Loose Nanofiltration-Based Electrodialysis For Highly Efficient Textile Wastewater
Loose Nanofiltration-Based Electrodialysis For Highly Efficient Textile Wastewater
Loose Nanofiltration-Based Electrodialysis For Highly Efficient Textile Wastewater
Wenyuan Ye, Riri Liu, Xiangyu Chen, Qin Chen, Jiuyang Lin, Xiaocheng Lin, Bart Van
der Bruggen, Shuaifei Zhao
PII: S0376-7388(20)30760-2
DOI: https://doi.org/10.1016/j.memsci.2020.118182
Reference: MEMSCI 118182
Please cite this article as: W. Ye, R. Liu, X. Chen, Q. Chen, J. Lin, X. Lin, B. Van der Bruggen, S. Zhao,
Loose nanofiltration-based electrodialysis for highly efficient textile wastewater treatment, Journal of
Membrane Science (2020), doi: https://doi.org/10.1016/j.memsci.2020.118182.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. 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.
6 Lin).
7
a
Fujian Provincial Key Laboratory of Soil Environmental Health and Regulation,
10
b
Fujian Provincial Engineering Research Center of Rural Waste Recycling
12 350116, China
13
c
College of Chemical Engineering, Fuzhou University, 350116, Fujian, China
14
d
Department of Chemical Engineering, Process Engineering for Sustainable Systems
16
e
Deakin University, Geelong, Institute for Frontier Materials, VIC 3216, Australia
17
18
19
20
1
21 Abstract: Effective fractionation of dyes and salts (NaCl) potentially paves a new
24 (ED) for dye and salt separation. The NF membrane with a molecular weight cut-off
25 of 678 Da showed a sufficient retention for reactive dyes with small molecular
26 weights (<627 Da) predominantly based on synergistic effect of steric repulsion and
28 transport for anions (i.e., Cl-), as its loose surface structure created extra
29 nano-channels for ion transfer. The anion conductive loose NF membrane based ED
32 process along with 99.4% dye recovery. Additionally, an extremely low fouling
34 charge of the loose NF membrane served as an intrinsic barrier to prevent the dye
35 species from entering the membrane pores through electrostatic repulsion. This study
38
41
42
2
43 1. Introduction
44 The textile industry has been considered one of the most polluted industrial sectors,
45 which potentially exerts a detrimental impact on the eco-system and public health [1,
46 2]. A large quantity of textile wastewater is generated in either dye synthesis or dyeing
47 processes. Generally, textile wastewater is highly loaded with inorganic salts (i.e., 6.0
48 wt.% NaCl), which is derived from the neutralization reaction for dye synthesis or
49 used as additives to enhance the uptake of dyes during dyeing [3]. In the advanced
51 potential resource [4-7]. Particularly, the dye species in textile wastewater can be
52 feasibly extracted and reused for fabric dyeing to reduce the dyeing expenditure;
54 resource for industrial applications, e.g., in pressure retarded osmosis [8], forward
58 exclusively focus on dye degradation or removal, failing to efficiently extract the salts
63 high transmission of NaCl, but sufficiently retains dye species, showing a promising
64 feasibility in fractionation of dyes and salts. For instance, a loose NF membrane with
3
65 an MWCO of 860 Da can completely reject direct dyes (i.e., >99.6%) with 96.7%
66 permeation for NaCl [24]. Ye et al. tailored the surface property of a loose NF
68 high rejections to reactive dyes (>99.3%) but a nearly free transmission for NaCl (i.e.,
69 97.1%) [25]. Zhu et al. designed a loose nanofiltration membrane functionalized with
71 to reactive dyes with outstanding NaCl permeation (i.e., 98.0%) [26]. However, the
73 which significantly reduces the permeation flux and deteriorates the fractionation
74 efficacy of dyes and salts during the diafiltration with considerable water
75 consumption [27], in view of enriching and purifying the dyes from textile
76 wastewater.
77 Apart from loose NF, electrodialysis (ED) also poses a potential in organics/salts
79 mixtures can be continuously transported through the ion exchange membranes under
82 an acid blue 9/salt mixture [31]. Initially, ca. 85% of salt species (NaCl) was removed
83 from this dye/salt binary mixture by ED. However, a compact dye cake layer evolved
84 and deposited onto the surface of the anion exchange membrane for pore blocking,
4
87 valorization from textile wastewater for acid/base generation [32]. Only a moderate
88 amount (i.e., ~70%) of inorganic salts in the reactive dye/salt mixtures was
89 desalinated and converted into acid and base, mainly due to the formation of a dye
90 cake layer on the anion exchange membrane. Such a cake layer promoted the dye
91 molecules to enter the pore structure and bind with the positively charged groups of
94 jeopardized the electrodialytic performance of ion exchange membranes [32, 33]. This
95 strongly urges for advanced strategies to effectively fractionate the dye and salt
97 Taking advantage of both ED and loose NF, a novel stack configuration “loose
98 NF-based ED” using loose NF membranes instead of anion exchange membranes was
99 designed to separate dyes from salts in this study. Especially, the NF membranes
100 allow for ion migration under an established current field [34-36], showing a
101 strategical promise in ion fractionation. Considering the difference in molecular sizes
102 between dyes and salts, it calls for an advanced selective membrane for efficient
103 dye/salt fractionation via ED. Herein, a loose NF membrane with an MWCO of 678
104 Da was applied as the anion conductive membrane in ED for fractionation of dyes and
105 NaCl. Initially, the filtration behavior of the loose NF membrane with NaCl and dye
106 solutions was investigated at an applied pressure of 2 bar to elucidate its selectivity
107 towards reactive dyes (MW<627 Da) and NaCl. Afterwards, the practicability of the
108 loose NF membrane as the anion conductive membrane in ED was demonstrated for
5
109 fractionation of dyes and NaCl under different operation conditions. Ultimately, the
110 fouling propensity of the loose NF membrane in the NF-based ED process was further
111 explored by comparing with a commercial anion exchange membrane to highlight its
113 2. Experimental
115 Three reactive dyes, namely reactive orange 1 (RO1, MW: 615.3 Da), reactive
116 orange 16 (RO16, MW: 617.5 Da) and reactive blue 19 (RB19, MW: 626.6 Da), were
117 used as the typical model dyes. Supplementary Fig. S1 displays the properties (i.e.,
118 chemical structure and size) of these tested reactive dyes. NaCl with 99.0% purity was
119 ordered from Sigma-Aldrich. Ultrapure water with electrical resistance of 18.2
120 MΩ·cm from Milli-Q system was supplied in all the experiments. All the chemicals
123 supplied by Guangdong Yinachuan Environmental Technology Co. (China) and used
124 as anion conductive membrane in the NF-based ED process. The cation (CJMC-1)
125 and anion (CJMA-1) exchange membranes were ordered from ChemJoy Polymer
126 Material Co., Ltd (China) and were used in a conventional ED process as comparison
127 with NF-based ED process. The main properties of these ion exchange membranes are
129
6
131 2.2.1 Surface pore size
132 The pore size of the tested NF membrane was measured by filtrating the aqueous
133 solutions of 100 mg·L-1 polyethylene glycols with various molecular weights (i.e.,
134 200, 300, 400, 600, and 1000 Da) at 4 bar. The concentration of the polyethylene
135 glycols was determined based on the total organic carbon (TOC) concentration
136 through TOC-LCpH/CPN analyzer (Shimadzu, Japan). The rejection coefficient (R)
137 of polyethylene glycols by the NF membrane was calculated via the concentration of
138 polyethylene glycols in the permeate (Cp) and feed (Cf) by Eq. 1:
C
139 R (% ) = 1 − p × 100 (1)
Cf
140 Specifically, the correlation between the Stokes diameter (dp) and molecular weight
143 Based on the retention of polyethylene glycols, the pore size distribution of the
145 function between Stokes diameters of polyethylene glycols and their rejection (R) [37,
146 38]:
dR ( d p ) ( ln d − ln µ )2
147 =
1
exp −
p p
(3)
d p ln σ p 2π ( p )
2
dd p
2 ln σ
148 where μp presents the mean effective pore size of the membranes, as determined at R
149 of 50%. σp denotes geometric standard deviation of pore size, defined as the ratio
150 between dp at R of 84.13% and μp. It is noted that the MWCO of the membrane is
7
152 2.2.2 Surface charge
153 Surface charge of the NF membrane is critical for screening the charged solutes,
154 significantly influencing the membrane performance. Generally, the surface charge of
155 the membrane was assessed in terms of zeta potential through the SurPASS
156 electrokinetic analyzer (Anton Paar GmbH, Austria) in a 10.0 mmol·L-1 NaCl aqueous
157 solution. The solution pH for surface charge measurement ranged from 2.9 to 9.3,
158 which was adjusted through titration with HCl or NaOH. Finally, the pH-dependent
159 zeta potential of the loose NF membrane was recorded and analyzed.
161 The filtration performance of the tested NF membrane was conducted with a
162 lab-made filtration system. Firstly, the NF membrane was initially compacted at 5 bar
163 through filtration of pure water to achieve a steady permeation flux, prior to the
164 filtration experiment. Afterwards, the separation of pure NaCl solutions with variable
165 salinities from 0.05 to 15.0 g·L-1 was carried out at 2 bar and 25 °C. In addition, the
166 filtration of pure reactive dye solutions (RB19, RO16, and RO1) with variable
167 concentrations (0.1, 0.5, and 1.0 g·L-1) was conducted. Finally, the separation of a
168 dye/NaCl (i.e., RO1/NaCl) binary mixture with different salinities (i.e., up to 15.0
169 g·L-1 NaCl) was further performed to evaluate the fractionation of reactive dyes and
171 The rejection (R) of reactive dyes or NaCl for the tested NF membrane was
8
Cp
173 R ( %) = 1 − ×100 (4)
Cf
174 where Cp and Cf are the concentrations of the dyes or NaCl in the permeate and feed,
175 respectively. Particularly, the concentration of the dyes was measured by a UV-1780
176 double beam spectrophotometer (Shimadzu); the concentration of Cl- ions was
179 In this case, the commercial NF membrane was used instead of commercial anion
181 performance for dye and NaCl fractionation. A lab-made ED stack was designed for
182 this experiment, comprised of titanium-based electrodes (i.e., an anode and a cathode)
183 coated with ruthenium and a membrane stack (Fig. 1). Specifically, three pieces of
184 cation exchange membrane (CJMC-1) and two pieces of NF membrane with effective
185 area of 1.94×10-3 m2 were alternatively inserted into the membrane stack. The NaCl
186 solution which had equivalent salt concentration with the feed was applied in the
187 concentrate compartment. A 3.0% Na2SO4 aqueous solution was applied as the
189
9
190 Fig. 1 Configuration of NF-based ED process for dye and salt fractionation (CM:
192
193 Initially, a 300 mL pure NaCl solution with variable concentrations (i.e., 6.0, 9.0,
194 12.0 and 15.0 g·L-1) was used as feed to test the transport behavior of ions through the
196 Subsequently, the transport of the ions through the NF membrane was studied at
197 different currents (i.e., 0.3, 0.4, 0.5, and 0.6 A) to investigate the effect of current on
199 Additionally, a 300 mL RO1/NaCl (i.e., 1.0 g·L-1 RO1 and 12.0 g·L-1 NaCl)
200 mixture was further employed as the feed in the NF-based ED process for separation
201 of RO1 dye and NaCl. The loose NF-based ED process was operated at 0.5 A and
202 25 °C. Once the conductivity of the feed in the diluate chamber was reduced to 1.0
203 mS·cm-1, the loose NF-based ED process was terminated as 1 cycle. Simultaneously,
204 the fouling propensity of the NF membrane as anion conductive membrane caused by
205 dye species was studied based on 5-cycle ED operation at the same condition. As a
206 comparison, the conventional ED process equipped with commercial anion exchange
209 voltage and current across the ED stack were recorded by a regulated power supply
10
211 The desalination efficiency (η) of the loose NF membrane during ED process was
C1
213 η ( % ) = 1 − ×100 (5)
C0
214 where C0 and C1 are the concentrations of NaCl in the feed before and after ED
215 process.
218 Generally, the larger pore size of the NF membranes would diminish the
219 electrostatic repulsion to the ionic species to some extent, which facilitates the
220 transmission of inorganic salts (e.g., NaCl) through the membranes for fractionation
221 of organics and salts [39]. NF membranes with loose surface structures (i.e., 500-1000
222 Da MWCO) are of increasing interest for addressing this challenge. As implied in Fig.
223 2, the NF membrane in this study had an MWCO of 678 Da, demonstrating a loose
224 surface structure for favorable permeation of salts. Additionally, the surface of this
225 loose NF membrane yielded a mean effective pore size (µ p) of 0.48 nm, which can
227
11
228 Fig. 2 Relationship between rejections to polyethylene glycols and the Stokes
229 diameters of the loose NF membrane and theoretical pore size distribution (inset)
230
232 can be clearly observed that the isoelectric point of the tested NF membrane was at
233 pH 3.0. This allows for a highly negative charge density on the loose NF membrane
234 surface at higher pH ranges, mainly due to the dissociation of carboxylic groups
235 present on the poly(piperazine)amide-based surface with low cross-linking degree [21,
236 40]. Specifically, the zeta potential of the NF membrane was -38 mV at neutral pH,
237 indicating an enhanced electrostatic repulsion to the charged organic solutes, such as
239
242 To effectively fractionate the dyes and NaCl in ED process, it calls for steric
243 retention of dyes and permeation of NaCl by the membrane. To study the potential of
245 solutions was initially conducted (Fig. 4). As demonstrated in Fig. 4A, the loose NF
12
246 membrane exhibited a very low rejection to NaCl over the NaCl concentration range.
247 Generally, the increasing salt concentration would weaken the electrostatic repulsion
248 force of the NF membrane surface to the salt solutes through suppressing the Debye
249 screening length for reduced salt retention and vice versa [41, 42]. However, only a
250 moderate rejection of 25.8±1.3% for NaCl was observed when filtrating a 0.05 g·L-1
251 NaCl solution. This is mainly attributed to the considerably large pore size of the
252 loose NF membrane, diminishing the electrostatic repulsion effect to some extent and
253 resulting in fast permeation of NaCl even at low salt concentrations. As the NaCl
254 concentration was increased to 15.0 g·L-1, the loose NF membrane showed a nearly
255 complete salt transport (i.e., 2.7±0.4% rejection). This suggests that the unique
256 surface structure of the loose NF membrane can create sufficient nano-channels for
257 salt transport. Additionally, the loose NF membrane endowed a complete rejection
258 (i.e., >99.2%) to all the reactive dyes with low molecular weights (<627 Da), based on
259 the synergistic effects of electrostatic repulsion and steric exclusion (Fig. 4B).
260
13
261
262 Fig. 4 Separation performance of the loose NF membrane with various solutions: (A)
263 pure NaCl solutions with different salinity; (B) pure reactive dye (RB19, RO16 and
265
266 In dye synthesis as well as in dyeing processes, the salts (e.g., NaCl) with high
267 content would be generated as byproduct, which potentially has a negative impact on
268 filtration performance of the membrane. Therefore, the separation behavior of the
269 loose NF membrane in the RO1/NaCl binary mixtures with different salinities (up to
270 15.0 g·L-1) was investigated (Fig. 4C). Although the elevated content of NaCl in the
271 RO1/NaCl mixture would intensify the shielding effect of the membrane surface for
272 reduced dye rejection [41, 43], an insignificant decline (i.e., ~0.2%) in rejection of
273 RO1 dye was observed over the NaCl concentration range. This suggests that steric
274 exclusion mainly dominates the separation performance of the loose NF membrane
275 for RO1 dye over the Donnan repulsion. On the other hand, the presence of RO1 had
276 no obvious impact on the permeation of NaCl by the loose NF membrane. Especially,
277 an expectably high transmission of NaCl in the RO1/NaCl binary mixtures was
278 yielded, consistently comparable to that in the pure NaCl solutions. Such a separation
14
279 performance demonstrates the superior selectivity of the loose NF membrane towards
283 The loose surface morphology of the tested NF membrane could create extra
284 nano-channels for NaCl permeation but effectively retain the reactive dyes through
286 this loose NF membrane as a role of anion conductive membrane for salt transport in
287 the ED process. Figs. 5 and 6 show the electrodialytic performance of the loose NF
288 membrane for salt transport in ED at different conditions (i.e., salt concentrations and
290
291
15
292 Fig. 5 Electrodialytic performance of the loose NF membrane and the anion exchange
293 membrane for desalination of NaCl solutions with different salinities at a current of
294 0.5 A. (A) Evolution of conductivity in the concentrate and diluate sides during
295 NF-based ED; (B) changes of voltage, current and stack resistance in NF-based ED;
296 (C) performance comparison between NF-based ED and conventional ED; (D)
298
299
300 Fig. 6 Electrodialytic performance of loose NF membrane for desalination of NaCl
301 solution (12 g·L-1) at different current densities. (A) Evolution in conductivity; (B)
303
304 Fig. 5A shows the electrodialytic performance of the loose NF membrane for
305 desalination of NaCl solutions with different salinities at a current of 0.5 A. The
306 conductivity of the brine (i.e., NaCl) solutions in the diluate compartment decayed
307 linearly with the operation time. Simultaneously, the conductivity of the solutions in
308 the concentrate compartment increased steadily, ascribed to the transfer of Na+ and Cl-
309 ions through the cation exchange membrane and the loose NF membrane, indicating
16
310 the anionic conductivities of the loose NF membrane for effective transport of Cl- ions.
311 As the NF-based ED process proceeded, more Na+ and Cl- ions in the diluate
312 compartment were transported through the cation exchange membrane and loose NF
313 membrane, respectively. Therefore, the conductivity of the brine solutions in the
314 diluate chamber was significantly reduced to ca. 0.40 mS·cm-1, demonstrating an
316 membrane at the established current field. For instance, the final content of NaCl in
317 the diluate chamber was reduced to 117.8 mg·L-1 when using a 12.0 g·L-1 NaCl
318 solution as feed, yielding a ca. 99.0% desalination efficiency for the NF-based ED
319 process.
320 Additionally, the voltage across the NF-based ED stack increased gradually at the
321 constant current mode, mainly ascribed to the increasing electrical resistance (Fig.
322 5B). As the Na+ and Cl- ions in the feed were further exhausted at the ultimate stage,
323 the NF-based ED stack showed a remarkable boost in the overall electrical resistance,
324 leading to an increase in voltage across the membrane stack. Once the voltage across
325 the membrane stack was beyond the range of the power supply (i.e., 60.4 V), the
326 NF-based ED stack would be operated at the constant voltage mode. In this case, the
327 current across the NF-based ED stack started to decrease with the continuously
328 increasing overall electrical resistance. On the other hand, increasing initial
329 concentration of NaCl in the feed solutions had no obvious effect on the transport of
330 NaCl through the loose NF membrane. Specifically, the NF-based ED process
331 exhibited a nearly identical decay tendency in conductivity of the brine solutions in
17
332 the diluate compartment over the tested salt concentration range, indicating the same
333 salt transfer rate at the fixed current field as driving force.
335 exchange membranes (CMJA-1) was operated (Fig. 5C). The CMJA-1 membrane
336 yielded a faster transport of Cl- ions than the loose NF membrane, as reflected from
337 the steeper decay in conductivity of the brine solution in the diluate compartment.
338 Generally, the CMJA-1 anion exchange membrane has the functional groups carrying
339 with positive charges and offers a lower area resistance (i.e., 2.52 Ω·cm2) for
340 enhanced ion transfer at a fixed direct current field (Supplementary Fig. S2).
341 However, the loose NF membrane with larger pore size displayed a negatively
342 charged surface, establishing an intrinsic barrier for transport of anions (e.g., Cl- ions)
343 via electrostatic repulsion to some extent, as demonstrated by the higher area
345 Fig. 6 depicts the electrodialytic performance of the loose NF membrane in a 12.0
346 g·L-1 NaCl solution at different current densities (i.e., from 0.3 to 0.6 A). As shown in
347 Fig. 6A, the feed (i.e., NaCl) solution in the diluate chamber had a conductivity that is
348 decreasing with operation time. Increasing the current density across the ED stack can
349 facilitate the transport of Na+ and Cl- ions through the membranes, as demonstrated by
350 a steeper decay in the conductivity of the feed solution. Specifically, a reversely
351 proportional relationship between the current density of the ED process and the
352 operation time was found. This indicates that a shorter operation time would be
353 required at increasing current density without sacrificing the desalination efficiency
18
354 (i.e., 99.0%).
356 In order to sufficiently fractionate the dyes and NaCl, the NF-based ED process
357 was studied for separation of the dye/salt mixtures at a current of 0.5 A, aiming at
359
360
361
19
362 Fig. 7 (A) NF-based ED for fractionation of RO1/NaCl mixture; (B) conventional ED
363 for fractionation of RO1/NaCl mixture; (C) digital pictures of the loose NF membrane
364 and conventional the anion exchange membrane (CMJA-1) before and after fouling
365 caused by dye during ED process; (D) fouling mechanism of the loose NF membrane
368
369 As illustrated in Fig. 7A, the NF-based ED stack yielded a nearly linear decay in
370 the conductivity of the diluate stream at an established current field, indicating the
371 effective transport of Na+ and Cl- ions in RO1/NaCl mixture through the cation
372 exchange membrane and the loose NF membrane. Compared to the case of a pure
373 NaCl solution with 12.0 g·L-1, the ion transport was limited to some extent at the
374 same operation conditions when using the RO1/NaCl mixture as feed, calling for a
375 longer operation duration for desalination. This is plausibly ascribed to the formation
376 of a dye cake layer on the loose NF membrane surface to impede the fast ion transport.
377 However, the dye cake layer had no obvious impact on the efficacy of ion transport.
378 Remarkably, the content of NaCl in the RO1/NaCl mixture was reduced from 12.0 to
379 ~0.13 g·L-1 with 98.9% desalination efficiency during NF-based ED process
380 (Supplementary Fig. S3). Furthermore, the recovery of dye reached 99.4% in this
381 case, due to the nano-channels of the loose NF membrane surface, which would
382 effectively retain the dye species through size exclusion effect (Fig. 4).
20
383 In order to explore the anti-fouling property of the loose NF membrane induced by
384 dye species at the established direct current field, the NF-based ED process was
385 performed in 5 cycles. The NF-based ED stack was observed to show a nearly
386 identical evolution in the conductivity of the concentrate and diluate in all the 5 cycles,
387 demonstrating the extremely low fouling propensity of loose NF membrane for stable
388 ED operation, which can be reflected by Fig. 7C and Supplementary Fig. S2 (i.e.,
389 slight increase in area resistance of the loose NF membrane after fouling). This is
390 mainly due to the negative surface charge of the loose NF membrane which can
391 effectively repel the dye species from entering the pore structure for pore blocking
392 through electrostatic repulsion, highlighting the superiority of the loose NF membrane
393 as anion conductive membrane in fractionation of dye and salt through ED.
394 In contrast, the conventional ED stack equipped with an anion exchange membrane
395 (i.e., CMJA-1) suffered from severe fouling caused by adsorption and deposition of
396 dye molecules into the pore structure of the CMJA-1 anion exchange membrane
397 through electrostatic attraction during a 5-cycle operation (Fig. 7B, 7C and 7D),
398 which remarkably boosted the area resistance of the CMJA-1 anion exchange
400 increased electrical resistance (i.e., up to 3020 Ω) of the ED stack. This would
401 significantly constraint the ion transport of the membrane stack with reduced
402 desalination efficiency (Fig. 7B and Supplementary Fig. S3). Specifically, the
403 content of Cl- ions remained above the level of ~0.27 g·L-1 in the feed for
404 conventional ED. On the other hand, the higher electrical resistance of conventional
21
405 ED stack would produce more heats and a risk of damage in the chemical structure of
406 the CMJA-1 anion exchange membrane (Fig. 7C), thus deteriorating the performance
407 and minimizing the lifespan of the commercial CMJA-1 anion exchange membrane.
408 4. Conclusion
409 In this study, a novel loose NF-based ED process was proposed to fractionate the
410 dye/salt mixture by using a loose NF membrane as anion conductive membrane. Due
411 to the intrinsic loose surface structure, the NF membrane with an MWCO of 678 Da
412 showed a high permeation for NaCl with a sufficient retention (>99.2%) for reactive
413 dyes (MW<627 Da), enabling an outstanding selectivity between dyes and NaCl.
415 conductive membrane during ED application in NaCl solutions, since the loose
416 surface structure created sufficient nano-channels to intensify ion transport. Based on
417 this, a one-step effective fractionation of dye and NaCl from the dye/NaCl mixture
418 can be successfully obtained by the loose NF-based ED process. Specifically, a 98.9%
419 desalination efficiency was achieved along with a 99.4% dye recovery. Compared
420 with the commercial anion exchange membrane (CMJA-1), the loose NF membrane
422 fractionation of dye/NaCl binary mixtures with reduced fouling propensity. This is
423 due to the fact that the intrinsic negative surface charge of the loose NF membrane
424 acted as an effective barrier to repel the dye species from entering the pore structure
425 for pore blocking via electrostatic repulsion. This study provides a conceptual
22
427 effective fractionation of organics and salt species from highly-saline wastewater, in
429 Acknowledgment
430 J. Lin, W. Ye and X. Lin would like to thank the funding support from the National
431 Natural Science Foundation of China (Grant Nos: 21706035, 21707018, and
432 21808038), the Natural Science Foundation of Fujian Province (Grant No:
433 2019Y0006), the Fujian Agriculture and Forestry University Program for
434 Distinguished Young Scholar (Grant No.: xjq201704), the Award Program for
435 Minjiang Scholar Professorship and the Valuable Instrument & Equipment Open Test
436 Fund of Fuzhou University (Grant No.: 2019T024) for this work.
437 References
438 [1] Y.C. Xu, Z.X. Wang, X.Q. Cheng, Y.C. Xiao, L. Shao. Positively charged
440 co-deposition for textile wastewater treatment. Chem. Eng. J. 303 (2016)
441 555-564.
442 [2] K. Paździor, L. Bilińska, S. Ledakowicz. A review of the existing and emerging
445 [3] W. Ye, K. Ye, F. Lin, H. Liu, M. Jiang, J. Wang, R. Liu, J. Lin. Enhanced
23
449 [4] J.S. Guest, S.J. Skerlos, J.L. Barnard, M.B. Beck, G.T. Daigger, H. Hilger, S.J.
451 Raskin, M.C.M. Van Loosdrecht, D. Yeh, N.G. Love. A new planning and design
452 paradigm to achieve sustainable resource recovery from wastewater. Environ. Sci.
454 [5] M. Kamali, D.P. Suhas, M.E. Costa, I. Capela, T.M. Aminabhavi. Sustainability
457 [6] J. Lin, F. Lin, X. Chen, W. Ye, X. Li, H. Zeng, B. Van der Bruggen. Sustainable
460 zero liquid discharge. Ind. Eng. Chem. Res. 58 (2019) 11003-11012.
461 [7] Y. Xu, J. Lin, C. Gao, B. Van der Bruggen, Q. Shen, H. Shao, J. Shen. Preparation
463 potential fractionation of dyes and Na2SO4. Ind. Eng. Chem. Res. 56 (2017)
464 11967-11976.
465 [8] B.E. Logan, M. Elimelech. Membrane-based processes for sustainable power
467 [9] T.Y. Cath, A.E. Childress, M. Elimelech. Forward osmosis: Principles,
468 applications, and recent developments. J. Membr. Sci. 281 (2006) 70-87.
469 [10] T. Chung, S. Zhang, K.Y. Wang, J. Su, M.M. Ling. Forward osmosis processes:
24
471 [11] S. Zhao, L. Zou, C.Y. Tang, D. Mulcahy. Recent developments in forward
472 osmosis: Opportunities and challenges. J. Membr. Sci. 396 (2012) 1-21.
473 [12] J.W. Post, H.V.M. Hamelers, C.J.N. Buisman. Energy recovery from controlled
474 mixing salt and fresh water with a reverse electrodialysis system. Environ. Sci.
476 [13] J. Lin, W. Ye, J. Huang, B. Ricard, M. Baltaru, B. Greydanus, S. Balta, J. Shen,
477 M. Vlad, A. Sotto, P. Luis, B. Van der Bruggen. Toward resource recovery from
478 textile wastewater: Dye Extraction, water and base/acid regeneration using a
479 hybrid NF-BMED process. ACS Sustain. Chem. Eng. 3 (2015) 1993-2001.
480 [14] A. Yurtsever, B. Calimlioglu, E. Sahinkaya. Impact of SRT on the efficiency and
482 for the treatment of textile industry wastewater. Chem. Eng. J. 314 (2017)
483 378-387.
486 for adsorptive removal of dye from textile wastewater in a fixed-bed column.
488 [16] T.A. Nguyen, R. Juang. Treatment of waters and wastewaters containing sulfur
25
493 [18] W. Ye, J. Lin, R. Borrego, D. Chen, A. Sotto, P. Luis, M. Liu, S. Zhao, C.Y.
495 loose nanofiltration membrane for digital ink-jet printing. Sep. Purif. Technol.
497 [19] J. Zhu, J. Wang, A.A. Uliana, M. Tian, Y. Zhang, Y. Zhang, A. Volodin, K.
502 [20] J. Wang, L. Qin, J. Lin, J. Zhu, Y. Zhang, J. Liu, B. Van der Bruggen. Enzymatic
503 construction of antibacterial ultrathin membranes for dyes removal. Chem. Eng. J.
505 [21] J. Lin, C.Y. Tang, C. Huang, Y.P. Tang, W. Ye, J. Li, J. Shen, R. Van den
506 Broeck, J. Van Impe, A. Volodin, C. Van Haesendonck, A. Sotto, P. Luis, B. Van
509 1-14.
510 [22] Y. Xu, M. Wu, S. Yu, Y. Zhao, C. Gao, J. Shen. Ultrathin and stable graphene
512 digital inkjet printing dye. J. Membr. Sci. 586 (2019) 15-22.
513 [23] W. Ye, H. Liu, M. Jiang, J. Lin, K. Ye, S. Fang, Y. Xu, S. Zhao, B. Van der
26
515 recovering humic substance as liquid fertilizer by loose nanofiltration. Water Res.
517 [24] J. Lin, W. Ye, H. Zeng, H. Yang, J. Shen, S. Darvishmanesh, P. Luis, A. Sotto, B.
518 Van der Bruggen. Fractionation of direct dyes and salts in aqueous solution using
520 [25] W. Ye, H. Liu, F. Lin, J. Lin, S. Zhao, S. Yang, J. Hou, S. Zhou, B. Van der
524 [26] J. Zhu, A. Uliana, J. Wang, S. Yuan, J. Li, M. Tian, K. Simoens, A. Volodin, J.
525 Lin, K. Bernaerts, Y. Zhang, B. Van der Bruggen. Elevated salt transport of
528 [27] J. Lin, C.Y. Tang, W. Ye, S. Sun, S.H. Hamdan, A. Volodin, C.V. Haesendonck,
532 [28] W. Pronk, M. Biebow, M. Boller. Electrodialysis for recovering salts from a
534 2414-2420.
535 [29] Y. Zhang, L. Pinoy, B. Meesschaert, B. Van der Bruggen. Separation of small
27
537 57 (2011) 2070-2078.
540 [31] C. Xue, Q. Chen, Y. Liu, Y. Yang, D. Xu, L. Xue, W. Zhang. Acid blue 9
542 [32] Y.W. Berkessa, Q. Lang, B. Yan, S. Kuang, D. Mao, L. Shu, Y. Zhang. Anion
543 exchange membrane organic fouling and mitigation in salt valorization process
548 [34] L. Ge, B. Wu, Q. Li, Y. Wang, D. Yu, L. Wu, J. Pan, J. Miao, T. Xu.
553 approach for sea water demineralization and the production of physiological
555 [36] F. Sheng, L. Hou, X. Wang, M. Irfan, M.A. Shehzad, B. Wu, X. Ren, L. Ge, T.
558 [37] W. Ye, R. Liu, F. Lin, K. Ye, J. Lin, S. Zhao, B. Van der Bruggen. Elevated
28
559 nanofiltration performance via mussel-inspired co-deposition for sustainable
560 resource extraction from landfill leachate concentrate. Chem. Eng. J. 388 (2020)
561 124200.
562 [38] M. Jiang, K. Ye, J. Lin, X. Zhang, W. Ye, S. Zhao, B. Van der Bruggen.
563 Effective dye purification using tight ceramic ultrafiltration membrane. J. Membr.
565 [39] M. Jiang, K. Ye, J. Deng, J. Lin, W. Ye, S. Zhao, B. Van der Bruggen.
568 [40] C.Y. Tang, Y. Kwon, J.O. Leckie. Effect of membrane chemistry and coating
570 membranes: I. FTIR and XPS characterization of polyamide and coating layer
572 [41] X. Wang, C. Zhang, P. Ouyang. The possibility of separating saccharides from a
573 NaCl solution by using nanofiltration in diafiltration mode. J. Membr. Sci. 204
575 [42] Y. Zeng, L. Wang, L. Zhang, J.Q. Yu. An acid resistant nanofiltration membrane
578 [43] J. Qin, X. Dai, Y. Zhou, L. Zhang, H. Chen, C. Gao. Desalting and recovering
29
Highlights:
separation
membrane
3. Loose NF-based ED yields 98.9% desalination efficiency and 99.4% dye recovery
membrane
Declaration of interests
■ The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.
☐The authors declare the following financial interests/personal relationships which may be considered
as potential competing interests: