当前位置:首页 >> IT/计算机 >>

5.2 Ultrasonic cleaning of nylon microfiltration membranes fouled by Kraft paper mill effluent


Journal of Membrane Science 205 (2002) 247–257

Ultrasonic cleaning of nylon micro?ltration membranes fouled by Kraft paper mill ef?uent
Jianxin Li, R.D. Sanderson? , E.P. Jacobs
Division of Polymer Science, Department of Chemistry, UNESCO Associated Center for Macromolecules and Materials, University of Stellenbosch, P.O. Bag X1, Matieland 7602, South Africa Received 30 April 2001; received in revised form 12 July 2001; accepted 20 March 2002

Abstract An ultrasonic technique was successfully applied to remove fouling and recover the permeate ?ux of ?at sheet micro?ltration (MF) membranes. Three kinds of cleaning methods were used, namely: forward?ushing, ultrasonic cleaning and ultrasound with forward?ushing, and their cleaning ef?ciencies were compared. It was found that ultrasound associated with forward?ushing was a new effective method for the recovery of permeate ?ux. Scanning electron microscopy (SEM) analysis indicated that this method was able to remove fouling layers from a membrane surface and restore the original structure of the membrane surface. The operating conditions during cleaning were investigated. In general, a high forward?ushing velocity and low cleaning solution (water) temperature, under the same ultrasonic conditions, gave higher cleaning ef?ciency. Moreover, online ultrasound can reduce membrane fouling and enhance permeate ?ux. The horn sonicator employed had a frequency of 20 kHz and a power of 375 W. Fouling and cleaning experiments were performed with nylon membranes with 0.2 m average pore diameter. The membranes were fouled by Kraft paper mill ef?uent. ? 2002 Elsevier Science B.V. All rights reserved.
Keywords: Ultrasonic cleaning; Forward?ushing; Micro?ltration; Nylon membrane; Paper ef?uent

1. Introduction Wastewater reclamation in the pulp and paper industry has been emphasized due to the massive amount of water used in this ?eld. It has gathered ever more attention as the regulations on the ef?uent become more stringent. Many technologies including adsorption, chemical oxidation, ion-exchange, evaporation and membrane processes have been used for puri?cation of wastewater from the paper industry. Of these technologies, membrane processes can be tailored according to the required degree of puri?cation [1]. One
? Corresponding author. Tel.: +27-21-8083172; fax: +27-21-8084967. E-mail address: rds@maties.sun.ac.za (R.D. Sanderson).

of the limiting aspects in applying micro?ltration (MF) and ultra?ltration (UF) for wastewater treatment is that of problems with membrane fouling and consequent ?ux reduction. Membrane fouling is characterized by an “irreversible” decline in ?ux. Most of the literature that has appeared over the past two decades focuses on fouling rather than cleaning, even though what appears to be a fouling problem may really be a cleaning problem. Considerable progress has been made in understanding the interactions between the foulants, the membrane and the operating conditions. However, although many techniques have been developed to overcome fouling, membrane cleaning techniques still seem to be practically inadequate for membrane ?ltration systems.

0376-7388/02/$ – see front matter ? 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 6 - 7 3 8 8 ( 0 2 ) 0 0 1 2 1 - 7

248

J. Li et al. / Journal of Membrane Science 205 (2002) 247–257

The typical methods of membrane cleaning that have been used in practice have been forward?ushing (spiralwound and tubular) and backwashing (hollow ?bre) [2–7]. Forward?ushing and backwashing may be especially useful with colloidal suspensions and some tubular membranes. Further, chemicals such as detergents, acids or alkalis are often used to clean fouled membranes [8,9]. However, the chemical methods used sometimes damage the membrane materials and causes secondary pollution. Electrical techniques have also been used to enhance permeate ?ux in membrane ?ltration [10–14]. Charged particles will move away from the membrane surface, depending on the electric ?eld strength applied, thus reducing the extent of concentration polarization and increasing ?ux. There is, however, the danger of electrolysis taking place at the electrodes and gas being generated. pH change of the product stream will be proportional to the applied voltage. Corrosion of electrodes and high power costs have inhibited the commercial practice of this technique [2]. Ultrasound has been widely used as a method for cleaning materials because of the cavitation phenomenon [15]. Cavitation is de?ned as the formation, growth and implosion collapse of bubbles, which are formed when a large negative pressure is applied to a liquid medium. When ultrasound is transmitted through a liquid medium, such as the feed solution, alternate compression and expansion cycles of the medium occur. The compression cycle can cause micro bubbles to collapse, with a release of energy, which causes cleaning of the membrane surface. The collapse of the cavities has suf?cient energy to overcome the interaction between the foulant and the membrane and remove the foulant from the surface of the polymer membrane. Many studies have been carried out to enhance the solvent permeate ?ux using ultrasound treatment. Harvey [16] proposed the use of an acoustic liquid whistle or ultrasound transducer to produce cavitation and to remove concentration polarization, in order to prevent clogging of the membrane in a water desalination process with reverse osmosis membranes. A patent [17] proposes to apply ultrasound to a puri?cation process in which water is passed through an apparatus containing an ion-exchange resin bed. Ultrasound, which is generated constantly during the adsorption and regeneration periods, breaks the

boundary layer and improves these processes. Kost and Langer’s patent [18] proposes a method for enhancing or controlling the permeability of small and large molecular weight molecules in a membrane system which is exposed to ultrasound of a selected intensity of 0.05–30 W/cm2 and a frequency between 10 kHz and 20 MHz for most polymeric membranes and an intensity of 0.05–3 W/cm2 and a frequency of 1–3 MHz for biological membranes. Some papers discuss the effect of ultrasound on the permeation of solutes through polymer membranes. Lenart and Auslander [19] studied the effect of ultrasound on the diffusion of some electrolytes—sodium, potassium and calcium chlorides—through cellophane membranes. The intensity of the ultrasound ?eld in their experiment varied from 1.2 to 6 W/cm2 at 1 MHz frequency. The results obtained demonstrated intensi?cation of the diffusion process by the ultrasound. The authors consider the main cause of acceleration of diffusion with ultrasound to be the appearance of acoustic microcurrents on liquid. Also taking part are radiation pressure, gravitation, cavitation and acoustic pressure. Li et al. [20,21] describe the results of investigations of the in?uence of ultrasound on the diffusion of electrolytes through a cellophane membrane. It was found that the diffusion velocity of electrolyte through the membrane with ultrasound irradiation is higher than that without ultrasound and the amount of solvent permeated increases with acoustic pressure. Band et al. [22] investigated the in?uence of specially modulated ultrasound signals on the water desalination process with a new polymeric material— ion-exchange hollow ?bers. For Na+ –H+ ion-exchange the enhancement effect increased with increasing power of the ultrasound. Chai et al. [23] applied the ultrasound technique to clean polymeric UF and MF membranes fouled by peptone permeation. Ultrasound with a frequency of 45 kHz and an output power of 2.73 W/cm2 was employed to examine the effect of ultrasound on the cleaning of fouled membranes in a cross-?owing ?ltration cell under various ?ltration operating conditions. It is suggested that the cleaning of fouled membranes by ultrasound in association with water cleaning is an effective new method. The above studies provided valuable information on ?ux enhancement by ultrasonic irradiation, but use of the ultrasonic method and its effectiveness is limited.

J. Li et al. / Journal of Membrane Science 205 (2002) 247–257

249

Online ultrasonic irradiation has low ef?ciency and high cost. An ultrasonic bath is only useful in laboratory studies because there is a high waste of acoustic energy in the bath during the cleaning process. In the present study, a horn ultrasonic cleaner (Model W-375, Ultrasonics Inc.), with a frequency of 20 kHz and a power density of 82.9 W/cm2 , was employed. A ?at cell was irradiated with the sonicator. The effect of ultrasound on the cleaning of nylon MF membranes fouled by ef?uent from the paper industry was studied. Research was focused on the effect of ultrasound on the cleaning of fouled membranes and a comparison of results obtained from cleaning by forward?ushing, ultrasonic cleaning and ultrasound associated with forward?ushing. The operating conditions during cleaning and the effect of online ultrasound on the permeate ?ux were also investigated.

Table 1 Characteristics of paper mill ef?uent Items pH Aluminium (Al) Calcium (Ca) Magnesium (Mg) Potassium (K) Bicarbonate (HCO3 ) Chloride (Cl) Turbidity Values (mg/l) 4.96 16 353 27 30 98 10 64 Items Conductivity ( s/cm) Boron (B) Iron (Fe) Sodium (Na) Sulphur (S) Nitrate (NO3 ) M-alkalinity (CaCO3 ) Suspended solids (SS) Values (mg/l) 5550 2 3 908 1075 50 27.2 6636

This data is from Mondi Piet Retief, Mondi Ltd., South Africa.

2. Experimental procedures 2.1. MF and ultrasonic cleaning system The major considerations in choosing an appropriate MF system included industrial relevance and wastewater-disposal problems. Consequently, the feed solution selected was the ef?uent from a wastewater treatment plant of the Mondi Kraft mill at Piet Retief, South Africa. This plant has experienced long term problems with membrane fouling. The ef?uent from

the paper mill contains an amount of breakdown products of lignin and lignosulphonate. The wastewater treatment plant comprised the following processes: pre-treatment, dissolved air ?otation (DAF), MF and UF. Samples were taken from the DAF product. The characteristics of the ef?uent are summarized in Table 1. The ?at-sheet MF experimental setup, shown in Fig. 1, allows for the accurate control of inlet pressure, retentate ?ow rate and temperature. In each MF experiment, continuous stirring in the feed tank was provided. The permeate ?ux was measured by a electrical balance, connected to a PC. During the experiments both the retentate and permeate were recycled to the feed tank after ?ux measurements. Fig. 1 is also a schematic representation of the ultrasonic cleaning

Fig. 1. Experimental setup for cross-?ow micro?ltration and ultrasonic cleaning.

250

J. Li et al. / Journal of Membrane Science 205 (2002) 247–257

system used. A horn ultrasonic cleaner (Model W-375, Ultrasonics Inc.), with a frequency of 20 kHz and a power density of 82.9 W/cm2 , was employed in this study. The cross-?ow ?ltration cell was immersed in a water bath during ultrasonic cleaning. The cell was irradiated with the sonicator. The ?at cell had the following speci?cs: made from Perspex, the thickness of the top plate 20 mm, a cavity 90 mm × 35 mm × 14 mm on the top plate, and the height of feed channel 2.5–3 mm. The horn transducer was ?xed by a rack so as to keep the same distance between transducer and cell in each experiment. 2.2. Cross-?ow ?ltration experiments The permeate ?ux was measured every 10–20 s by a balance. Each experiment commenced with pure water being circulated through the system at a ?xed ?ow rate and applied pressure for about half an hour to compress the membrane and to build up a stable ?ow ?eld. The ?ux values for new, clean membrane, referred to as Jw , were recorded after 30 min of pure water ?ltration. The pure water was then replaced by ef?uent. The ?ow rate of the feed and operating pressure were ?xed at 0.125 m/s and 50 kPa, respectively, for the duration of the ?ltration experiments. The turbidity of the feed and permeate solution were 64 and 1, respectively. 2.3. Cleaning experiments After the membranes were fouled with paper mill ef?uent, the feed solution was changed from ef?uent to water. The fouled membranes were cleaned by three methods, namely: forward?ushing, ultrasonic cleaning, and ultrasound associated with forward?ushing. The duration was 10 min. Forward?ushing was performed with valve V2 closed. Valve V1 was opened only after the transmembrane pressure differential over the membrane subsided and reached zero, under the pre-determined ?ushing velocity. Ultrasound associated with forward?ushing was performed with intermittent ?ushing under a pre-determined ?ushing velocity during ultrasonic irradiation. To investigate the cleaning ef?ciency of each cleaning method, the cleaned membrane was used to ?lter pure water under the same operating condition as used during the fouling phase.

3. Results and discussion 3.1. Effect of online ultrasound on the ?ux during the fouling process The following experiment was conducted in order to investigate the effect of online ultrasound on the permeate ?ux during fouling. The fouling experiment was carried out with a feed of paper mill ef?uent at 0.125 m/s axial velocity and 50 kPa, after an initial 30 min period of pure water ?ltration. The sonicator was turned on after fouling ?ltration for 90 min. Fig. 2 shows the in?uence of online ultrasound on the permeate ?ux. A rapid decline in permeate ?ux is observed because of the presence and formation of a fouling layer on the nylon membrane surface. The permeate ?ux value ?nally reached was 21.45 l/(m2 h) after 90 min of fouling operation. The ?ux increased to 28.8 l/(m2 h) when the sonicator was turned on for 5 min, then slightly decreased from 28.8 to 25.98 l/(m2 h) after ultrasonic irradiation for 30 min. Fig. 2 also shows the results of online ultrasonic cleaning for 20 min, after 90 min of fouling operation. Online ultrasonic cleaning with feed can enhance the permeate ?ux from 21.45 to 29.4 l/(m2 h). But the ?ux also slightly decreased from 29.4 to 23.04 l/(m2 h) after 30 min of fouling operation. Further experiments were carried out with and without ultrasonic irradiation during the paper mill

Fig. 2. Permeate ?ux changes with ?ltration time during paper mill ef?uent fouling experiment in the absence and presence of ultrasound (20 kHz), and fouling again after online ultrasonic cleaning for 20 min, after fouling time of 90 min. The operating pressure was 50 kPa.

J. Li et al. / Journal of Membrane Science 205 (2002) 247–257

251

Fig. 3. Changes in the permeate ?ux with and without ultrasonic irradiation during paper mill ef?uent fouling experiment at 50 kPa and 0.125 m/s.

ef?uent fouling experiment. Fig. 3 shows the changes in the permeate ?ux with and without ultrasonic irradiation. Without ultrasonic irradiation, the ?ux value declined from an initial 172 to 66 and 40 l/(m2 h) at 10 and 30 min. On the other hand, when ultrasonic irradiation was used at the beginning of fouling, the ?ux decline was slower and the ?ux values obtained were 81 and 51 l/(m2 h) at 10 and 30 min. Results of this experiment suggestion that ultrasonic irradiation can enhance the ?ux during MF. However, no ?ux difference was found between membranes treated with and without ultrasonic irradiation after 50 min of fouling operation. There may be two reasons why the ?ux enhancement cannot be obtained over long periods of online ultrasonic treatment. First, the sound energy will decrease because of the increased temperature of the transducer during operation. Second, the agitation due to the ultrasonic irradiation is likely to be complicated by the substantial increase in the number of cavitation nucleation sites generated on the new particle surface. When larger particles exist in the feed (average particle size: 0.947 m in the paper mill ef?uent, tested by Zetasizer 1000HS), ultrasound can lead to a reduced ?ux rate. Presumably the momentum needed to re-suspend these particles is more than that which can be supplied by the application of the ultrasound, and the motion that is imparted to the particles in the fouling layers causes them to ‘jostle’ and pack more densely [13].

3.2. Recovery of permeate ?ux by different cleaning methods The effect of membrane cleaning was investigated by three different cleaning methods: forward?ushing, ultrasonic cleaning and ultrasound associated with forward?ushing. To compare the ef?ciencies of these methods, nylon membranes were fouled, under the same conditions, with ef?uent fouled by the DAF product from the paper industry, after pure water ?ltration. Each cleaning process was then applied, individually, for 10 min, after 80 min of fouling operation. These nylon membranes were used to ?lter pure water, under the same conditions and their permeate ?uxes determined. A graphical representation of permeate ?ux versus time is shown in Fig. 4. At the beginning of fouling operation the permeate ?ux declined rapidly, followed by a more gradual decline after fouling for 20 min (Fig. 4). The permeate ?ux was 52.8 l/(m2 h) after an 80 min fouling period. When the feed was changed from the ef?uent to pure water, the permeate ?ux increased to 86.4 l/(m2 h) after 30 min of pure water ?ltration, because of a reduction in concentration polarization. Forward?ushing increased the water permeate ?ux from 86.4 to 122.4 l/(m2 h) at 110 min of operation. An increase in the water permeate ?ux was observed after ultrasound cleaning. It is suggested that ultrasonic irradiation can effectively clean membrane fouling in MF. The value of 350 l/(m2 h) at 30 min of pure water operation

252

J. Li et al. / Journal of Membrane Science 205 (2002) 247–257

Fig. 4. Effect of different cleaning methods on the water ?ux after 80 min of fouling operation by paper mill ef?uent: water ?ltering, forward?ushing, ultrasonic cleaning and ultrasound with forward?ushing. The water ?ltration is used for examining the recovery of permeate ?ux after 10 min of cleaning. Operating temperature 23 ? C, pressure 50 kPa ?ow rate 0.125 m/s, and cleaning time 10 min.

obtained after ultrasonic cleaning was, however, still lower than the original water ?ux of 900 l/(m2 h). Hence, ultrasound associated with forward?ushing was applied in an effort to restore the permeate ?ux of the membrane. The water permeate ?ux of the membrane cleaned by ultrasound with forward?ushing was 738 l/(m2 h) at 30 min of pure water operation. In this case results showed that ultrasound associated with ?ushing was the most effective of these methods. In an effort to obtain a better understanding of the fouling and cleaning processes, we calculated the various resistances of the membrane during the fouling and cleaning procedures. The permeate ?ux during MF is usually written in terms of transmembrane pressure difference ( P) and a total resistance, according to the resistance model [24,25]: P J = (?Rt ) (1)

where Rm , the resistance of new or clean membranes, can be calculated from Eq. (1), written for pure water, i.e. P Jw = (3) (?w Rm ) where Jw is the pure water permeate ?ux for a new membrane (m/s) (at 30 min of pure water operation); ?w the viscosity of pure water (Pa s). Rr is a reversible resistance, i.e. particle polarization, cake layer, Rf is the fouling resistance, an irreversibly adsorbed layer which cannot be removed by water cleaning. Rr is removed by cleaning the membrane with water at a low ?ow rate (<12.5 cm/s). The water ?ux measured after different cleaning methods is Jw . Therefore: Rm + Rf = P (?w Jw ) (4)

where J is the permeate ?ux of solution (m/s), P the transmembrane pressure difference (kPa), Rt the total resistance (m?1 ) and ? is the viscosity of solution (Pa s). The viscosity of the paper ef?uent in the experiments was measured to be 1.02 × 10?3 Pa s at 20 ? C. In this study, the total resistance Rt is de?ned as Rt = Rm + Rr + Rf (2)

Each resistance can be readily calculated using the experimental data and Eqs. (1)–(4). Table 2 shows various resistances and their percentages during fouling. The membrane resistance was 6% of the total resistance. Therefore, the main resistances of the fouled membrane were Rr and Rf . This is because concentration polarization and fouling layers appeared on the membrane surface during the fouling period. The reversible resistance, Rr , is 34.9% of the total resistance. It is known that changing

J. Li et al. / Journal of Membrane Science 205 (2002) 247–257 Table 2 Various resistances R (×1012 m?1 ) and their percentages during fouling Resistances Rm Rr Rf Rt R 0.4 2.33 3.95 6.68 % 6 34.9 59.1 100

253

97.8%. This suggests that both procedures are necessary to completely remove the foulant from a membrane. It is known that in many liquids, high-energy ultrasound (a horn transducer with a frequency of 20 kHz and a power density of 82.9 W/cm2 in this study) produces observable effects such as caviatation, the rapid movement of ?uid caused by variations of sonic pressure, and microstreaming [14]. 3.3. SEM microscopy To investigate the effect of different cleaning methods on the fouling layer of a nylon membrane, SEM micrographs of new, fouled and cleaned MF membranes were recorded. They are shown in Figs. 5–7. The image of a new membrane surface (Fig. 5a) shows the typical membrane surface structure, including the membrane pores. Fig. 5b shows that after 80 min of operation, only some larger pores remained partially unblocked. Inductively coupled plasma (ICP) analysis showed that the main chemical composition of the fouling layer is breakdown products of lignin or lignosulphonate [26]. The surface of the fouled membrane appeared to have more pores after forward?ushing (Fig. 6a). This result indicated that forward?ushing could partly reduce fouling resistance. It was, however, dif?cult to remove all fouling matter by ?ushing alone. Ultrasonic irradiation can loosen, break and disperse the fouling layer because of ultrasonic cavitation and acoustic streaming (Fig. 6b). Hence, the permeate ?ux increased after ultrasonic cleaning although some fouling still remained on the membrane surface. The clear pore structure of the cleaned membrane (Fig. 5) appeared again after ultrasound with forward?ushing because this cleaning method can completely remove the fouling from the membrane surface. 3.4. Effect of different cross-?ow velocities during ultrasonic irradiation Fouling involves the accumulation and compaction of retained material at a membrane surface, while the solvent (water) passes through the pores. The deposited fouling layer is expected to become re-suspended and swept away by tangential- or cross?ow [27]. So, the different forward?ushing velocities resulted in different cleaning ef?ciencies.

Rm , the resistance of new or clean membranes; Rr , a reversible resistance; Rf , fouling resistance; Rt , total resistance.

operating conditions or using ?ow destabilization can reduce reversible resistance, enhance the mass transfer coef?cient and increase permeate ?ux. Fouling resistance, Rf , is 59.1% of the total resistance. Rf was 1.7 Rr , meaning that fouling was the main cause of the decrease in permeates ?ux in MF. To compare the effectiveness of these cleaning methods, the cleaning ef?ciency Ec (%) was de?ned as follows: (Rf ? Rc ) Ec = × 100 (5) Rf where Rf and Rc represent the resistance (m?1 ) of fouled and cleaned membrane, respectively. Rc was calculated from Eq. (4). Table 3 shows Rc and Ec for each cleaning method. It is seen that ?ushing can decrease the fouling resistance by only 8.8%. It is suggested that paper ef?uent fouling is dif?cult to clean using only water cleaning or ?ushing. In this case, membrane fouling is caused by the adsorption of foulant both on and inside the membrane. This membrane fouling is irreversible and additional cleaning methods are needed. Ultrasonic cleaning is an effective method to decrease membrane fouling in MF. It was determined that ultrasonic irradiation can reduce the fouling resistance by 84%. Under similar conditions, the cleaning ef?ciency of ultrasound associated with ?ushing was
Table 3 Cleaned membrane resistance Rc (×1011 m?1 ) and cleaning ef?ciency Ec (%) of different cleaning methods Cleaning method Flushing Ultrasound Ultrasonic ?ushing Rc 36 6.3 0.88 Ec (%) 8.8 84 97.8

254

J. Li et al. / Journal of Membrane Science 205 (2002) 247–257

Fig. 5. SEM micrographs of nylon membranes: (a) new membrane; (b) fouled membrane, magni?cation 20,000×.

The effect of different forward?ushing velocities under ultrasonic irradiation is shown in Fig. 8. Forward?ushing slightly increased the water permeate ?ux (compared with water ?ow) even when the velocity was increased from 0.21 to 0.56 m/s. This indicated that a fouling layer accumulated and compacted on/within the membrane surface and that it was

dif?cult to clean it and sweep it away by the forward?ushing method only. Fig. 8 also shows that ultrasound with forward?ushing can apparently enhance water permeate ?ux, as the forward?ushing velocity increased from 0.21 to 0.56 m/s under ultrasonic irradiation. This is because of the effectiveness of ultrasound and forward?ushing. The forward?ushing

Fig. 6. SEM micrographs of nylon membranes: (a) cleaned membrane by forward?ushing; (b) cleaned membrane by ultrasonic irradiation, magni?cation 20,000×.

J. Li et al. / Journal of Membrane Science 205 (2002) 247–257

255

Fig. 7. SEM micrographs of nylon membrane cleaned by ultrasonic forward?ushing, magni?cation 20,000×.

produces a high tangential or cross-?ow rate on fouling layers as the forward?ushing velociy increases. 3.5. Effect of different cleaning temperatures on permeate ?ux under sonication Since temperature in?uences the cleaning ef?ciency of ultrasound, the cleaning performance of ?ushing

under ultrasound was examined at different cleaning temperatures. Nylon membranes were again fouled for 80 min, then cleaned by ultrasound associated with forward?ushing (velocity, 0.4–0.5 m/s) at temperatures of 23, 30 and 40 ? C for 10 min. The pure water ?uxes of the cleaned membranes were determined under the same operating conditions. Fig. 9 shows the changes in the permeate ?uxes during fouling and after cleaning.

Fig. 8. Effects of different forward?ushing velocities on the recovery of permeate ?ux. The ?ow rate and pressure are 18 l/h (0.125 m/s) and 50 kPa during the fouling phase. The water ?ltration is used for cleaning and examining the recovery of permeate ?ux after an 80 min fouling period at the same pressure and ?ow rate as used in the fouling phase. The forward?ushing velocities are 0.21, 0.42 and 0.56 m/s. The cleaning time is 10 min.

256

J. Li et al. / Journal of Membrane Science 205 (2002) 247–257

Fig. 9. Effects of different temperatures of ultrasound on the recovery of permeate ?ux. The ?ow rate and pressure are 18 l/h (0.125 m/s) and 50 kPa during fouling phase. The water ?ltration is used for examining the recovery of permeate ?ux after an 80 min fouling period. The operating temperatures during the fouling and cleaning phases are 23, 30 and 40 ? C. The cleaning time is 10 min.

It is seen that the water permeate ?ux of the nylon membrane decreases as the cleaning temperature increases from 23 to 40 ? C. Firstly, this is due to a change in the ultrasonic cavitation intensity as the temperature decreases [28]. Ultrasonic cleaning of the MF membrane is mainly caused by ultrasonic cavitation and the acoustically excited bubble break-ups on the membrane surface. An increase in solution temperature gives rise to an increase in the vapor saturation pressure in the bubble so that the shock-wave intensity during the bubble break-up is reduced. In other words, the cavitation effectiveness decreases. Secondly, the ?uid viscosity also changes with temperature and could have caused the change in ultrasonic effectiveness. An increase in the viscosity of a liquid will result in an increase in the threshold of cavitation [29]. There will probably be several different types of cavity in the liquid: (a) an empty cavity (true cavitation); (b) a vapor-?lled cavity; (c) a gas-?lled cavity; (d) a combination of vapor- and gas-?lled cavities. A lower temperature favors the formation of empty cavities. 4. Conclusions This study describes the effect of ultrasound on the cleaning of nylon MF membranes. Nylon membrane

were fouled by pulp and paper ef?uent and cleaned by the methods of forward?ushing, ultrasonic cleaning and ultrasound associated with forward?ushing. The cleaning ef?ciencies were determined by pure water ?ux comparisons. Experimental results showed that ultrasound associated with ?ushing was the most effective of these cleaning methods. Cleaning by this method can clean fouled membranes and restore the original membrane morphology completely. Resistance analysis indicated that the cleaning ef?ciencies of ultrasonic cleaning and ultrasound with forwrd?ushing were 87 and 97.8%, respectively. It was also found that high forward?ushing velocity combined with ultrasound was also bene?cial for cleaning. To be effective, the ?ushing velocity should be greater than the normal operating velocity. At different cleaning temperatures, under ultrasonic irradiation, the pure water ?ux increased as the cleaning temperature decreased from 40 to 23 ? C. Acknowledgements This study was supported by the Water Research Commission of South Africa. Jianxin Li gratefully acknowledges the ?nancial support from the South African National Research Foundation (NRF) and THRIP (Human Resources and Industry Programs).

J. Li et al. / Journal of Membrane Science 205 (2002) 247–257

257

References
[1] J. Nourtila-Jokinen, M. Nystrom, J. Membr. Sci. 119 (1996) 99–115. [2] M. Cheryan, Ultra?ltration Handbook, Technomic. Pub. Co. Inc., Lancaster, 1998, p. 267. [3] M. Kennedy, S.M. Kim, I. Muteryo, L. Broens, J. Schippers, Intermittent cross?ushing of hollow ?ber ultra?ltration system, Desalination 118 (1998) 175–188. [4] S.G. Redkar, R.H. Davis, Enhancement of cross-?ow micro?ltration performance using high frequency reverse ?ltration, AIChEJ 41 (1995) 501–508. [5] C.S. Parnham, R.H. Davis, Protein recovery from bacterial cell debris using cross-?ow micro?ltration with backpulsing, J. Membr. Sci. 118 (1996) 259–268. [6] S. Redkar, V. Kuberkar, R.H. Davis, Modeling of concentration polarization and depolarization with high frequency backpulsing, J. Membr. Sci. 121 (1996) 229–242. [7] M. Pontie, X. Chasseray, D. Lemordant, et al., The streaming potential method for the characterization of ultra?ltration organic membranes and the control of cleaning treatments characterization, J. Membr. Sci. 129 (1997) 125–133. [8] A.E. Jaffar, The application of a novel chemical treatment program to mitigate scaling and fouling in reverse osmosis units, Desalination 96 (1994) 71–79. [9] D. Mukherjee, A. Kulkarni, W.N. Gill, Chemical treatment for improved performance of reverse osmosis membranes, Desalination 104 (1996) 239–249. [10] S. Lentdch, P. Aimar, J.L. Orozco, Enhanced separation of albumin-poly(ethylene glycol) by combination of ultra?ltration and electrophoresis, J. Membr. Sci. 80 (1993) 221–232. [11] H.S. Muralidhara, Enhance separations with electricity, Chemtechnology (5) (1994) 36–41. [12] H.M. Huotari, G. Tragardh, I.H. Huisman, Cross-?ow membrane ?ltration enhanced by an external dc electric ?eld: a review, Chem. Eng. Res. Design 77 (1999) 461–468. [13] E.S. Tarleton, R.J. Wakeman, Electroacoustic cross-?ow micro?ltration, Filtr. & Sep. 29 (9/10) (1992) 425–432. [14] R.J. Wakeman, E.S. Tarleton, An experimental study of electroacoustic cross-?ow micro?ltration, Chem. Eng. Res. Design 69 (1991) 386–397.

[15] G.J. Price (Ed.), Current Trends in Sonochemistry, Royal Society of Chemistry, Cambridge, 1992, pp. 1–7. [16] R. Harvey, US Patent no. 3,206,397 (1965). [17] W. Shimichi, Japan Patent no. 07,31974 (1995). [18] J. Kost, R. Langer, Ultrasound enhancement of membrane permeability, US Patent no. 4,7802,212 (1988). [19] I. Lenart, D. Auslander, Ultrasonic 9 (1980) 216. [20] H. Li, E. Ohdaria, M. Ide, Enhancement in diffusion of electrolyte through membranes using ultrasonic dialysis equipment with plane membrane, Jpn. J. Appl. Phys. 35 (5B) (1995) 2725. [21] H. Li, E. Ohdaria, M. Ide, Effect of ultrasonic irradiation on permeability of dialysis membrane, Jpn. J. Appl. Phys. 35 (5B) (1995) 3225. [22] M. Band, M. Gutman, V. Faerman, E. Korngold, J. Kost, P.J. Plath, V. Gantar, In?uence of specially modulated ultrasound on the water desalination process with ion-exchange hollow ?bres, Desalination 109 (1997) 303–313. [23] Xijun Chai, Takaomi Kobayashi, Nobuyuki Fujii, Ultrasoundassociated cleaning of polymeric membranes for water treatment, Sep. Purif. Technol. 15 (1999) 139–146. [24] R. Jiraratannon, A. Chanachai, A study of fouling in the ultra?ltration of passion fruit juice, J. Membr. Sci. 111 (1996) 39–48. [25] Marcel Mulder. Basic Principles of Membrane Technology, Kluwer Academic Publishers, Dordrecht, 1991, p. 181. [26] G.S. Domingo, E.P. Jacobs, P. Swart, Characterization of foulants present in ef?uents emanating from the Piet Retief Mondi Kraft paper mill, in: Proceedings of the 4th WISA-MTD Symposium on Membranes: Science and Engineering, Vol. 3, Stellenbosch, South Africa, 2001, pp. 26–27. [27] S.A. Perusich, R.C. Alkire, Ultrasonically induced cavitation studies of electrochemical passivity and transport mechanisms, I. Theoretical J. Electrochem. Soc. 138 (3) (1991) 700–707. [28] Chao Zhu, Guangliang Liu, Modeling of ultrasonic enhancement on membrane distillation, J. Membr. Sci. 176 (2000) 31–34. [29] T.J. Mason, J.P. Lorimer, Sonochemistry: Theory, Applications and Uses of Ultrasound in Chemistry, Ellis Horwood, Chichester, 1989, p. 29.


相关文章:
浸没式中空纤维膜反应器的在线超声清洗_图文.pdf
1.2 试验方法 MBR 在恒定膜通量下, 连续运行 3 d, 污染过程 P  ...Ultrasonic cleaning of nylon micro-filtration membranes fouled by Kraft paper ...
Investigation of microfiltration for treatment of e....pdf
Investigation of microfiltration for treatment of emulsified oily wastewater from...the fouled membrane could be recovered by using conventional cleaning methods....
Microfiltration of milk with backpulsing.pdf
Microfiltration of milk with backpulsing_电子/电路...ceramic membranes with pore sizes of 0.1 and ...fouled quicker than when backpulsing was not used...
Anti-fouling preadsorb on membranes_图文.pdf
of the UF and microfiltration (MF) membranes was...0.5 vol.% alkaline cleaning and disinfectant ...of the membranes fouled by the synthetic polymers...
biofouling on membranes-....ppt
Biofouling on Membranes-causes,consequences, control...Microfiltration: The Role of Extracellular Matrix',...accounts for delta P rise in biofouled modules....
RevOsmosisP反渗透_图文.ppt
Microfiltration Range Particle size Removes Macro ...fouled: - Increasing operating costs - Reducing ...Hydrolysis Factors of RO Membranes CA MEMBRANES ?...
超滤膜污染的预防与控制....pdf
of microfiltration membranes by NOM after ...膜科学与技术 ,2010, 4 5 周 玮 ,王国栋 , 梁...et al.Consecutive chemical cleaningoffouledPVC ...
HFK131超滤膜.pdf
cleaning cycle and with water of acceptable ...HFK-131 membranes may be irreversibly fouled if ...microfiltration, and ultrafiltration membranes and membrane...
更多相关标签: