective apparel, packaging materials, roofing covers and humidity control in confined spaces, involving air conditioning in buildings, aviation and space flight. Steam recovery is also in this category. Vapour permeation, like pervaporation, works according to a solution-diffusion mechanism. The molecular interaction between the membrane and the separated species is the determining factor rather than the molecular size. The main component of the process is the membrane material which determines the permeation and selectivity and hence the separation properties of the process. The driving force for the mass transfer of permeate from the feed side to the permeate side of the membrane is the chemical potential gradient established by applying a difference in the partial pressures of permeate across the membrane (Aptel et al., 1972; Neel, 1991). The membrane performance is expressed as the membrane flux J in P.P J l where P is the permeability coefficient for a vapour, l is thickness of the membrane, and P is the difference in pressure between the feed and permeate sides (Scott, 1998). 2. Water Permeability of Membrane Systems The water vapour and nitrogen permeability behaviour of 19 polymers reported in the literature has been summarised (Metz et al., 2005). The data extrapolated to zero water activity are given in Table 1. Many of the values were obtained from pure gas permeabilities 2

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT
98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 by calculating the ratios for the mixtures. The permeabilties are given in Barrer, a non-SI unit of gas permeability (specifically, oxygen permeability) used in the contact lens industry (Alter, 1962): one Barrer = 10-11 (cm3 O2) cm cm-2 s-1 mmHg-1. In real mixtures however, water may swell the membrane so that its effect on the slower species is not known with certainty. Generally, for binary mixtures of permanent gases a higher selectivity is accompanied by a lower permeability, but such a relationship does not hold for water vapour in a mixture with a permanent gas, where most of the highly selective polymers also have a very high permeability. There can be a variation of over seven orders of magnitude in selectivity and five orders of magnitude in permeability. The effect of polymer structure on performance is variable. Completely non-polar membranes such as those based on natural rubber, polystyrene, polypropylene and polyethylene tend to have poorer permeabilities. The low permeability of poly(vinyl alcohol) or PVA is surprising, as this is the basis of many commercial membranes in the area (Jonquires et al., 2002). The presence of some polarity such as ether or sulphonate groups is desirable. The introduction of sulphonate or carboxylate groups into the PVA structure would seem to be worthwhile. What is absent in the list of polymers tested here for vapour permeation are polysalts formed from cationic and anionic polyelectrolytes, which have good performance in pervaporation dehydration applications. For the removal of water from water/organic liquid or vapour mixtures, hydrophilic organic polymers are generally used because water is readily incorporated and diffused through these materials. The hydrophilicity is caused by groups present in the polymer chain that are able to interact with water molecules. Examples of hydrophilic polymers are: cellulose acetate, ionic polymers, poly(vinyl alcohol) and polyacrylonitrile. Hydrophilic features are found to be essential for good pervaporation performance in the dehydration of ethanol (Bolto et al., 2011). 2.1 Dehydration of natural gas

2.2 Drying of compressed air For the drying of compressed air the preferred membrane material is polydimethylsiloxane or cellulose acetate, because of their acceptable H2O/N2 selectivities and high permeabilities (Sijbesma et al., 2008). Increasing the level of sulphonate groups in poly(ether ether ketone) membranes has been shown to not only increase the permeability of water vapour, but to decrease gas permeability (Liu et al., 2001). The permeability towards water is doubled in poly(aryl ether sulphone) membranes by introducing carboxylate groups, and the H2O/N2 selectivity can be raised to more than 105 (Wang et al., 2001).

AC C

For the dehydration of natural gas the preferred highly selective membranes are sulphonated poly(ether sulphone) and a (polybutylene terephthalate)/(poly(ethylene oxide) or PBT/PEO block copolymer of molecular weight 1000 Da (Metz et al., 2005). The latter polymer (Fig. 1) had soft rubbery amorphous hydrophilic and hard rigid crystalline hydrophobic segments. It has been found that the water vapour permeability decreases with increasing temperature, which was attributed to the decrease in water solubility on heating (Metz et al., 2002). A cellulose ether composite membrane of high water vapour permeability and selectivity is proposed for the removal of water vapour from pressurized gases or gas mixtures (Ohlrogge et al., 2002). The removal of water vapour from a gas mixture can be accelerated by impregnating a membrane composed of regenerated cellulose with a hygroscopic electrolyte such as lithium bromide. The flux is increased by a factor of 2 to 5, and there is no adverse effect on selectivity (Van Wijk and Jansen, 1990).

EP

TE D

M AN U
3

SC

RI PT

ACCEPTED MANUSCRIPT
148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 2.3 Flue gas dehydration Appropriate polymers for flue gas dehydration are suggested to be sulphonated poly(ether ether ketone), or SPEEK, and the commercial polymer PEBAX 1074, which contains 45wt% of a polyether block amide (nylon 12) and 55wt% PEO (Sijbesma et al., 2008). Data for the latter material have been included in Table 1.xperiments have been conducted on this polymer using an artificial flue gas containing 11.2 vol% water, the remaining gases being nitrogen, CO2 and oxygen. A continuous removal of 0.6-1 kg/m2h was achieved over a 150 h test at 50C. Long term tests on a real, aggressive flue gas stream gave an average water vapour removal rate of 0.2-0.46 kg/m2h over 5300 h. Although the flux decreased with time because of the deposition of fly ash dust and gypsum crystals on the membrane surface, performance was judged to be adequate. The product water quality was not sufficient for reuse in the steam cycle, but was satisfactory for use as feed for a demineralised water plant. 2.4 Dehydration of ethanol

The production of 99.9% ethanol from 94% ethanol has been achieved by vapour permeation (Sander and Janssen, 1991). Although the membrane material is not revealed, the authors remark on the use of the same materials for vapour permeation and pervaporation of aqueous alcohol, and in an earlier article quote PVA as the membrane of choice for this application (Sander and Soukup, 1988). Vapour permeation was claimed to have the advantages of a lower required membrane area, higher flow rates, and the avoidance of harsh chemical reaction conditions. The volumetric flow rate is about 400 times higher for a vapour than a liquid feed at a given mass flow rate and comparable temperature and pressure conditions. Nevertheless, pervaporation seems to be well established in this area. A composite PVA membrane commercialized for pervaporation when tested in vapour permeation mode has yielded fluxes as high as 5.5 kg/m2h for 90 wt% aqueous ethanol at a pressure of 500 kPa, down to 0.7 kg/m2h when the pressure was 100 kPa (Jansen et al., 1992). The addition of sodium montmorillonite clay to PVA decreased the water vapour permeation rate from 0.44 to 0.38 kg/m2h for 90 wt% ethanol when there was 10 wt% clay present (Yeh et al., 2003). This was ascribed to an increase in the tortuosity of the diffusion pathway because of the barrier properties of the dispersed clay. Composite membranes of PVA coated on Nylon-4 had a flux of 0.088 kg/m2h in the vapour permeation dehydration of 10 wt% aqueous ethanol at 25C (Lee et al., 1992). In pervaporation mode the flux was 0.42 kg/m2h, but the respective separation factors were 94 and 14. Using microporous silica membranes formed by coating a microporous silica layer onto a ceramic porous tube (Gallego-Lizon et al., 2002), a water flux increase of approximately 100 times to 10 kg/m2h has been reported for an increase in temperature from 25 to 120C during the dehydration of a 96 wt% ethanol/water mixture, whilst the ethanol flux remained low at less than 0.1 kg/m2 h (van Veen et al., 2001). The dehydration of aqueous ethanol by vapour permeation has been studied using tubular NaY zeolite membranes at 90-110C (Sato et al., 2008). For a 90 wt% ethanol feed at 110C a flux of 20 kg/m2 h was obtained, to give a retentate that was 96.1 wt% ethanol. At a flow rate of 5 kg/m2 h the product was 98.5 wt% ethanol. For a pure water feed a flux of 80 kg/m2 h was achieved. With NaA zeolite membranes and 10 wt% water content a vapour permeation flux of 11 kg/m2 h at 125C has been obtained (Richter et al., 2006). Further results have been obtained at lower temperatures: 2.5 kg/m2 h at 90C and 5 kg/m2 h at 110C (Sommer and Melin, 2005). Some change in the crystal structure of LTA zeolite has been observed after its use in dehydrating

AC C

EP

TE D

M AN U
4

SC

RI PT

ACCEPTED MANUSCRIPT
197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 aqueous ethanol (Kyotani et al., 2009). A small deterioration in the water permeance was observed, and some 10% of sodium ions were detached, indicating proton exchange. Membranes composed of aromatic polyamides made from the reaction of bis[4-(4aminophenoxy) phenyl]diphenylmethane with the diacids terephthalic acid, 5-tert-butylisophthalic acid or 4,4-hexafluoroisopropylidenedibenzoic acid have been used in vapour permeation experiments (Fan et al., 2002). The highest flux of 0.2 kg/m2 h was obtained with the fluorinated species. It was found that the permeation rate could be increased by introducing a bulky group into the polymer backbone. Other fluorinated polymers have given similar results, with best flux being 0.38 kg/m2 h (Teng et al., 2000). Polyimide membranes have been blended with sulphonated polyethersulphone to give enhanced results of 0.64 kg/m2 h for an air sweep mode of operating vapour permeation at 80-100C (Wu et al., 2002). Earlier work on a series of polyimide membranes had shown large variations in performance in treating 10 wt% aqueous ethanol at 75C, the permeation of water being some 10 times greater in the polyimide made from 3,3,4,4-biphenyltetracarboxylic anhydride and diaminodiphenylsulphone compared to that prepared from pyromellitic anhydride and 4,4oxydianiline (Okamoto et al., 1992). This was attributed to the high rigidity and bulkiness of the polymer backbone in the former case preventing close chain packing and giving a more open structure for water transport. The vapour permeation performance of symmetrical and asymmetrical polycarbonate membranes, prepared by dry-phase and wet-phase inversion methods, has been studied for aqueous ethanol (Wang et al., 2005). In the case of the symmetrical polycarbonate membrane, vapour permeation was found to have a significantly increased separation factor and a slightly decreased permeation rate compared to pervaporation, the respective permeation rates being 0.16 and 0.18 kg/m2h for a 20 wt% aqueous ethanol at 25C. 2.5 Dehydration of isopropanol

2.6 Dehydration of acetonitrile In a comparison of vapour permeation and pervaporation for dehydrating aqueous acetonitrile with a silica membrane, a total flux of 3.9 kg/m2h at 70C has been measured (Fontalvo et al., 2005). In an economical evaluation it was ascertained that at high water concentrations or just for overcoming the azeotropic composition, vapour permeation was preferred, but for water concentrations lower than the azeotropic amount, pervaporation was the better option.

AC C

The performance of commercially available crosslinked PVA and microporous silica membranes for the dehydration of IPA/water mixtures by pervaporation/vapour permeation has been studied (Gallego-Lizon et al., 2002). The PVA membranes used in the study belong to a generic family of asymmetric composite membranes for water permeation. These membranes have a supporting layer of nonwoven porous polyester onto which an ultrafiltration membrane is cast and finally a layer of crosslinked PVA. For the range of conditions investigated in the study, water fluxes generally increased with water concentration and increased with the operating temperature from 70 to 105C. The water flux through the microporous silica membrane was found be to up to three times higher than that through the PVA membrane. A water flux of 21.5 kg/m2 h across the microporous silica membrane was reported at 105C for a 12.5 wt% water mixture. However, the IPA flux through the microporous silica membrane was much higher than that through the PVA membrane

EP

TE D

M AN U
5

SC

RI PT

ACCEPTED MANUSCRIPT
247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 2.7 Steam recovery Water vapour as steam is the most universal energy carrier. Its application is wide spread and can be found in all aspects of industrial processes. Industry converts more than 70% of the fuel it purchases for energy into steam. Waste steam can be found almost in every plant/factory where steam is used, from large industrial establishments such as refineries, power plants, chemical factories, steel makers, ore mining, to medium and small plants such as sugar mills and food processing facilities. It is usually contaminated and cannot be reused. A common practice in dealing with it is to use a condenser to collect water or to discharge the steam to the atmosphere. Discharging the spent steam to the atmosphere is not only an energy loss, but is at the same time an environmental issue as water vapour is a major driver of greenhouse gas-induced climate change (Hoang and Nguyen, 2009). The steam consumption in a typical thermal power station of a 1,000 MW capacity is about 2,800 t/h, which translates to about 800 kg/s of condensate. A 10,000 tonnes/day discharge of waste steam represents a loss of ~0.7 million m3 of natural gas/day or ~$40M/year (at 0.4 cent/MJ wholesale to industry). While no direct figures are available in Australia, recycling of steam has significant potential to improve the overall energy efficiency of industry which accounts for 25 Mt CO2/year emissions. With higher energy costs and a growing concern regarding environmental impact, it is highly desirable to recover the energy loss by recycling the spent steam. Membrane processes are being sought that effectively separate contaminants and recover the cleaned industrial waste steam. A high rate of diffusion of water vapour through a non-porous ionic membrane is reported as the key to producing high purity steam at temperatures approaching or exceeding 100C (Spiegelman and Blethen, 2006). The preferred membrane polymer was a copolymer of perfluorinated ethylene and perfluorinated vinyl compound containing an acid group (sulphonic or carboxylic) or its salt. An example is the sulphonic acid copolymer Nafion, a copolymer of tetrafluoroethylene and perfluoro(4-methyl-3,6-dioxa-7-octene-1-sulphonic acid. A membrane that is substantially gas impermeable is desirable, and this copolymer has a permeability of water vapour that is more than three orders of magnitude greater than the permeability of CO2 or CO, and some six orders of magnitude greater than the permeability of oxygen or nitrogen. PVA/silica nanoparticle composite membranes have been tested for steam recovery (Hoang and Nguyen, 2009). Water vapour fluxes of 70-150 kg/m2h were achieved at a differential pressure of 6 bars. A steam permeation plant that uses an LTA zeolite membrane has been installed at a sugar works. It treats 93 wt% bio-ethanol, obtaining a permeate that is below 0.1 wt% ethanol (Caro and Noack, 2008). Quite high fluxes are reported: 11.9, 14.9, 17.6 and 22.4 kg/m2h at 100, 110, 120 and 130C respectively. 2.8 Miscellaneous water removal applications Vapour permeation studies of mixtures propanol-methanol-water and propanol-methanol have been performed with a commercial hydrophilic PVA-polyacrylonitrile composite membrane (Will and Lichtenthaler, 1992). Also, the binary systems ammonia-water and methylamine-water were investigated using a commercial amine resistant PVA-polysulphone composite membrane. With the exception of the non-aqueous propanol-methanol system the separation factors and fluxes obtained for the binary systems were sufficiently high for practical application. In the case of the ternary mixture, vapour permeation also showed a much better separation than pervaporation and a reasonable flux. This was not true for the non-aqueous system though, when the flux was extremely small at <1% of that for vapour permeation. In general, the vapour permeation separation factors were larger, but the flux

AC C

EP

TE D

M AN U
6

SC

RI PT

ACCEPTED MANUSCRIPT
297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 was usually smaller than for pervaporation, so that separation via vapour permeation was possible when pervaporation failed. In practical applications, however, this is confined to those cases where the flux is sufficiently high. 3. Membranes Used in Vapour Permeation Processes A summary of the range of results viewed so far for water separation systems is given in Table 2. For enhanced passage of water vapour, introducing polar groups is a distinct advantage. There appears to be a further advantage in having an inorganic rather than a hydrophilic organic polymer membrane for the dehydration of ethanol, with silica and zeolite types being the best. 4. Comparison of Vapour Permeation and Pervaporation

Figure 2 presents a schematic of the vapour permeation and pervaporation processes, which shows how closely aligned are the engineering aspects of the processes (Kujawski, 2000). As mentioned earlier, vapour permeation is similar in principal to pervaporation, the only difference being that the feed for vapour permeation is a mixture of vapour or vapour and gases, whereas pervaporation involves a liquid feed and a liquid-vapour phase change to achieve the separation (Neel 1991). There is no phase change in vapour permeation, so the addition of heat equivalent to the enthalpy of vaporisation is not required in the membrane unit and there is no temperature drop along the membrane (Ito et al. 1997, Kujawski 2000). In vapour permeation, a slightly superheated vapour is usually employed to prevent condensation. However, the flux decreases with increasing superheating. Therefore, on the feed side of the membrane model, the local pressure should be as close as possible to the local saturation pressure in order to obtain a higher flux. This is very important with the feed side pressure losses of a vapour permeation model. Even a small degree of superheating may cause a significant decrease in flux, whereas selectivity seems not to be sensitive in this regard. Also, vapour condensation on the feed side of model cannot be avoided, depending on the operating conditions. The membrane hence has to be installed vertically so that if condensation occurs, the condensate flows downwards to the bottom of the cell instead of accumulating on the surface of the membrane. For a long membrane lifetime this may be essential (Chen and Lichtenthaler, 1995). A direct comparison of vapour permeation and pervaporation performance is not easy. The data available are summarised in Table 3. The application of NaA zeolite membranes from different sources using the two techniques for ethanol dehydration has been reported (Sato et al., 2008). With vapour permeation and 10 wt% aqueous ethanol a flux of 11 kg/m2 h is possible at 125C, versus pervaporation mode results of 10 kg/m2 h at 120C (Richter et al., 2006). With vapour permeation, composite membranes of PVA coated onto Nylon-4 had a flux of 0.088 kg/m2h in the dehydration of 10 wt% water/ethanol at 25C, whereas in pervaporation mode the flux was much greater, at 0.42 kg/m2h (Lee et al., 1992). There was a higher separation factor for vapour permeation (94 versus 14). The results can be explained by membrane swelling, which will not be significant for inorganic membranes, but for membranes made from hydrophilic organic polymers direct contact of the membrane with the feed as in pervaporation will mean greater swelling and a higher flux rate. In some cases, as with certain polyimide membranes, the swelling can be so great that membrane performance is spoiled (Okamoto et al., 1992). This is in contrast with vapour permeation where the membrane is not in contact with the liquid feed. There is a larger permselectivity for water for vapour permeation, demonstrated by the significantly higher concentration of water in the permeate (Teng et al., 2000).

AC C

EP

TE D

M AN U
7

SC

RI PT

ACCEPTED MANUSCRIPT
347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 A more detailed study of hybrid processes with distillation has shown that for water removal from acetonitrile with a silica membrane at high water concentrations or just for overcoming the azeotropic composition, vapour permeation is preferred, but for water concentrations lower than the azeotropic amount, pervaporation is the better option (Fontalvo et al. 2005). Total and energy costs are slightly lower for vapour permeation than for pervaporation for the high water content mixtures for this application. In summary, vapour permeation, relative to pervaporation is capable of higher fluxes of water in certain systems (Sander and Soukup, 1988), but this may be diminished with highly swollen membranes, which favour greater fluxes for pervaporation (Fan et al., 2002) increases the permselectivity of water: for 10% aqueous ethanol at 25C there is 97.7% water in the permeate, versus 54.5% for pervaporation (Teng et al., 2000) is more heat sensitive, as judged from Arrhenius plots (Fan et al., 2002) has a diffusion coefficient of the vapour phase that is about 400 times higher than that of the liquid phase (Sander and Janssen, 1991) a high selectivity is not necessarily accompanied by a lower permeability, as such a relationship does not hold always for water vapour in a mixture with a permanent gas, where most of the highly selective polymers also have a very high permeability the addition of heat equivalent to the enthalpy of evaporation is not required as there is no phase change (Ito et al., 1997) requires a smaller membrane area (Sander and Janssen, 1991, Fontalvo et al., 2005) has less likelihood of membrane damage caused by impurities in the feed (Sander and Soukup, 1988) operation in the vapour phase eliminates concentration polarisation (Kujawaski, 2000) membrane life is expected to be longer because of less swelling of the membrane (Hayashi et al., 2000). 5. Conclusions

Pervaporation is well established for dehydration applications, but it is likely that vapour permeation will intrude into this territory, its emphasis to date being on the extraction and recovery of organic species from a wide range of gas streams. The vapour permeation process is already used commercially for the dehydration of natural gas and organic solvents, especially alcohol, as well as for the removal of water from air. However, there may be further opportunities for the design of improved membranes of more ionic character, including polysalts. In a newer area, very high fluxes have been achieved now for steam separation and recovery at high temperatures using a composite PVA/silica nanoparticle membrane. Thus composite organic-inorganic membranes clearly have a promising future. The energy conservation and emission savings aspects of this process warrant further research to complete development. Although many advances have been made, the technology is still its early stages, and several challenges lie ahead, among them membrane stability at high temperatures. Again, composite membranes will have a role here. References Alter, H., 1962. A critical investigation of polyethylene gas permeability. Journal of Polymer Science 57, 925-935. Aptel, P., Cuny, J., Jozefowicz, J., Morel, G., Neel, J., 1972. Liquid transport through membranes prepared by grafting of polar monomers onto poly(tetrafluoroethylene) films. I.

AC C

EP

TE D

M AN U
8

SC

RI PT

ACCEPTED MANUSCRIPT
396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 Some fractionations of liquid mixtures by pervaporation. Journal of Applied Polymer Science 16(5), 1061-1076. Baker, R.W., Wijmans, J.G., Kaschemekat, J.H., 1998. The design of membrane vapor-gas separation systems. Journal of Membrane Science 151(1), 55-62. Bolto, B., Hoang, M., Xie, Z., 2011. A review of membrane selection for the dehydration of aqueous ethanol by pervaporation. Chemical Engineering and Processing: Process Intensification 50(3), 227-235. Brinkmann, T., Dijkstra, M., Ebert, K., Ohlrogge, K., 2003. Improved simulation of a vapour permeation module. Journal of Chemical Technology & Biotechnology 78(2-3), 332-337. Caro, J., Noack, M., 2008. Zeolite membranes - Recent developments and progress. Microporous and Mesoporous Materials 115(3), 215-233. Chen, Y., and Lichtenthaler, R. N. 1995. Membrane separations technology: Principals and application, edited by R.D. Noble and S .A. Stern, Elsevier Science, pp. 101-105. Fan, S.-C., Li, C.-L., Wang, Y.-C., Lee, K.-R., Liaw, D.-J., Lai, J.-Y., 2002. Application of aromatic polyamide membranes for pervaporation and vapor permeation. Desalination 148(13), 43-48. Fontalvo, J., Cuellar, P., Timmer, J.M.K., Vorstman, M.A.G., Wijers, J.G., Keurentjes, J.T.F., 2005. Comparing Pervaporation and Vapor Permeation Hybrid Distillation Processes. Industrial & Engineering Chemistry Research 44(14), 5259-5266. Gallego-Lizon, T., Ho, Y.S., Freitas dos Santos, L., 2002. Comparative study of commercially available polymeric and microporous silica membranes for the dehydration of IPA/water mixtures by pervaporation/vapour permeation. Desalination 149(1-3), 3-8. Hayashi, Y., Yuzaki, S., Kawanishi, T., Shimizu, N., Furukawa, T., 2000. An efficient ethanol concentration process by vapor permeation through asymmetric polyimide membrane. Journal of Membrane Science 177(1-2), 233-239. Hoang, M., Nguyen, C., 2009. Membrane and process for steam separation, purification and recovery. Australian Patent PCT/AU2009/000386. Ishida, M., Tasaka, Y. and Asaeda, M. 2005. A study on vapor permeation and pervaporation of acetic acid/water mixtures by porous silica membranes. Journal of Chemical Engineering of Japan 38(1), 903-907. Ito, A., Feng, Y., Sasaki, H., 1997. Temperature drop of feed liquid during pervaporation. Journal of Membrane Science 133(1), 95-102. Jansen, A.E., Versteeg, W.F., van Engelenburg, B., Hanemaaijer, J.H., ter Meulen, B.P., 1992. Methods to improve flux during alcohol/water azeotrope separation by vapor permeation. Journal of Membrane Science 68(3), 229-239. Jonquires, A., Clment, R., Lochon, P., Nel, J., Dresch, M., Chrtien, B., 2002. Industrial state-of-the-art of pervaporation and vapour permeation in the western countries. Journal of Membrane Science 206(1-2), 87-117. Kujawaski, W., 2000. Application of pervaporation and vapour permeation in environmental protetction. Polish Journal of Environmental Studies 9(1), 13-26. Kyotani, T., Ikeda, T., Saito, J., Nakane, T., Hanaoka, T., Mizukami, F., 2009. Crystal Structure of Tubular Na LTA Zeolite Membrane Used for a Vapor Permeation Process: Unusual Distribution of Adsorbed Water Molecules. Industrial & Engineering Chemistry Research 48(24), 10870-10876. Lee, K.-R., Chen, R.-Y., Lai, J.-Y., 1992. Plasma deposition of vinyl acetate onto Nylon-4 membrane for pervaporation and evapomeation separation of aqueous alcohol mixtures. Journal of Membrane Science 75(1-2), 171-180. Leemann, M., Eigenberger, G., Strathmann, H., 1996. Vapour permeation for the recovery of organic solvents from waste air streams: separation capacities and process optimization. Journal of Membrane Science 113(2), 313-322.

AC C

EP

TE D

M AN U
9

SC

RI PT

ACCEPTED MANUSCRIPT
446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 Liu, S., Wang, F., Chen, T., 2001. Synthesis of Poly(ether ether ketone)s with High Content of Sodium Sulfonate Groups as Gas Dehumidification Membrane Materials. Macromolecular Rapid Communications 22(8), 579-582. Metz, S.J., Potreck, J., Mulder, M.H.V., Wessling, M., 2002. Water vapor and gas transport through a poly(butylene terephthalate) poly(ethylene oxide) block copolymer. Desalination 148(1-3), 303-307. Metz, S.J., van de Ven, W.J.C., Potreck, J., Mulder, M.H.V., Wessling, M., 2005. Transport of water vapor and inert gas mixtures through highly selective and highly permeable polymer membranes. Journal of Membrane Science 251(1-2), 29-41. Neel, J., 1991. Introduction to pervaporation in: Pervaporation Membrane, Separation Process, Elseviedr, Amsterdam. Ohlrogge, K., Nitsche, V., Wind, J., 2002. Arrangement for removing water vapour from pressurised gases or gas mixtures, US Patent 6,485,545. Okamoto, K., Tanihara, N., Watanabe, H., Tanaka, K., Kita, H, Nakamura, A., Kusuki, Y. and Nakagawa, K. 1992 Vapor permeation and pervaporation separation of water-ethanol mixtures through polyimide membranes. Journal of Membrane Science, 68, 53-63. Richter, H., Voigt, I., Khnert, J.-T., 2006. Dewatering of ethanol by pervaporation and vapour permeation with industrial scale NaA-membranes. Desalination 199(1-3), 92-93. Sander, U., Janssen, H., 1991. Industrial application of vapour permeation. Journal of Membrane Science 61, 113-129. Sander, U., Soukup, P., 1988. Design and operation of a pervaporation plant for ethanol dehydration. Journal of Membrane Science 36, 463-475. Sato, K., Sugimoto, K., Nakane, T., 2008. Mass-production of tubular NaY zeolite membranes for industrial purpose and their application to ethanol dehydration by vapor permeation. Journal of Membrane Science 319(1-2), 244-255. Scott, K., 1998. Handbook of Industrial Membranes, Elsevier Advanced Technology, London. Sijbesma, H., Nymeijer, K., van Marwijk, R., Heijboer, R., Potreck, J., Wessling, M., 2008. Flue gas dehydration using polymer membranes. Journal of Membrane Science 313(1-2), 263-276. Sommer, S., Melin, T., 2005. Influence of operation parameters on the separation of mixtures by pervaporation and vapor permeation with inorganic membranes. Part 1: Dehydration of solvents. Chemical Engineering Science 60(16), 4509-4523. Spiegelman, J.J., Blethen, R.D., 2006. Method of producing high purity steam. US Patent PCT/US2006/035790. Teng, M.-Y., Lee, K.-R., Fan, S.-C., Liaw, D.-J., Huang, J., Lai, J.-Y., 2000. Development of aromatic polyamide membranes for pervaporation and vapor permeation. Journal of Membrane Science 164(1-2), 241-249. van Veen, H.M., van Delft, Y.C., Engelen, C.W.R., Pex, P.P.A.C., 2001. Dewatering of organics by pervaporation with silica membranes. Separation and Purification Technology 22-23, 361-366. Van Wijk, H.F., Jansen, A.E., 1990. Method and membrane for the removal of water vapour from a gas/vapour mixture by means of vapour permeation. Wang, Z., Chen, T., Xu, J., 2001. Gas and Water Vapor Transport through a Series of Novel Poly(aryl ether sulfone) Membranes. Macromolecules 34(26), 9015-9022. Wang, Y.-C., Teng M.-Y., Lee, K.-R. and Lai, J.-Y. 2005. Comparison between the pervaporation and vapor permeation performances of polycarbonate membranes. European Polymer Journal 41, 1667-1673. Will, W. and Lichtenthaler, R. N. 1992. Comparison of the separation of mixtures by vapor permeation and by pervaporation using PVA composite membranes. II. The binary systems

AC C

EP

TE D

M AN U
10

SC

RI PT

ACCEPTED MANUSCRIPT
496 497 498 499 500 501 502 503 504 505 506 ammonia-water, methylamine-water, l-propanol-methanol and the ternary system 1-propanolmethanol-water. Journal of Membrane Science 68, 127-131. Wu, Y., Peng, X., Liu, J., Kong, Q., Shi, B., Tong, M., 2002. Study on the integrated membrane processes of dehumidification of compressed air and vapor permeation processes. Journal of Membrane Science 196(2), 179-183. Yeh, J.-M., Yu, M.-Y., Liou, S.-J., 2003. Dehydration of wateralcohol mixtures by vapor permeation through PVA/clay nanocomposite membrane. Journal of Applied Polymer Science 89(13), 3632-3638.

AC C

EP

TE D
11

M AN U

SC

RI PT

Table Click here to download Table: VP WR tables.docx

ACCEPTED MANUSCRIPT
Table 1 Water vapour permeabilities and water/nitrogen selectivities at 30C for various organic polymers (Metz et al. 2005, Sijbesma et al. 2008) Water Selectivity, Permeability, H2O/N2 Barrer PEBAX 1074a 160,000 200,000 PBT/PEO block copolymerb 85,500 40,500 Sulphonated poly(ether ether ketone) 61,000 10,200,000 Polydimethylsiloxane 40,000 143 Sulphonated poly(ether sulphone) 15,000 214,000 Ethyl cellulose 20,000 6,060 Cellulose acetate 6,000 24,000 Poly(phenylene oxide) 4,060 1,068 Poly(ether sulphone) 2,620 10,500 Natural rubber 2,600 299 Polysulphone 2,000 8,000 Polycarbonate 1,400 4,670 Polystyrene 970 388 Polyimide 640 5,330,000 Polyacrylonitrile 300 1,880,000 Poly(vinyl chloride) 275 12,500 Polyamide 6 275 11,000 Polypropylene 68 227 Poly(vinyl alcohol) 19 33,300 Polyethylene 12 6 a PEBAX 1074 is a blend of polyether block amide (nylon 12) and poly(ethylene oxide) b (Polybutylene terephthalate)/poly(ethylene oxide) block copolymer Polymer

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Table 2 Water transfer data for membranes used for vapour permeation processes
Process Membrane Water Flux, kg/m2h (Temp., C) Reference

NaY zeolite NaA zeolite

TE D

Microporous silica

Isopropanol dehydration Acetonitrile dehydration Steam recovery

PVA Microporous silica Silica

EP

AC C

Nafion

PVA inorganic hybrid LTA zeolite

M AN U
0.38 (25) 0.64 (80-100) 0.1 (25) 10 (120) 20 (110) 11 (125) 2.5 (90) 5 (110) 7 (105) 21.5 (105) 3.9 (70) 68 (106) 70-150 (153) 14.9 (110) 22.4 (130)

Ethanol dehydration

PVA PVA PVA/clay PVA on Nylon-4 PVA composite Polyamide & fluorinated version Fluorinated polyamide Modified polyimide

0.44 0.38 0.088 (25) 0.7-5.5 (89) 0.20

SC

Drying of compressed air Flue gas dehydration

Polydimethylsiloxane Cellulose acetate SPEEK PEBAX 1074

0.2-0.46 (50) -

Sijbesma et al. 2008

Sijbesma et al. 2008

Sander and Soukup 1988 Yeh et al. 2003 Lee et al. 1992 Jansen et al. 1992 Fan et al. 2002 Teng et al. 2000 Wu et al. 2002 van Veen et al. 2001 Sato et al. 2008 Richter et al. 2006 Sommer and Melin 2005 Gallego-Lizon et al. 2002

Fontalvo et al. 2005

Spiegelman and Blethen 2006 Hoang and Nguyen 2009 Caro and Noack 2008

RI PT

Natural gas dehydration

PEO/PBT copolymer Sulphonated poly(ether sulphone)

Metz et al. 2005

ACCEPTED MANUSCRIPT
Table 3 Comparison of water fluxes in vapour permeation and pervaporation processes Process Membrane Vapour Permeation Water Flux, kg/m2h (Temp., C) 10 wt%: 11 (125) 18 wt%: 15 (125) 10 wt%: 0.088 (25) 44 wt%: 0.087 (25) 10 wt%: 0.20 (55) 90 wt%: 0.14 (55) 10 wt%: 0.38 (25) 90 wt%: 0.285 (25) Higher for high water content Pervaporation Water Flux, kg/m2h (Temp., C) 10 wt%: 10 (120) 18 wt%: 13 (120) Reference

PVA on Nylon-4 PVA composite Polyamide

10 wt%: 0.42 (25) 50 wt%: 0.90 (25) 10 wt%: 0.33 (55) 90 wt%: 0.22 (55)

SC

10 wt%: 0.47 (25) 90 wt%: 0.293 (25)

Acetonitrile dehydration, distillation hybrid

Silica

M AN U

Higher for low water content

AC C

EP

TE D

RI PT

Ethanol dehydration, varying water contents

NaA zeolite

Richter et al. 2006 Lee et al. 1992 Fan et al. 2002 Teng et al. 2000 Fontalvo et al. 2005

Figure

ACCEPTED MANUSCRIPT

Fig.1. Structure of PBT/PEO block copolymer (Metz et al., 2005).

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Fig. 2. Flow diagram for the vapour permeation (VP) and pervaporation (PV) processes (after Kujawski, 2000).

AC C

EP

TE D

M AN U

SC

RI PT