- Email: [email protected]
Study on the behavior of nanofiltration membranes using for chromium(III) recovery from salt mixture solution
Desalination 315 (2013) 115–123
Contents lists available at SciVerse ScienceDirect
Desalination journal homepage: www.elsevier.com/locate/desal
Study on the behavior of nanofiltration membranes using for chromium(III) recovery from salt mixture solution P. Religa a,⁎, A. Kowalik-Klimczak b, P. Gierycz b, c a b c
Department of Environmental Protection, Radom University of Technology, Chrobrego 27, Radom, Poland Faculty of Chemical and Process Engineering, Warsaw University of Technology, Waryńskiego 1, Warsaw, Poland Institute of Physical Chemistry Polish Academy of Sciences, Kasprzaka 44/52, Warsaw, Poland
H I G H L I G H T S ► ► ► ► ►
Low pH and high salt concentration in chromium solution modified NF membrane surface Two washing procedures to reproduce the properties of NF membranes was analyzed Potential zeta, SEM and permeability of membranes was determined Membrane bathing with HCl reduced its scaling but not the surface charge reproduced Mixed HCl–NaOH bath leads both reduce of membrane scaling and recovery its charge
a r t i c l e
i n f o
Article history: Received 13 August 2012 Received in revised form 26 October 2012 Accepted 27 October 2012 Available online 20 November 2012 Keywords: Nanofiltration (NF) Chromium(III) recovery Membrane charge Permeability Cleaning agents
a b s t r a c t Results of studies of the effects of concentrate salt solutions characterized by low pH, on the nanofiltration membrane surface properties used for the separation of chromium(III) have been presented in this paper. It was shown that the low pH of the concentrate salt solutions and cleaning bath with hydrochloric acid irreversibly altered the charge of tested membranes. As the consequence an instability of permeability and selectivity of the membrane during the process was noticed. The effect of alkaline bath used after cleaning with a solution of hydrochloric acid to regenerate the surface charge of tested membranes was also examined. The results showed that the use of bath in the form of sodium hydroxide leads to the partial recovery of a low surface charge of the membranes. In addition a significant improvement in the stability of the tested NF membranes used for the separation of chromium(III) from concentrate salt solutions at low pH was observed. Moreover, SEM images obtained for the tested membranes cleaned with solutions of HCl and NaOH indicated no mechanical defects in the structure of the examined membranes. © 2012 Elsevier B.V. All rights reserved.
1. Introduction Nanofiltration is a pressure driven process, which assures great process efficiency, as well as enables its great selectivity [1]. On account of these advantages, the nanofiltration is successfully applied to the treatment of surface and underground waters, above all for their softening [2–5] and for the separation of metal ions from industrial wastewaters [6–8]. Several other studies [9–11] as well as our previous works [12–14] demonstrate very interesting possibilities of nanofiltration application for chromium(III) recovery from concentrated salts mixture characterized by low pH. Such solutions are used, among others, in a tanning industry. Separation of chromium(III) from concentrated salts mixture depends on the feed composition [12,13] and properties of the used nanofiltration membranes [14]. The presence of mono- and multivalent negative ions and sufficiently high number of ⁎ Corresponding author. Tel.: +48 48 3617583; fax: +48 48 3617598. E-mail address: [email protected] (P. Religa). 0011-9164/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.desal.2012.10.036
monovalent positive ions produces a Donnan phenomenon which can be additionally increased by the selection of the nanofiltration membrane demonstrating the proper chemical-physical capacity of surface layer. The NF membrane characterized by a negative zeta potential [14] is especially useful for chromium(III) recovery from concentrated salts mixture characterized by low pH. The use of such membrane lowers its polarization causing the increase of permeate flux, and at the same time growth of chromium(III) concentration factor. However, even at optimum feed composition and the use of membrane with appropriate surface properties decrease of its productivities during the nanofiltration process was observed [12]. It was caused by the change of membrane zeta potential by the low pH. The consequence of this fact is the adsorption of ions on the surface and/or in the pores and the deterioration of the membranes selectivity [12]. The change in the membranes selectivity causes changes in the retention of feed components, which may lead to the membrane polarization increase and as a consequence to reduction of its permeability. Because the maintenance of high membrane permeability is preferred for chromium(III)
116
P. Religa et al. / Desalination 315 (2013) 115–123
recovery, it seems to be necessary to remove the adsorbed ions layer from the surface and/or NF membrane pores to keep low values of zeta potential. The aim of this study was to analyze the changes of surface properties of nanofiltration membranes used for chromium(III) recovery from concentrated salt solutions at low pH and to examine the different washing procedures to reproduce the properties of the tested nanofiltration membranes — characteristic zeta potential, as well as their hydraulic permeability. 2. Experimental The experiments were carried out at laboratory scale in cross flow cell made of stainless steel operated in batch mode with circulation presented in Fig. 1. The nanofiltration of model concentrated salt solutions containing 2 gCr 3 +/dm 3, 10 gCl −/dm 3, 10 gSO42 −/dm 3 and characterized by a pH ≈ 4 was conducted for transmembrane pressure on the level of 14 bar and retentate flow equal to 800 dm 3/h. The composition of solution and operation conditions were fixed based on our previous studies [12–14]. The temperature of feed solution during the process was maintained at 25 ± 1 °C by a thermostat. The feed solution was pumped from the feed tank toward to nanofiltration membrane, obtaining a retentate that was returned to the feed/retentate tank and a permeate that was collected in the permeate tank. Two kinds of commercial nanofiltration flat sheet membranes (under symbol DL and HL) provided by GE Osmonics were used in the experiments. The tested nanofiltration membranes had an active layer made of the poly(piperazine-amide) (Fig. 2). The isoelectric point (IP) was less than 3.0 and 3.3 for DL and HL membrane, respectively [14–16]. Such membrane IP indicated on its negative zeta potential under experimental pH. The DL membrane had active three-layer — dense membrane structure, while the HL membrane had active two-layer — loose membrane structure. Each of these membranes has an effective area of 0.0155 m2. After nanofiltration the tested membranes were cleaned with acid and alkaline baths according with procedures presented in Tables 1 and 2. pH of cleaning baths was chosen on the basis of the tested membranes characteristics regarding their chemical resistance [14]. Temperature of the acidic and alkaline cleaning bath was constant and equal to 18 ± 1 °C. The changing of membrane permeability under the influence of concentrated salt solutions and the cleaning according to proposed procedures was analyzed based on the JP = f(ΔP) dependence designated for the deionized water (TMP = 10–24 bar, QR = 800 dm 3/h, t = 25 ± 1 °C) (Tables 1 and 2). Samples of permeate and retentate have been collected for determination of chromium(III) concentration in well-defined time intervals.
O
O
C
C
N
NH
COOH Fig. 2. Poly(piperazine-amide) formula [17–19].
After the end of the experiment, samples of permeate and retentate have been collected for determination of the chloride concentration. The samples of permeate, feed and retentate have been analyzed using the following methods: • chromium(III) — spectrophotometer NANOCOLOR UV/VIS using 1,5-difenylokarbazyde method with wave length λ = 540 nm, • chlorides — the Mohr titration method. The feed solution has been prepared using the following chemicals: CrCl3·6H2O (Sigma-Aldrich), pure NaCl (Chempur®), pure Na2SO4 (Chempur®) and the deionized water. The feed solution was characterized by pH ≈ 4. For initial pH correction the pure HCl (Lachner) was used. The pH was measured by pH-meter (Mettler Toledo SevenEasy). Membrane surface zeta potential was determined by streaming potential using an apparatus and the procedure described in the literature [20]. KCl (Chempur®) solution (0.001 M) was used as the electrolyte solution to measure the streaming potential of nanofiltration membranes. The pH was set by adding NaOH (Chempur®) and HCl (Chempur®). The zeta potential was calculated from the streaming potential using the Helmholtz–Smoluchowski equation taking into account a dielectric constant, viscosity and electrolytic conductivity of the solution. The analysis of membranes surface was determined by scanning electron microscope PHENOM G2 (FEI). For SEM analysis, the deposition of a gold layer of about 2 nm in thickness was done using K550x Sputter Coater (Technologies Quorum). 3. Results and discussion 3.1. The effect of concentrated salt solution on nanofiltration membranes properties In the first stage of the study, the influence of concentrated salt solutions on the surface properties of the tested DL and HL nanofiltration membranes was analyzed. For this purpose, nanofiltration membranes
Fig. 1. Schema of the laboratory plant: 1 — feed/retentate tank, 2 — thermostat, 3 — measurement of temperature, 4 — measurement of pH, 5 — high pressure pump, 6 — manometer, 7 — NF membrane, 8 — permeate tank, 9 — flowmeter, P — permeate, F — feed, and R — retentate.
P. Religa et al. / Desalination 315 (2013) 115–123
117
3.2. The effect of cleaning agents on nanofiltration membranes properties
Table 1 Single-step cleaning. Step
Time (min)
Solution
Rinsing
10
Acidic cleaning
10
Rinsing
20
Deionized water pH = 7 Hydrochloric acid pH = 2 for DL membrane pH = 3 for HL membrane Deionized water pH = 7
were soaked for 20 hours in the concentrated salt solution characterized by low pH. Subsequently, the zeta potentials of the new and soaked, tested membranes in a concentrated salt solution were determined. The obtained results are presented in Fig. 3. In the case of new tested membranes it was found that the HL membrane is characterized by a less negative zeta potential than the DL membrane. In process conditions (pH ≈ 4) the HL membrane demonstrated zeta potential on the level of − 4, whereas the DL membrane was characterized by a zeta potential equal to − 14. Similar results were observed in other works [15,21,22]. According to the examinations conducted by Tanga et al. [17,23] for solution with defined pH the membrane zeta potential depends on the ratio between acidic and basic surface groups. The more acidic groups on the membrane surface the more negative the membrane zeta potential. Hence, analysis of the obtained results (Fig. 3) allows to conclude that the less negative zeta potential of HL than DL membranes is caused by the fact that the HL membrane has a higher density of amine than the carboxyl groups on its active layer. The results show also that the concentrated salt solution characterized by low pH has a significant impact on the change in zeta potential of the tested membranes. Both in the case of DL as well as HL membranes soaked for 20 hours in concentrated salt solutions the change of the zeta potential from negative to positive was observed. The change of the zeta potential of tested membranes caused the formation of the ionic adsorption layer on their surfaces. The long time and high concentrations of ions presence in concentrate salt solution led to the scaling of the tested membranes. It is visible on the SEM images of DL and HL membranes surface (Fig. 4). The results indicate that the membranes used for separation of chromium(III) from concentrated salt solutions characterized by low pH were modified under the influence of these solutions. Because of the deposition of the solution components and scaling the total membrane surface charge (i.e. zeta potential) was changing. Therefore, it is obvious that to perform the efficient process in the studied conditions the membranes must be periodically cleaned. It is necessary to find a cleaning bath, which will enable the removal of scaling from the surface of nanofiltration membranes and regenerate their surface charge.
Table 2 Dual-step cleaning. Step
Time (min)
Solution
Rinsing
10
Acidic cleaning
10
Rinsing
20
Basic cleaning
10
Rinsing
20
Deionized water pH = 7 Hydrochloric acid pH = 2 for DL membrane pH = 3 for HL membrane Deionized water pH = 7 Sodium hydroxide pH = 11 for DL membrane pH = 9 for HL membrane Deionized water pH = 7
3.2.1. Hydrochloric acid The stability of the DL and HL membranes during frequent use for the nanofiltration of concentrated salt solutions characterized by low pH was analyzed. After the process, tested membranes were cleaned with the hydrochloric acid according to the procedure presented in Table 1. The influence of the cleaning bath application prepared on the basis of hydrochloric acid on the surface properties of DL and HL membranes was examined. The obtained results showed a change of the zeta potential of the tested (new and after working at 6 and 20 hours) membranes (Fig. 5). It was observed that the zeta potential of HL membrane received a positive value for the process conditions (pH ≈ 4) after 6 hours of working. The zeta potential determined for the DL membrane was also changed during the nanofiltration of concentrated salt solutions. However, after 20 hours of working it was still negative (pH ≈ 4). The obtained results showed that the low pH of the model solutions and cleaning bath (solutions of hydrochloric acid) caused an increase in the density of positively charged groups on the surface and in the pores of the tested nanofiltration membranes. Such behavior is characteristic for the poly(piperazine-amide) membranes. At pH ≈ 4, the protonation of the amine groups is more pronounced than the dissociation of the carboxyl groups, resulting in a less negative surface charge [16,17,23–25]. According to research conducted by Bandini et al. [26–29] amine functional groups, becoming active at the low pH and being carriers of positive charges, create convenient conditions for the adsorption of anions both on the surface of nanofiltration membranes and in their internal structure (Eqs. (1) and (2)). þ
þ
−
þ
R−NH þ H3 O ↔R−NH2 þ H2 O þ
−
R−NH2 þ Cl ↔R−NH2 Cl
ð1Þ ð2Þ
Hence, when the zeta potential of the membrane was positive, the negative chlorides and sulfates ions present in the model solution were ‘immobilized’ on the surface and inside the membrane. It can result in formation of ionic adsorption layer causing new conditions in the system, including the strengthening effect of concentration polarization and scaling of the NF membranes. This phenomenon has been described in details in our previous work [12]. HL membrane has a higher density of amine than carboxyl groups in its active layer what causes that this type of membrane is more susceptible to the adsorption of negative ions and in consequence showed less stable work under experiment conditions. The observed changes in surface properties of the tested membranes, determined using zeta potential measurements (Fig. 5), explain the results presented in Tables 3 and 4. In the case of the DL membrane characterized by a negative zeta potential, even after 20 hours of working in concentrated salt solution, the retention of chromium(III) remained stable at the level of 96%. This membrane is also characterized by a low rate of chlorides retention equal to 8% and a constant permeate flux on level of 28 × 10 − 6 m 3/(m 2s). The analysis of SEM images of the new DL membrane surface (Fig. 4a) and its surface after 20 hours of working (Fig. 6b) allowed to conclude that the active layer of this membrane was solid, without visible cracks and leakiness, while, in case of the HL membrane, for which the change of the zeta potential from negative on positive was observed, a steady reduction of permeate flux was stated. The positive zeta potential of the membrane caused the high retention of ions present in solution. This led to the increase of concentration polarization of the membrane and the reduction of permeate flux, which in turn adversely affected the efficiency of the process. After 20 hours of working in concentrated salt solution, the stability of the HL membrane was seriously shaken. The analysis of the SEM images (Fig. 6a) showed the structural change of HL membrane surface and mechanical damage of the surface.
118
P. Religa et al. / Desalination 315 (2013) 115–123
Fig. 3. Zeta potential of new and “soaked” DL (a) and HL (b) membrane (after 20 hours in salt mixture solution: 2 gCr3+/dm3, 10 gCl−/dm3, 10 gSO42−/dm3, pH ≈ 4) at various pH, (□ — new DL membrane, ○ — DL membrane after soaking at 20 hours, ◊ — new HL membrane, Δ — HL membrane after soaking at 20 hours).
Fig. 4. SEM images of new DL (a), HL (c) membranes and DL (b), HL (d) membranes after 20 hours in salt mixture solution: 2 gCr3+/dm3, 10 gCl−/dm3, 10 gSO42−/dm3, pH ≈ 4.
P. Religa et al. / Desalination 315 (2013) 115–123
119
Fig. 5. Zeta potential of new and “worked” DL and HL membrane (after working at 6 and 20 hours in salt mixture solution: 2 gCr3+/dm3, 10 gCl−/dm3, 10 gSO42−/dm3, pH≈ 4, TMP = 14 bar and cleaning by HCl) at various pH, (□— new membrane, ▲ — membrane after working at 6 hours, ○ — membrane after working at 20 hours).
According to some researchers [6,8,30] nanofiltration membranes made from polyamide can oxidize and hydrolyze in solutions characterized by a low pH. In case of conducted examinations, concentrated Table 3 Comparison of the permeate flux stability of the DL and HL membranes new and frequently used for the separation of chromium(III) from concentrated salt solutions and cleaned with hydrochloric acid solutions. JP × 106, m3/(m2s)
New After After After After After After
6h 9h 12 h 15 h 18 h 20 h
DL membrane
HL membrane
28 28 28 28 29 28 28
21 19 19 17 17 17 22
salt solutions were characterized by pH ≈ 4. Moreover, the HL membrane after each use was cleaned with a solution of hydrochloric acid of pH = 3. Conditions, in which the membrane was tested, causes the disadvantageous changes in its properties. A further consequence of these changes is mechanical damage in the membrane structure (Fig. 6a), which probably caused the significant increase in permeate flux and decrease of chlorides retention observed for HL membrane after 20 hours of working in concentrated salt solution (Tables 3 and 4). It was found that the DL membrane characterized by a smaller than the HL membrane number of the amine groups in its structure [16,31] is less sensitive to low pH. Therefore it maintains constant negative zeta potential which substantially reduces the scaling and allows for an efficient process. In addition, active three-layer of DL membrane [15,30] ensured its high resistance to chemical and mechanical damage [15,30]. Unfortunately, HL membrane did not have these properties. The obtained results showed that the application
Table 4 Comparison of the stability of the DL and HL membranes new and frequently used for the separation of chromium(III) from concentrated salt solutions and cleaned with hydrochloric acid solutions. Concentration of chromium (g/dm3)
Retention of chromium (%)
Concentration of chloride (g/dm3) Permeate
Retentate
Retention of chloride (%)
Permeate
Retentate
(a) DL membrane New After 6 h After 9 h After 12 h After 15 h After 18 h After 20 h
0.08 0.09 0.09 0.09 0.09 0.09 0.09
2.44 2.44 2.44 2.44 2.44 2.44 2.44
96 96 96 96 96 96 96
8.8 8.8 8.8 8.8 8.8 8.8 8.8
9.6 9.6 9.6 9.6 9.8 9.7 9.6
8 8 8 8 10 9 8
(b) HL membrane New After 6 h After 9 h After 12 h After 15 h After 18 h After 20 h
0.08 0.06 0.06 0.06 0.09 0.09 0.09
2.45 2.47 2.47 2.47 2.44 2.44 2.43
97 97 97 97 96 96 96
7.3 7.4 7.5 7.6 7.6 7.6 8.0
9.9 10.0 10.0 10.2 10.1 10.1 10.2
26 26 25 25 25 25 21
120
P. Religa et al. / Desalination 315 (2013) 115–123
Fig. 6. SEM images of NF membranes type of HL (a) i DL (b) after working at 20 h and cleaning of HCl.
of hydrochloric acid for cleaning of the nanofiltration membranes used for separation of chromium(III) from concentrated salt solutions allows to remove scaling formed on the surface of the tested membranes (Fig. 6b). In addition, the results suggested that the use of hydrochloric acid prevents the regeneration of the surface charge of tested membranes (Fig. 5). Thus, cleaning of the HL membrane according to the proposed procedure does not lead to full recovery of its properties. Such membrane demonstrates the work instability and the susceptibility to mechanical damages. In the case of a DL membrane the cleaning with hydrochloric acid allows for its stable and efficient work for a long time of use. 3.2.2. Hydrochloric acid and sodium hydroxide The new procedure of the membrane cleaning was offered to regenerate the HL membranes surface charge. The membrane should be soaked in alkaline bath after its cleaning in the solution of hydrochloric
acid. The DL and HL membranes after the nanofiltration of concentrated salt solutions characterized by low pH,were cleaned with hydrochloric acid and sodium hydroxide, as proposed in the procedure presented in Table 2. As can be observed, the use of alkaline bath in the form of sodium hydroxide, after cleaning with hydrochloric acid, leads to the recovery of the negative surface charge of nanofiltration membranes (Fig. 7). Analyzing the obtained results it was found that both DL and HL membranes are characterized by a negative zeta potential for the process conditions (pH≈4) even after 20 hours of work in concentrated salt solutions. The obtained results (Tables 5 and 6) stated that the use of alkaline bath in the form of sodium hydroxide after cleaning with hydrochloric acid leads to significant improvement in stability of DL and HL membranes during nanofiltration of concentrated salt solutions characterized by pH ≈ 4. It is especially important for the HL membrane which during the process conditions is characterized by a permeate flux on
Fig. 7. Zeta potential of new and “worked” DL and HL membrane (after working at 6 and 20 hours in salt mixture solution: 2 gCr3+/dm3, 10 gCl−/dm3, 10 gSO42−/dm3, pH≈ 4, TMP = 14 bar and cleaning by HCl and NaOH) at various pH, (□ — new membrane, ▲ — membrane after working at 6 hours, ○ — membrane after working at 20 hours).
P. Religa et al. / Desalination 315 (2013) 115–123 Table 5 Comparison of the permeate flux stability of the DL and HL membranes: new and frequently used for the separation of chromium(III) from concentrated salt solutions and cleaned with hydrochloric acid and sodium hydroxide solutions. JP × 106, m3/(m2s)
New After After After After After After
6h 9h 12 h 15 h 18 h 20 h
DL membrane
HL membrane
28 28 27 27 26 26 27
21 21 22 22 21 22 22
121
Moreover, conditions ruling in the arrangement caused hydrolysis and oxidizing of the tested membranes active layer. These phenomena could have a direct impact on instability of permeability and selectivity of the membrane as well as its mechanical damage. On the other hand the use of alkaline bath in the form of sodium hydroxide leads to the partial recovery of a low surface charge of the membranes. As a consequence significant working stability of the tested NF membranes was noticed. Moreover no mechanical defects in the structure of the examined membranes cleaned by HCl/NaOH bath were observed. 3.3. The effect of cleaning agents on nanofiltration membrane permeabilities
the level of 22× 10 −6 m 3/(m2s). HL membrane also kept the selective properties manifesting with fixed rate of the chlorides retention on the level of 25%. Analyzing SEM images of DL and HL membranes, the lack of mechanical changes in properties of the surface structure was found. The active layer was uniform, without any visible cracks and leaks even in the case of HL membrane (Fig. 8). Summarizing, it was shown that the low pH of the cleaning bath with hydrochloric acid eliminated the membrane scaling but at the same time irreversibly altered the charge of both tested membranes.
The relationship JP = f(ΔP) allows to determine the permeability coefficients of deionized water for both new and repeatedly used DL and HL membranes during the nanofiltration of concentrated salt solutions and cleaning according to the proposed procedures (Tables 1 and 2). The obtained results are summarized in Fig. 9a and b. Analyzing the obtained results for the new tested membranes (Fig. 9a and b) one can find that the HL membrane has a higher permeability coefficient of deionized water than the DL membrane. It is caused by the thinner active layer [12,15,32]. In the case of HL membrane cleaned with hydrochloric acid after nanofiltration of concentrated salt solutions the decrease in permeability (36%) after 18 hours of work in
Table 6 Comparison of the stability of the DL and HL membranes new and frequently used for the separation of chromium(III) from concentrated salt solutions and cleaned with hydrochloric acid and sodium hydroxide solutions. Concentration of chromium (g/dm3)
Retention of chromium (%)
Concentration of chloride (g/dm3) Permeate
Retentate
Retention of chloride (%)
Permeate
Retentate
(a) DL membrane New After 6 h After 9 h After 12 h After 15 h After 18 h After 20 h
0.08 0.06 0.06 0.06 0.06 0.06 0.09
2.44 2.47 2.47 2.47 2,47 2.47 2.47
96 98 98 98 98 98 96
8.8 8.8 8.8 8.8 8.8 8.8 8.8
9.6 9.6 9.6 9.6 9.6 9.6 9.7
8 8 8 8 8 8 9
(a) HL membrane New After 6 h After 9 h After 12 h After 15 h After 18 h After 20 h
0.08 0.06 0.06 0.06 0.06 0.06 0.07
2.45 2.47 2.47 2.44 2.44 2.44 2.43
97 98 98 97 97 97 97
7.3 7.5 7.5 7.5 7.5 7.5 7.5
9.9 10.0 10.0 10.0 10.0 10.0 10.0
26 25 25 25 25 25 25
Fig. 8. SEM images of NF membranes type of HL (a) i DL (b) after working at 20 h and cleaning of HCl and NaOH.
122
P. Religa et al. / Desalination 315 (2013) 115–123
Fig. 9. Correlation between the permeability coefficient of DL (a) and HL (b) membranes and time of working in concentration salt solution (□ — membrane after cleaning of HCl, ■ — membrane after cleaning of HCl and NaOH).
relation to the initial permeability (coefficient appointed for the new membrane) is shown. A sudden increase of permeability coefficient after 20 hours was probably caused by mechanical damage in the structure of the HL membrane surface (Fig. 6a). In the case of applying of sodium hydroxide after cleaning with the hydrochloric acid, the permeability coefficient after 20 hours of work declined only by the 18% towards the initial appointed for the new HL membrane. Moreover, in the case of HL membrane cleaning by HCl and NaOH, a sudden increase of the permeability coefficient which could suggest mechanical damages in the structure of this membrane after several hours use was not observed. Permeability coefficient of the DL membrane repeatedly applied for the nanofiltration of concentrated salt solutions and cleaning according to the single-step cleaning (Table 1) after significant decrease of about 14% in the first period of the work was kept at a steady level during further use. In the case of a DL membrane cleaned according to the dual-step cleaning (Table 2), the stability of permeability coefficient was observed (Fig. 9a). Its means that the membranes cleaning according to this procedure guarantees both for HL and DL membranes stable and higher permeability than in the case of the membranes cleaning with only a hydrochloric acid (Fig. 9a and b). In the case of a new DL and HL membranes cleaned with HCl solution a higher decrease of permeate flux obtained for the deionized water than in the case of integrated cleaning (solutions of HCl and NaOH) was observed. Similar results were observed in ref. [15]. 4. Conclusions The results show that the diversity and high concentrations of ions, for the low pH of tested solutions and cleaning bath in the form of hydrochloric acid, have irreversibly altered the surface charge of the HL membrane. After 20 hours of work in concentrate salt solution and cleaning with hydrochloric acid, the zeta potential of HL membranes received a positive value for pH≈ 4. Moreover, prevailing conditions in the arrangement caused hydrolysis and oxidizing of the HL membrane material. These phenomena can have a direct impact on the mechanical damage found on the surface of the tested membrane. As a result an instability of permeability and selectivity of the HL membrane during nanofiltration of concentrated salt solutions and cleaning with solution
of hydrochloric acid was observed. However, the DL membrane, which after 20 hours of work was still characterized by a negative zeta potential for the pH≈ 4, worked stably and efficiently. The influence of the alkaline bath applied after cleaning with hydrochloric acid on the regeneration properties of DL and HL membranes was examined. Obtained results confirmed that applying of the alkaline bath in the form of the sodium hydroxide after cleaning with solution of the hydrochloric acid can lead to recovering of the low surface charge of HL membranes. It significantly improves both the membrane stability and better resistance to mechanical damage in the structure of the tested membranes.
Acknowledgments This work has been supported by the European Union in the framework of European Social Fund through the Warsaw University of Technology Development Programme, realized by Center for Advanced Studies.
References [1] A.B. Koltuniewicz, E. Drioli, Membrane in clean technology, in: Theory and practice, vol.1-2, Wiley-VCH Verlag GmbH & Co. KGaA, 2008. [2] A.L. Ahmad, B.S. Ooi, A. Wahab Mohammad, J.P. Choudhury, Development of a highly hydrophilic nanofiltration membrane for desalination and water treatment, Desalination 168 (2004) 215–221. [3] S. Ghizellaoui, A. Chibani, S. Ghizellaoui, Use of nanofiltration for partial softening of very hard water, Desalination 179 (2005) 315–322. [4] A. Orecki, M. Tomaszewska, K. Karakulski, A.W. Morawski, Surface water treatment by the nanofiltration method, Desalination 162 (2004) 47–54. [5] C. Bellona, J.E. Drewes, Viability of a low-pressure nanofilter in treating recycled water for water reuse applications: a pilot-scale study, Water Res. 41 (2007) 3948–3958. [6] Y. Ku, S. Chen, W. Wang, Effect of solution composition on the removal of copper ions by nanofiltration, Sep. Purif. Technol. 43 (2005) 135–142. [7] L.M. Ortega, R. Lebrun, J.F. Blais, R. Hausler, Removal of metal ions from an acidic leachate solution by nanofiltration membranes, Desalination 227 (2008) 204–216. [8] Z. Wang, G. Liu, Z. Fan, X. Yang, J. Wang, S. Wang, Experimental study on treatment of electroplating wastewater by nanofiltration, J. Membr. Sci. 305 (2007) 185–195. [9] L.M. Ortega, R. Lebrun, I.M. Noël, R. Hausler, Application of nanofiltration in the recovery of chromium(III) from tannery effluents, Sep. Purif. Technol. 44 (2005) 45–52.
P. Religa et al. / Desalination 315 (2013) 115–123 [10] C. Das, P. Patel, S. De, S. DasGupta, Treatment of tanning effluent using nanofiltration followed by reverse osmosis, Sep. Purif. Technol. 50 (2006) 291–299. [11] S. Gomes, S.A. Cavaco, M.J. Quina, L.M. Gando-Ferreira, Nanofiltration process for separating Cr(III) from acid solutions: experimental and modelling analysis, Desalination 254 (2010) 80–89. [12] P. Religa, A. Kowalik, P. Gierycz, A new approach to chromium concentration from salt mixture solution using nanofiltration, Sep. Purif. Technol. 82 (2011) 114–120. [13] P. Religa, A. Kowalik, P. Gierycz, Application of nanofiltration for chromium concentration in the tannery wastewater, J. Hazard. Mater. 186 (2011) 288–292. [14] P. Religa, A. Kowalik, P. Gierycz, Effect of membrane properties on chromium(III) recirculation from concentrate salt mixture solution by nanofiltration, Desalination 274 (2011) 164–170. [15] A. Al-Amoudi, P. Williams, S. Mandale, R.W. Lovitt, Cleaning results of new andfouled nanofiltration membrane characterized by zeta potential and permeability, Sep. Purif. Technol. 54 (2007) 234–240. [16] M. Mänttäri, A. Pihlajamäki, M. Nyström, Effect of pH on hydrophilicity and charge and their effect on the filtration efficiency of NF membranes at different pH, J. Membr. Sci. 280 (2006) 311–320. [17] C.Y. Tanga, Y.-N. Kwon, J.O. Leckie, Effect of membrane chemistry and coating layer on physicochemical properties of thin film composite polyamide RO and NF membranes. I. FTIR and XPS characterization of polyamide and coating layer chemistry, Desalination 242 (2009) 149–167. [18] M. Pontié, H. Dach, J. Leparc, M. Hafsi, A. Lhassani, Novel approach combining physico-chemical characterizations and mass transfer modelling of nanofiltration and low pressure reverse osmosis membranes for brackish water desalination intensification, Desalination 221 (2008) 174–191. [19] M. Jahanshahi, A. Rahimpour, M. Peyravi, Developing thin film composite poly(piperazine-amide) and poly(vinyl-alcohol) nanofiltration membranes, Desalination 257 (2010) 129–136. [20] Z. Adamczyk, M. Zaucha, M. Zembala, Zeta potential of mica covered by colloid particles: a streaming potential study, Langmuir 26 (12) (2010) 9368–9377.
123
[21] A.I.C. Morao, A.M.B. Alves, M.D. Afonso, Concentration of clavulanic acid broths: influence of the membrane surface charge density on NF operation, J. Membr. Sci. 281 (2006) 417–428. [22] K. Boussu, C. Vandecasteele, B. Van der Bruggen, Relation between membrane characteristic and performance in nanofiltration, J. Membr. Sci. 310 (2008) 51–65. [23] C.Y. Tanga, Y.-N. Y-N. Kwon, J.O. Leckie, Effect of membrane chemistry and coating layer on physicochemical properties of thin film composite polyamide RO and NF membranes. II. Membrane physiochemical properties and their dependence on polyamide and coating layers, Desalination 242 (2009) 168–182. [24] C. Zhao, S. Nie, M. Tang, S. Sun, Polymeric pH-sensitive membranes — a review, Prog. Polym. Sci. 36 (2011) 1499–1520. [25] V. Freger, T.C. Arnot, J.A. Howell, Separation of concentrated organic/inorganic salt mixture by nanofiltration, J. Membr. Sci. 178 (2000) 185–193. [26] S. Bandini, Modelling the mechanism of charge formation in NF membranes: theory and application, J. Membr. Sci. 264 (2005) 75–86. [27] L. Bruni, S. Bandini, The role of the electrolyte on the mechanism of charge formation in polyamide nanofiltration membranes, J. Membr. Sci. 308 (2008) 136–151. [28] A. Szymczyk, P. Fievet, S. Bandini, On the amphoteric behaviour of Desal DK nanofiltration membranes at low salt concentration, J. Membr. Sci. 533 (2010) 60–68. [29] S. Bandini, C. Mazzoni, Modelling the amphoteric behaviour of polyamide nanofiltration membranes, Desalination 184 (2005) 327–336. [30] S. Platt, M. Nyström, A. Bottino, G. Capannelli, Stability of NF membranes under extreme acidic conditions, J. Membr. Sci. 239 (2004) 91–103. [31] A.E. Childress, M. Elimelech, Nanofiltration membrane performance to membrane charge (electrokinetic) characteristics, Environ. Sci. Technol. 34 (2000) 3710–3716. [32] A. Al-Amoundi, P. Williams, A.S. Al-Hobaib, R.W. Lovitt, Cleaning results of new and fouled nanofiltration membrane characterized by contact angle, updated DSPM, flux and salts rejection, Appl. Surf. Sci. 254 (2008) 3983–3992.
Contents lists available at SciVerse ScienceDirect
Desalination journal homepage: www.elsevier.com/locate/desal
Study on the behavior of nanofiltration membranes using for chromium(III) recovery from salt mixture solution P. Religa a,⁎, A. Kowalik-Klimczak b, P. Gierycz b, c a b c
Department of Environmental Protection, Radom University of Technology, Chrobrego 27, Radom, Poland Faculty of Chemical and Process Engineering, Warsaw University of Technology, Waryńskiego 1, Warsaw, Poland Institute of Physical Chemistry Polish Academy of Sciences, Kasprzaka 44/52, Warsaw, Poland
H I G H L I G H T S ► ► ► ► ►
Low pH and high salt concentration in chromium solution modified NF membrane surface Two washing procedures to reproduce the properties of NF membranes was analyzed Potential zeta, SEM and permeability of membranes was determined Membrane bathing with HCl reduced its scaling but not the surface charge reproduced Mixed HCl–NaOH bath leads both reduce of membrane scaling and recovery its charge
a r t i c l e
i n f o
Article history: Received 13 August 2012 Received in revised form 26 October 2012 Accepted 27 October 2012 Available online 20 November 2012 Keywords: Nanofiltration (NF) Chromium(III) recovery Membrane charge Permeability Cleaning agents
a b s t r a c t Results of studies of the effects of concentrate salt solutions characterized by low pH, on the nanofiltration membrane surface properties used for the separation of chromium(III) have been presented in this paper. It was shown that the low pH of the concentrate salt solutions and cleaning bath with hydrochloric acid irreversibly altered the charge of tested membranes. As the consequence an instability of permeability and selectivity of the membrane during the process was noticed. The effect of alkaline bath used after cleaning with a solution of hydrochloric acid to regenerate the surface charge of tested membranes was also examined. The results showed that the use of bath in the form of sodium hydroxide leads to the partial recovery of a low surface charge of the membranes. In addition a significant improvement in the stability of the tested NF membranes used for the separation of chromium(III) from concentrate salt solutions at low pH was observed. Moreover, SEM images obtained for the tested membranes cleaned with solutions of HCl and NaOH indicated no mechanical defects in the structure of the examined membranes. © 2012 Elsevier B.V. All rights reserved.
1. Introduction Nanofiltration is a pressure driven process, which assures great process efficiency, as well as enables its great selectivity [1]. On account of these advantages, the nanofiltration is successfully applied to the treatment of surface and underground waters, above all for their softening [2–5] and for the separation of metal ions from industrial wastewaters [6–8]. Several other studies [9–11] as well as our previous works [12–14] demonstrate very interesting possibilities of nanofiltration application for chromium(III) recovery from concentrated salts mixture characterized by low pH. Such solutions are used, among others, in a tanning industry. Separation of chromium(III) from concentrated salts mixture depends on the feed composition [12,13] and properties of the used nanofiltration membranes [14]. The presence of mono- and multivalent negative ions and sufficiently high number of ⁎ Corresponding author. Tel.: +48 48 3617583; fax: +48 48 3617598. E-mail address: [email protected] (P. Religa). 0011-9164/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.desal.2012.10.036
monovalent positive ions produces a Donnan phenomenon which can be additionally increased by the selection of the nanofiltration membrane demonstrating the proper chemical-physical capacity of surface layer. The NF membrane characterized by a negative zeta potential [14] is especially useful for chromium(III) recovery from concentrated salts mixture characterized by low pH. The use of such membrane lowers its polarization causing the increase of permeate flux, and at the same time growth of chromium(III) concentration factor. However, even at optimum feed composition and the use of membrane with appropriate surface properties decrease of its productivities during the nanofiltration process was observed [12]. It was caused by the change of membrane zeta potential by the low pH. The consequence of this fact is the adsorption of ions on the surface and/or in the pores and the deterioration of the membranes selectivity [12]. The change in the membranes selectivity causes changes in the retention of feed components, which may lead to the membrane polarization increase and as a consequence to reduction of its permeability. Because the maintenance of high membrane permeability is preferred for chromium(III)
116
P. Religa et al. / Desalination 315 (2013) 115–123
recovery, it seems to be necessary to remove the adsorbed ions layer from the surface and/or NF membrane pores to keep low values of zeta potential. The aim of this study was to analyze the changes of surface properties of nanofiltration membranes used for chromium(III) recovery from concentrated salt solutions at low pH and to examine the different washing procedures to reproduce the properties of the tested nanofiltration membranes — characteristic zeta potential, as well as their hydraulic permeability. 2. Experimental The experiments were carried out at laboratory scale in cross flow cell made of stainless steel operated in batch mode with circulation presented in Fig. 1. The nanofiltration of model concentrated salt solutions containing 2 gCr 3 +/dm 3, 10 gCl −/dm 3, 10 gSO42 −/dm 3 and characterized by a pH ≈ 4 was conducted for transmembrane pressure on the level of 14 bar and retentate flow equal to 800 dm 3/h. The composition of solution and operation conditions were fixed based on our previous studies [12–14]. The temperature of feed solution during the process was maintained at 25 ± 1 °C by a thermostat. The feed solution was pumped from the feed tank toward to nanofiltration membrane, obtaining a retentate that was returned to the feed/retentate tank and a permeate that was collected in the permeate tank. Two kinds of commercial nanofiltration flat sheet membranes (under symbol DL and HL) provided by GE Osmonics were used in the experiments. The tested nanofiltration membranes had an active layer made of the poly(piperazine-amide) (Fig. 2). The isoelectric point (IP) was less than 3.0 and 3.3 for DL and HL membrane, respectively [14–16]. Such membrane IP indicated on its negative zeta potential under experimental pH. The DL membrane had active three-layer — dense membrane structure, while the HL membrane had active two-layer — loose membrane structure. Each of these membranes has an effective area of 0.0155 m2. After nanofiltration the tested membranes were cleaned with acid and alkaline baths according with procedures presented in Tables 1 and 2. pH of cleaning baths was chosen on the basis of the tested membranes characteristics regarding their chemical resistance [14]. Temperature of the acidic and alkaline cleaning bath was constant and equal to 18 ± 1 °C. The changing of membrane permeability under the influence of concentrated salt solutions and the cleaning according to proposed procedures was analyzed based on the JP = f(ΔP) dependence designated for the deionized water (TMP = 10–24 bar, QR = 800 dm 3/h, t = 25 ± 1 °C) (Tables 1 and 2). Samples of permeate and retentate have been collected for determination of chromium(III) concentration in well-defined time intervals.
O
O
C
C
N
NH
COOH Fig. 2. Poly(piperazine-amide) formula [17–19].
After the end of the experiment, samples of permeate and retentate have been collected for determination of the chloride concentration. The samples of permeate, feed and retentate have been analyzed using the following methods: • chromium(III) — spectrophotometer NANOCOLOR UV/VIS using 1,5-difenylokarbazyde method with wave length λ = 540 nm, • chlorides — the Mohr titration method. The feed solution has been prepared using the following chemicals: CrCl3·6H2O (Sigma-Aldrich), pure NaCl (Chempur®), pure Na2SO4 (Chempur®) and the deionized water. The feed solution was characterized by pH ≈ 4. For initial pH correction the pure HCl (Lachner) was used. The pH was measured by pH-meter (Mettler Toledo SevenEasy). Membrane surface zeta potential was determined by streaming potential using an apparatus and the procedure described in the literature [20]. KCl (Chempur®) solution (0.001 M) was used as the electrolyte solution to measure the streaming potential of nanofiltration membranes. The pH was set by adding NaOH (Chempur®) and HCl (Chempur®). The zeta potential was calculated from the streaming potential using the Helmholtz–Smoluchowski equation taking into account a dielectric constant, viscosity and electrolytic conductivity of the solution. The analysis of membranes surface was determined by scanning electron microscope PHENOM G2 (FEI). For SEM analysis, the deposition of a gold layer of about 2 nm in thickness was done using K550x Sputter Coater (Technologies Quorum). 3. Results and discussion 3.1. The effect of concentrated salt solution on nanofiltration membranes properties In the first stage of the study, the influence of concentrated salt solutions on the surface properties of the tested DL and HL nanofiltration membranes was analyzed. For this purpose, nanofiltration membranes
Fig. 1. Schema of the laboratory plant: 1 — feed/retentate tank, 2 — thermostat, 3 — measurement of temperature, 4 — measurement of pH, 5 — high pressure pump, 6 — manometer, 7 — NF membrane, 8 — permeate tank, 9 — flowmeter, P — permeate, F — feed, and R — retentate.
P. Religa et al. / Desalination 315 (2013) 115–123
117
3.2. The effect of cleaning agents on nanofiltration membranes properties
Table 1 Single-step cleaning. Step
Time (min)
Solution
Rinsing
10
Acidic cleaning
10
Rinsing
20
Deionized water pH = 7 Hydrochloric acid pH = 2 for DL membrane pH = 3 for HL membrane Deionized water pH = 7
were soaked for 20 hours in the concentrated salt solution characterized by low pH. Subsequently, the zeta potentials of the new and soaked, tested membranes in a concentrated salt solution were determined. The obtained results are presented in Fig. 3. In the case of new tested membranes it was found that the HL membrane is characterized by a less negative zeta potential than the DL membrane. In process conditions (pH ≈ 4) the HL membrane demonstrated zeta potential on the level of − 4, whereas the DL membrane was characterized by a zeta potential equal to − 14. Similar results were observed in other works [15,21,22]. According to the examinations conducted by Tanga et al. [17,23] for solution with defined pH the membrane zeta potential depends on the ratio between acidic and basic surface groups. The more acidic groups on the membrane surface the more negative the membrane zeta potential. Hence, analysis of the obtained results (Fig. 3) allows to conclude that the less negative zeta potential of HL than DL membranes is caused by the fact that the HL membrane has a higher density of amine than the carboxyl groups on its active layer. The results show also that the concentrated salt solution characterized by low pH has a significant impact on the change in zeta potential of the tested membranes. Both in the case of DL as well as HL membranes soaked for 20 hours in concentrated salt solutions the change of the zeta potential from negative to positive was observed. The change of the zeta potential of tested membranes caused the formation of the ionic adsorption layer on their surfaces. The long time and high concentrations of ions presence in concentrate salt solution led to the scaling of the tested membranes. It is visible on the SEM images of DL and HL membranes surface (Fig. 4). The results indicate that the membranes used for separation of chromium(III) from concentrated salt solutions characterized by low pH were modified under the influence of these solutions. Because of the deposition of the solution components and scaling the total membrane surface charge (i.e. zeta potential) was changing. Therefore, it is obvious that to perform the efficient process in the studied conditions the membranes must be periodically cleaned. It is necessary to find a cleaning bath, which will enable the removal of scaling from the surface of nanofiltration membranes and regenerate their surface charge.
Table 2 Dual-step cleaning. Step
Time (min)
Solution
Rinsing
10
Acidic cleaning
10
Rinsing
20
Basic cleaning
10
Rinsing
20
Deionized water pH = 7 Hydrochloric acid pH = 2 for DL membrane pH = 3 for HL membrane Deionized water pH = 7 Sodium hydroxide pH = 11 for DL membrane pH = 9 for HL membrane Deionized water pH = 7
3.2.1. Hydrochloric acid The stability of the DL and HL membranes during frequent use for the nanofiltration of concentrated salt solutions characterized by low pH was analyzed. After the process, tested membranes were cleaned with the hydrochloric acid according to the procedure presented in Table 1. The influence of the cleaning bath application prepared on the basis of hydrochloric acid on the surface properties of DL and HL membranes was examined. The obtained results showed a change of the zeta potential of the tested (new and after working at 6 and 20 hours) membranes (Fig. 5). It was observed that the zeta potential of HL membrane received a positive value for the process conditions (pH ≈ 4) after 6 hours of working. The zeta potential determined for the DL membrane was also changed during the nanofiltration of concentrated salt solutions. However, after 20 hours of working it was still negative (pH ≈ 4). The obtained results showed that the low pH of the model solutions and cleaning bath (solutions of hydrochloric acid) caused an increase in the density of positively charged groups on the surface and in the pores of the tested nanofiltration membranes. Such behavior is characteristic for the poly(piperazine-amide) membranes. At pH ≈ 4, the protonation of the amine groups is more pronounced than the dissociation of the carboxyl groups, resulting in a less negative surface charge [16,17,23–25]. According to research conducted by Bandini et al. [26–29] amine functional groups, becoming active at the low pH and being carriers of positive charges, create convenient conditions for the adsorption of anions both on the surface of nanofiltration membranes and in their internal structure (Eqs. (1) and (2)). þ
þ
−
þ
R−NH þ H3 O ↔R−NH2 þ H2 O þ
−
R−NH2 þ Cl ↔R−NH2 Cl
ð1Þ ð2Þ
Hence, when the zeta potential of the membrane was positive, the negative chlorides and sulfates ions present in the model solution were ‘immobilized’ on the surface and inside the membrane. It can result in formation of ionic adsorption layer causing new conditions in the system, including the strengthening effect of concentration polarization and scaling of the NF membranes. This phenomenon has been described in details in our previous work [12]. HL membrane has a higher density of amine than carboxyl groups in its active layer what causes that this type of membrane is more susceptible to the adsorption of negative ions and in consequence showed less stable work under experiment conditions. The observed changes in surface properties of the tested membranes, determined using zeta potential measurements (Fig. 5), explain the results presented in Tables 3 and 4. In the case of the DL membrane characterized by a negative zeta potential, even after 20 hours of working in concentrated salt solution, the retention of chromium(III) remained stable at the level of 96%. This membrane is also characterized by a low rate of chlorides retention equal to 8% and a constant permeate flux on level of 28 × 10 − 6 m 3/(m 2s). The analysis of SEM images of the new DL membrane surface (Fig. 4a) and its surface after 20 hours of working (Fig. 6b) allowed to conclude that the active layer of this membrane was solid, without visible cracks and leakiness, while, in case of the HL membrane, for which the change of the zeta potential from negative on positive was observed, a steady reduction of permeate flux was stated. The positive zeta potential of the membrane caused the high retention of ions present in solution. This led to the increase of concentration polarization of the membrane and the reduction of permeate flux, which in turn adversely affected the efficiency of the process. After 20 hours of working in concentrated salt solution, the stability of the HL membrane was seriously shaken. The analysis of the SEM images (Fig. 6a) showed the structural change of HL membrane surface and mechanical damage of the surface.
118
P. Religa et al. / Desalination 315 (2013) 115–123
Fig. 3. Zeta potential of new and “soaked” DL (a) and HL (b) membrane (after 20 hours in salt mixture solution: 2 gCr3+/dm3, 10 gCl−/dm3, 10 gSO42−/dm3, pH ≈ 4) at various pH, (□ — new DL membrane, ○ — DL membrane after soaking at 20 hours, ◊ — new HL membrane, Δ — HL membrane after soaking at 20 hours).
Fig. 4. SEM images of new DL (a), HL (c) membranes and DL (b), HL (d) membranes after 20 hours in salt mixture solution: 2 gCr3+/dm3, 10 gCl−/dm3, 10 gSO42−/dm3, pH ≈ 4.
P. Religa et al. / Desalination 315 (2013) 115–123
119
Fig. 5. Zeta potential of new and “worked” DL and HL membrane (after working at 6 and 20 hours in salt mixture solution: 2 gCr3+/dm3, 10 gCl−/dm3, 10 gSO42−/dm3, pH≈ 4, TMP = 14 bar and cleaning by HCl) at various pH, (□— new membrane, ▲ — membrane after working at 6 hours, ○ — membrane after working at 20 hours).
According to some researchers [6,8,30] nanofiltration membranes made from polyamide can oxidize and hydrolyze in solutions characterized by a low pH. In case of conducted examinations, concentrated Table 3 Comparison of the permeate flux stability of the DL and HL membranes new and frequently used for the separation of chromium(III) from concentrated salt solutions and cleaned with hydrochloric acid solutions. JP × 106, m3/(m2s)
New After After After After After After
6h 9h 12 h 15 h 18 h 20 h
DL membrane
HL membrane
28 28 28 28 29 28 28
21 19 19 17 17 17 22
salt solutions were characterized by pH ≈ 4. Moreover, the HL membrane after each use was cleaned with a solution of hydrochloric acid of pH = 3. Conditions, in which the membrane was tested, causes the disadvantageous changes in its properties. A further consequence of these changes is mechanical damage in the membrane structure (Fig. 6a), which probably caused the significant increase in permeate flux and decrease of chlorides retention observed for HL membrane after 20 hours of working in concentrated salt solution (Tables 3 and 4). It was found that the DL membrane characterized by a smaller than the HL membrane number of the amine groups in its structure [16,31] is less sensitive to low pH. Therefore it maintains constant negative zeta potential which substantially reduces the scaling and allows for an efficient process. In addition, active three-layer of DL membrane [15,30] ensured its high resistance to chemical and mechanical damage [15,30]. Unfortunately, HL membrane did not have these properties. The obtained results showed that the application
Table 4 Comparison of the stability of the DL and HL membranes new and frequently used for the separation of chromium(III) from concentrated salt solutions and cleaned with hydrochloric acid solutions. Concentration of chromium (g/dm3)
Retention of chromium (%)
Concentration of chloride (g/dm3) Permeate
Retentate
Retention of chloride (%)
Permeate
Retentate
(a) DL membrane New After 6 h After 9 h After 12 h After 15 h After 18 h After 20 h
0.08 0.09 0.09 0.09 0.09 0.09 0.09
2.44 2.44 2.44 2.44 2.44 2.44 2.44
96 96 96 96 96 96 96
8.8 8.8 8.8 8.8 8.8 8.8 8.8
9.6 9.6 9.6 9.6 9.8 9.7 9.6
8 8 8 8 10 9 8
(b) HL membrane New After 6 h After 9 h After 12 h After 15 h After 18 h After 20 h
0.08 0.06 0.06 0.06 0.09 0.09 0.09
2.45 2.47 2.47 2.47 2.44 2.44 2.43
97 97 97 97 96 96 96
7.3 7.4 7.5 7.6 7.6 7.6 8.0
9.9 10.0 10.0 10.2 10.1 10.1 10.2
26 26 25 25 25 25 21
120
P. Religa et al. / Desalination 315 (2013) 115–123
Fig. 6. SEM images of NF membranes type of HL (a) i DL (b) after working at 20 h and cleaning of HCl.
of hydrochloric acid for cleaning of the nanofiltration membranes used for separation of chromium(III) from concentrated salt solutions allows to remove scaling formed on the surface of the tested membranes (Fig. 6b). In addition, the results suggested that the use of hydrochloric acid prevents the regeneration of the surface charge of tested membranes (Fig. 5). Thus, cleaning of the HL membrane according to the proposed procedure does not lead to full recovery of its properties. Such membrane demonstrates the work instability and the susceptibility to mechanical damages. In the case of a DL membrane the cleaning with hydrochloric acid allows for its stable and efficient work for a long time of use. 3.2.2. Hydrochloric acid and sodium hydroxide The new procedure of the membrane cleaning was offered to regenerate the HL membranes surface charge. The membrane should be soaked in alkaline bath after its cleaning in the solution of hydrochloric
acid. The DL and HL membranes after the nanofiltration of concentrated salt solutions characterized by low pH,were cleaned with hydrochloric acid and sodium hydroxide, as proposed in the procedure presented in Table 2. As can be observed, the use of alkaline bath in the form of sodium hydroxide, after cleaning with hydrochloric acid, leads to the recovery of the negative surface charge of nanofiltration membranes (Fig. 7). Analyzing the obtained results it was found that both DL and HL membranes are characterized by a negative zeta potential for the process conditions (pH≈4) even after 20 hours of work in concentrated salt solutions. The obtained results (Tables 5 and 6) stated that the use of alkaline bath in the form of sodium hydroxide after cleaning with hydrochloric acid leads to significant improvement in stability of DL and HL membranes during nanofiltration of concentrated salt solutions characterized by pH ≈ 4. It is especially important for the HL membrane which during the process conditions is characterized by a permeate flux on
Fig. 7. Zeta potential of new and “worked” DL and HL membrane (after working at 6 and 20 hours in salt mixture solution: 2 gCr3+/dm3, 10 gCl−/dm3, 10 gSO42−/dm3, pH≈ 4, TMP = 14 bar and cleaning by HCl and NaOH) at various pH, (□ — new membrane, ▲ — membrane after working at 6 hours, ○ — membrane after working at 20 hours).
P. Religa et al. / Desalination 315 (2013) 115–123 Table 5 Comparison of the permeate flux stability of the DL and HL membranes: new and frequently used for the separation of chromium(III) from concentrated salt solutions and cleaned with hydrochloric acid and sodium hydroxide solutions. JP × 106, m3/(m2s)
New After After After After After After
6h 9h 12 h 15 h 18 h 20 h
DL membrane
HL membrane
28 28 27 27 26 26 27
21 21 22 22 21 22 22
121
Moreover, conditions ruling in the arrangement caused hydrolysis and oxidizing of the tested membranes active layer. These phenomena could have a direct impact on instability of permeability and selectivity of the membrane as well as its mechanical damage. On the other hand the use of alkaline bath in the form of sodium hydroxide leads to the partial recovery of a low surface charge of the membranes. As a consequence significant working stability of the tested NF membranes was noticed. Moreover no mechanical defects in the structure of the examined membranes cleaned by HCl/NaOH bath were observed. 3.3. The effect of cleaning agents on nanofiltration membrane permeabilities
the level of 22× 10 −6 m 3/(m2s). HL membrane also kept the selective properties manifesting with fixed rate of the chlorides retention on the level of 25%. Analyzing SEM images of DL and HL membranes, the lack of mechanical changes in properties of the surface structure was found. The active layer was uniform, without any visible cracks and leaks even in the case of HL membrane (Fig. 8). Summarizing, it was shown that the low pH of the cleaning bath with hydrochloric acid eliminated the membrane scaling but at the same time irreversibly altered the charge of both tested membranes.
The relationship JP = f(ΔP) allows to determine the permeability coefficients of deionized water for both new and repeatedly used DL and HL membranes during the nanofiltration of concentrated salt solutions and cleaning according to the proposed procedures (Tables 1 and 2). The obtained results are summarized in Fig. 9a and b. Analyzing the obtained results for the new tested membranes (Fig. 9a and b) one can find that the HL membrane has a higher permeability coefficient of deionized water than the DL membrane. It is caused by the thinner active layer [12,15,32]. In the case of HL membrane cleaned with hydrochloric acid after nanofiltration of concentrated salt solutions the decrease in permeability (36%) after 18 hours of work in
Table 6 Comparison of the stability of the DL and HL membranes new and frequently used for the separation of chromium(III) from concentrated salt solutions and cleaned with hydrochloric acid and sodium hydroxide solutions. Concentration of chromium (g/dm3)
Retention of chromium (%)
Concentration of chloride (g/dm3) Permeate
Retentate
Retention of chloride (%)
Permeate
Retentate
(a) DL membrane New After 6 h After 9 h After 12 h After 15 h After 18 h After 20 h
0.08 0.06 0.06 0.06 0.06 0.06 0.09
2.44 2.47 2.47 2.47 2,47 2.47 2.47
96 98 98 98 98 98 96
8.8 8.8 8.8 8.8 8.8 8.8 8.8
9.6 9.6 9.6 9.6 9.6 9.6 9.7
8 8 8 8 8 8 9
(a) HL membrane New After 6 h After 9 h After 12 h After 15 h After 18 h After 20 h
0.08 0.06 0.06 0.06 0.06 0.06 0.07
2.45 2.47 2.47 2.44 2.44 2.44 2.43
97 98 98 97 97 97 97
7.3 7.5 7.5 7.5 7.5 7.5 7.5
9.9 10.0 10.0 10.0 10.0 10.0 10.0
26 25 25 25 25 25 25
Fig. 8. SEM images of NF membranes type of HL (a) i DL (b) after working at 20 h and cleaning of HCl and NaOH.
122
P. Religa et al. / Desalination 315 (2013) 115–123
Fig. 9. Correlation between the permeability coefficient of DL (a) and HL (b) membranes and time of working in concentration salt solution (□ — membrane after cleaning of HCl, ■ — membrane after cleaning of HCl and NaOH).
relation to the initial permeability (coefficient appointed for the new membrane) is shown. A sudden increase of permeability coefficient after 20 hours was probably caused by mechanical damage in the structure of the HL membrane surface (Fig. 6a). In the case of applying of sodium hydroxide after cleaning with the hydrochloric acid, the permeability coefficient after 20 hours of work declined only by the 18% towards the initial appointed for the new HL membrane. Moreover, in the case of HL membrane cleaning by HCl and NaOH, a sudden increase of the permeability coefficient which could suggest mechanical damages in the structure of this membrane after several hours use was not observed. Permeability coefficient of the DL membrane repeatedly applied for the nanofiltration of concentrated salt solutions and cleaning according to the single-step cleaning (Table 1) after significant decrease of about 14% in the first period of the work was kept at a steady level during further use. In the case of a DL membrane cleaned according to the dual-step cleaning (Table 2), the stability of permeability coefficient was observed (Fig. 9a). Its means that the membranes cleaning according to this procedure guarantees both for HL and DL membranes stable and higher permeability than in the case of the membranes cleaning with only a hydrochloric acid (Fig. 9a and b). In the case of a new DL and HL membranes cleaned with HCl solution a higher decrease of permeate flux obtained for the deionized water than in the case of integrated cleaning (solutions of HCl and NaOH) was observed. Similar results were observed in ref. [15]. 4. Conclusions The results show that the diversity and high concentrations of ions, for the low pH of tested solutions and cleaning bath in the form of hydrochloric acid, have irreversibly altered the surface charge of the HL membrane. After 20 hours of work in concentrate salt solution and cleaning with hydrochloric acid, the zeta potential of HL membranes received a positive value for pH≈ 4. Moreover, prevailing conditions in the arrangement caused hydrolysis and oxidizing of the HL membrane material. These phenomena can have a direct impact on the mechanical damage found on the surface of the tested membrane. As a result an instability of permeability and selectivity of the HL membrane during nanofiltration of concentrated salt solutions and cleaning with solution
of hydrochloric acid was observed. However, the DL membrane, which after 20 hours of work was still characterized by a negative zeta potential for the pH≈ 4, worked stably and efficiently. The influence of the alkaline bath applied after cleaning with hydrochloric acid on the regeneration properties of DL and HL membranes was examined. Obtained results confirmed that applying of the alkaline bath in the form of the sodium hydroxide after cleaning with solution of the hydrochloric acid can lead to recovering of the low surface charge of HL membranes. It significantly improves both the membrane stability and better resistance to mechanical damage in the structure of the tested membranes.
Acknowledgments This work has been supported by the European Union in the framework of European Social Fund through the Warsaw University of Technology Development Programme, realized by Center for Advanced Studies.
References [1] A.B. Koltuniewicz, E. Drioli, Membrane in clean technology, in: Theory and practice, vol.1-2, Wiley-VCH Verlag GmbH & Co. KGaA, 2008. [2] A.L. Ahmad, B.S. Ooi, A. Wahab Mohammad, J.P. Choudhury, Development of a highly hydrophilic nanofiltration membrane for desalination and water treatment, Desalination 168 (2004) 215–221. [3] S. Ghizellaoui, A. Chibani, S. Ghizellaoui, Use of nanofiltration for partial softening of very hard water, Desalination 179 (2005) 315–322. [4] A. Orecki, M. Tomaszewska, K. Karakulski, A.W. Morawski, Surface water treatment by the nanofiltration method, Desalination 162 (2004) 47–54. [5] C. Bellona, J.E. Drewes, Viability of a low-pressure nanofilter in treating recycled water for water reuse applications: a pilot-scale study, Water Res. 41 (2007) 3948–3958. [6] Y. Ku, S. Chen, W. Wang, Effect of solution composition on the removal of copper ions by nanofiltration, Sep. Purif. Technol. 43 (2005) 135–142. [7] L.M. Ortega, R. Lebrun, J.F. Blais, R. Hausler, Removal of metal ions from an acidic leachate solution by nanofiltration membranes, Desalination 227 (2008) 204–216. [8] Z. Wang, G. Liu, Z. Fan, X. Yang, J. Wang, S. Wang, Experimental study on treatment of electroplating wastewater by nanofiltration, J. Membr. Sci. 305 (2007) 185–195. [9] L.M. Ortega, R. Lebrun, I.M. Noël, R. Hausler, Application of nanofiltration in the recovery of chromium(III) from tannery effluents, Sep. Purif. Technol. 44 (2005) 45–52.
P. Religa et al. / Desalination 315 (2013) 115–123 [10] C. Das, P. Patel, S. De, S. DasGupta, Treatment of tanning effluent using nanofiltration followed by reverse osmosis, Sep. Purif. Technol. 50 (2006) 291–299. [11] S. Gomes, S.A. Cavaco, M.J. Quina, L.M. Gando-Ferreira, Nanofiltration process for separating Cr(III) from acid solutions: experimental and modelling analysis, Desalination 254 (2010) 80–89. [12] P. Religa, A. Kowalik, P. Gierycz, A new approach to chromium concentration from salt mixture solution using nanofiltration, Sep. Purif. Technol. 82 (2011) 114–120. [13] P. Religa, A. Kowalik, P. Gierycz, Application of nanofiltration for chromium concentration in the tannery wastewater, J. Hazard. Mater. 186 (2011) 288–292. [14] P. Religa, A. Kowalik, P. Gierycz, Effect of membrane properties on chromium(III) recirculation from concentrate salt mixture solution by nanofiltration, Desalination 274 (2011) 164–170. [15] A. Al-Amoudi, P. Williams, S. Mandale, R.W. Lovitt, Cleaning results of new andfouled nanofiltration membrane characterized by zeta potential and permeability, Sep. Purif. Technol. 54 (2007) 234–240. [16] M. Mänttäri, A. Pihlajamäki, M. Nyström, Effect of pH on hydrophilicity and charge and their effect on the filtration efficiency of NF membranes at different pH, J. Membr. Sci. 280 (2006) 311–320. [17] C.Y. Tanga, Y.-N. Kwon, J.O. Leckie, Effect of membrane chemistry and coating layer on physicochemical properties of thin film composite polyamide RO and NF membranes. I. FTIR and XPS characterization of polyamide and coating layer chemistry, Desalination 242 (2009) 149–167. [18] M. Pontié, H. Dach, J. Leparc, M. Hafsi, A. Lhassani, Novel approach combining physico-chemical characterizations and mass transfer modelling of nanofiltration and low pressure reverse osmosis membranes for brackish water desalination intensification, Desalination 221 (2008) 174–191. [19] M. Jahanshahi, A. Rahimpour, M. Peyravi, Developing thin film composite poly(piperazine-amide) and poly(vinyl-alcohol) nanofiltration membranes, Desalination 257 (2010) 129–136. [20] Z. Adamczyk, M. Zaucha, M. Zembala, Zeta potential of mica covered by colloid particles: a streaming potential study, Langmuir 26 (12) (2010) 9368–9377.
123
[21] A.I.C. Morao, A.M.B. Alves, M.D. Afonso, Concentration of clavulanic acid broths: influence of the membrane surface charge density on NF operation, J. Membr. Sci. 281 (2006) 417–428. [22] K. Boussu, C. Vandecasteele, B. Van der Bruggen, Relation between membrane characteristic and performance in nanofiltration, J. Membr. Sci. 310 (2008) 51–65. [23] C.Y. Tanga, Y.-N. Y-N. Kwon, J.O. Leckie, Effect of membrane chemistry and coating layer on physicochemical properties of thin film composite polyamide RO and NF membranes. II. Membrane physiochemical properties and their dependence on polyamide and coating layers, Desalination 242 (2009) 168–182. [24] C. Zhao, S. Nie, M. Tang, S. Sun, Polymeric pH-sensitive membranes — a review, Prog. Polym. Sci. 36 (2011) 1499–1520. [25] V. Freger, T.C. Arnot, J.A. Howell, Separation of concentrated organic/inorganic salt mixture by nanofiltration, J. Membr. Sci. 178 (2000) 185–193. [26] S. Bandini, Modelling the mechanism of charge formation in NF membranes: theory and application, J. Membr. Sci. 264 (2005) 75–86. [27] L. Bruni, S. Bandini, The role of the electrolyte on the mechanism of charge formation in polyamide nanofiltration membranes, J. Membr. Sci. 308 (2008) 136–151. [28] A. Szymczyk, P. Fievet, S. Bandini, On the amphoteric behaviour of Desal DK nanofiltration membranes at low salt concentration, J. Membr. Sci. 533 (2010) 60–68. [29] S. Bandini, C. Mazzoni, Modelling the amphoteric behaviour of polyamide nanofiltration membranes, Desalination 184 (2005) 327–336. [30] S. Platt, M. Nyström, A. Bottino, G. Capannelli, Stability of NF membranes under extreme acidic conditions, J. Membr. Sci. 239 (2004) 91–103. [31] A.E. Childress, M. Elimelech, Nanofiltration membrane performance to membrane charge (electrokinetic) characteristics, Environ. Sci. Technol. 34 (2000) 3710–3716. [32] A. Al-Amoundi, P. Williams, A.S. Al-Hobaib, R.W. Lovitt, Cleaning results of new and fouled nanofiltration membrane characterized by contact angle, updated DSPM, flux and salts rejection, Appl. Surf. Sci. 254 (2008) 3983–3992.