- Email: [email protected]
Enhanced chemical oxygen demand removal and flux reduction in pulp and paper wastewater treatment using laccase-polymerized membrane filtration
Journal of Hazardous Materials 181 (2010) 763–770
Contents lists available at ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Enhanced chemical oxygen demand removal and flux reduction in pulp and paper wastewater treatment using laccase-polymerized membrane filtration Chun-Han Ko a , Chihhao Fan b,∗ a b
School of Forest and Resources Conservation, National Taiwan University, No. 1, Sec. 4, Roosevelt Rd, Da-An District, Taipei 10617, Taiwan Department of Safety, Health, and Environmental Engineering, Ming Chi University of Technology, 84 Gung-Juan Rd, Taishan 24301, Taipei County, Taiwan
a r t i c l e
i n f o
Article history: Received 30 January 2010 Received in revised form 16 May 2010 Accepted 17 May 2010 Available online 25 May 2010 Keywords: Chemical oxygen demand Enzymatic degradation Permeate flux Pulp and paper Polymerization
a b s t r a c t The purpose of this present study is to investigate the removal efficiency of chemical oxygen demand (COD) from pulp and paper wastewater using laccase-polymerized membrane filtration process. The membranes with molecular weight cut-off (MWCO) of 5000 and 10,000, 30,000 and 54,000 were used in a cross-flow module to treat the pulp and paper wastewater containing high phenolic constituents and COD. With 2.98 IU/L of activated laccase applied at room temperature for 180 min, the contaminants in raw wastewater and second effluent were polymerized to form larger molecules with average molecular weight of 1300 and 900 Da (Dalton), respectively. With laccase polymerization prior to filtration, over 60% removals of COD by the four investigated membranes were observed, compared with low COD removal without laccase polymerization. Moreover, the addition of laccase resulted in 4–14% reduction of membrane permeability during the first 180 min filtration operation due to gel layer formation by the polymerization. No further flux decline was observed afterwards indicating the steady state was reached and the membranes could be used to remove the polymerized pollutants without significant fouling. The maximum apparent resistance occurrence for raw wastewater treated with laccase also supported the effectiveness for COD removal with laccase polymerization before membrane filtration. Additionally, pretreatment by inactivated laccase only caused further flux reduction without additional removal of COD. © 2010 Elsevier B.V. All rights reserved.
1. Introduction The pulp and paper industry has close relationship to the economic development, and the increasing amount of paper utilization makes the pulp and paper industry even more important in the process of societal modernization. Unfortunately, pulp and paper mills generate a considerable amount of wastewater containing various types of contaminants [1–3]. In the literature, it has been reported that high concentrations of chlorinated phenolic acids and chemical oxygen demand (COD) are a basic characteristic of pulp and paper wastewater [1]. These phenolic compounds are recalcitrant in the environment, difficult to biodegrade and detrimental to the public health. Also, the dark brown colour of the wastewater due to the presence of abundant partially oxidized organic lignins inhibits the natural process of photosynthesis, and many times, results in unacceptable odourous problems [1,4–7]. Without appropriate treatment before discharge, the pulp and paper wastewater becomes a severe threat to the aquatic environment as well as the riverine ecology.
∗ Corresponding author. Tel.: +886 229089899x4656; fax: +886 229080783. E-mail address: [email protected] (C. Fan). 0304-3894/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jhazmat.2010.05.079
In order to protect the aquatic environment from being polluted by pulp and paper wastewater, many contaminant-removing facilities were installed to improve the effluent water quality of pulp and paper mills. The frequent-used conventional treatment units to remove COD include sedimentation and floatation, coagulation and precipitation, adsorption, chemical oxidation, and so many. Although many physical and chemical methods are quite effective in treating pulp and paper wastewater, much effort still has been expended in the innovation of wastewater treatment technology due to the high capital and energy costs and instability in operation of the conventional treatment technologies [8]. Besides, the elevated living standard requests by the public and the increasingly stringent effluent discharge standards have brought on the inevitable advancement in wastewater treatment efficiency for pulp and paper mills. There has been an increasing popularity in membrane technology applied to treating wastewater from hardwood kraft mills over the past two decades because membrane processes offer a high level of contaminant separation and relatively low energy consumption. Pressure-driven membrane processes, including reverse osmosis, nanofiltration, ultrafiltration and microfiltration, are of a great deal of interests because they may prevent a wide range of small particles from penetrating membranes, depending on the
764
C.-H. Ko, C. Fan / Journal of Hazardous Materials 181 (2010) 763–770
Fig. 1. The polystyrene sulfonate standards (straight lines) at 2220, 910, and 220 Da. and the average molecular weights of raw wastewater and the secondary effluent before (dash-dotted line) and after (solid line) laccase addition.
pore size or molecular weight cut-off (MWCO) of the membrane. Recently enzymatic polymerization has been demonstrated to be able to effectively treat the wastewater containing toxic phenolic compounds through physical separation [9,10], although other researches indicated the contaminants from pulp and paper mills might be removed or degraded through different enzymatic reaction mechanisms [11–13]. Enzymes may catalyze the oxidation of phenolic substances and certain amines while reducing oxygen to water [14]. In the literature, it has been reported that adding laccase with 10 IU/mL to wastewater may facilitate the removal of phenolic compounds in pulp mill effluents by inducing cross-linking and precipitation among contaminants [15]. Laccase is a blue copper containing polyphenol oxidase which plays an important role both in lignin biosynthesis and in biodegradation [14]. Compared to other enzymes, laccase is rather unspecific and suitable for use in wide range of applications, such as pulp deligninfication, wastewater detoxification and denim decolourization [16,17]. A successful application of laccase-assisted membrane separation technology to treating various synthetic wastewaters each containing one selected phenol has been reported [18]. In this present study, laccase with the activity of 2.98 IU/L (i.e., 0.3 g protein/L reaction solution) was introduced to polymerize the organic contaminants in the pulp and paper wastewater. The COD removal efficiencies of the resulting polymerized wastewater were evaluated by cross-flow filtration using 4 different membranes with the MWCO of 5000, 10,000, 30,000, and 54,000 Da, respectively. The raw wastewater and secondary effluent of a consolidated kraft pulping and fine grade papermaking mill were used. The flux penetrating each tested membrane at pH value of 7.5 and room temperature was analyzed and the apparent resistance was calculated as well.
the Danish company of Novozyme. The membranes were obtained from GE Osmonics. The membranes of 5000 and 10,000 Da MWCO were made of polyethersulfone. The membranes of 30,000 and 54,000 Da MWCO were made of polysulfone.
2. Materials and methods
2.3. Gel permeation chromatography (GPC)
2.1. Wastewater samples and materials
The molecular weights of the phenolic compounds were analyzed by gel permeation chromatography (GPC). The GPC is operated by the principle of size exclusion, in which target molecules are separated based on their difference in sizes. The UV absorbing organics were eluted in the descending orders of molecular weights, i.e., organics with larger molecular weights were eluted at earlier times since they have difficulties to migrate into the finer micropore structure of column packing materials. The system used in this study consisted of a Jasco 880 HPLC PU pump, a Jasco 875 UV detector, connected to a guard column of Shodex SBG, followed by Shodex SB-802 HQ and SB-803 HQ 8 mm × 300 mm columns (Showa Denko K.K., Japan) in series. The serial columns used were of polyhydroxymethacrylate packing gels with following molecular weight exclusion limits of eluent conditions: 4000 and
The wastewater samples used were the raw wastewater and secondary effluent collected from a kraft pulp and paper mill operating at 47,000 CMD in eastern Taiwan. The wastewater was treated through a series of physical/chemical/biological units of stabilization, coagulation, flocculation, sedimentation, aeration followed by activated sludge, and acidification before being discharged. All the collected wastewater samples were filtered by 0.5 m PE membrane prior to laccase pretreatment and membrane filtration. For each wastewater sample, the pH was adjusted to 7.5 using 5 mM of phosphate buffer. All the chemicals used were of reagent grade purchased from Sigma. The water used was deionized (DI) water. The enzyme used in this study was Laccase 51003 obtained from
2.2. Enzymatic polymerization In pretreatment process, the laccase was added into the water sample and stirred for 1 h, resulting in the activity of 2.98 IU/L prior to filtration tests. The activity of laccase was measured by the rate of oxidation of syringaldazine. One international unit (IU) is defined as the amount of enzyme which oxidizes 1 milli molar (mM) syringaldazine per minute under the condition of pH 7.5 and 30 ◦ C. After 60 min of reaction, 4% (w/v) sodium azide solution, resulting in the total concentration of 0.004%, was added to stop the enzymatic reaction. The average molecular weights of the constituents in treated samples were monitored at the reaction times 1 and 3 h. The monitoring result showed that most of the polymerization occurred during the first 1 h. Therefore, the samples for subsequent molecular weight analysis were collected after 1 h polymerization reaction. For additional comparison, the laccase was inactivated by sterilization at 120 ◦ C for 2 h, and the resulting inactivated enzyme solution was added to water samples to evaluate its impact on COD removal from pulp and paper wastewater as well. The overall laccase activity in the stock solution was 42,973 IU/mL, with the COD concentration of 30,800 mg/L. The protein concentration of laccase stock solution employed in the present work was 4320 mg/L. Since laccase was added in reaction mixture at the dosage of 2.98 IU/L, the contributions for both COD (i.e., 0.0021 mg/L) and protein mass loadings (i.e., 0.0003 mg/L) from laccase addition were assumed negligible.
C.-H. Ko, C. Fan / Journal of Hazardous Materials 181 (2010) 763–770
Fig. 2. COD variations for raw wastewater treated with 54,000, 30,000, 10,000, and 5000 MWCO membranes.
Fig. 3. COD variations for secondary effluent treated with 54,000, 30,000, 10,000, and 5000 MWCO membranes.
765
766
C.-H. Ko, C. Fan / Journal of Hazardous Materials 181 (2010) 763–770
100,000 Da, respectively. The 0.005 M K2 HPO4 and DI water were used as the mobile phases for the raw wastewater and secondary effluent, respectively. The flow rate was controlled at 0.8 mL/min. The molecular weight distribution was monitored by a UV detector at the wavelength of 280 nm. For system calibration, the aqueous polystyrene sulfonate standards purchased from Polymer Standards Service GmbH, Mainz, Germany were used. These standards with molecular weights of 2220, 910 and 220 Da were eluted at 18.58, 22.42 and 25.82 min, respectively. The molecular weights were regressed with the obtained elution times, and the resulting calibration was employed as the reference for molecular weight estimation in the subsequent analysis using chromatography manager software [18]. The COD removal efficiency was calculated taking the COD concentration difference between influent and effluent divided by influent COD concentration. 2.4. Membrane filtration The cross-flow membrane filtration module used in this study consisted of a liquid reservoir, a gear pump (Cole-Parmer Model 75211-10) and a flat sheet plate with filtration area of 0.0084 m2 (Chi-Guang, Taiwan). For each tested wastewater, two samples were pretreated with laccase and inactivated laccase, respectively, prior to the filtration experiment. Additionally, another wastewater sample was used for filtration experiments without further pretreatment. Each filtration experiment was conducted at the operation pressure of 45 psi (3 kg/cm2 ) under room temperature (22 ± 2 ◦ C) for 9 h. Aqueous samples were collected in duplicate and the COD concentration was determined according to the standard methods [19]. The total phenols contents were estimated with a spectrophotometer method based on the Folin and Ciocalteu reagent [20].
Fig. 4. COD removal efficiencies of (a) raw wastewater and (b) the secondary effluent for the membrane MWCO at 5000, 10,000, 30,000, and 54,000.
3. Results and discussion 3.1. Molecular weights increase by laccase polymerization Fig. 1 shows the chromatograms of eluted average molecular weights of the raw wastewater and secondary effluent before and after the addition of laccase. The empty boxes are the responses of selected standards at their respective elution times. The dotted and solid curves are the chromatographic graphs for tested water samples without and after laccase pretreatment, respectively. For raw wastewater, four major peaks were observed at 1464, 1107, 590 and 304 Da, and three peaks were observed at 1711, 1344 and 795 Da after laccase addition. For the secondary effluent, four major peaks at 1292, 824, 558 and 251 Da and three peaks at 1280, 931 and 777 Da were observed prior to and after laccase addition, respectively. The above observation indicated that the laccase polymerization shifted the molecular distribution to a higher range for both investigated wastewaters. The disappearance of constituents with smaller molecular weights would contribute the enhanced removal of COD by following membrane separation. Further integrated calculation of these chromatograms revealed that the raw wastewater and secondary effluent had the averaged molecular weights of 914 and 660 Da prior to laccase polymerization. The average molecular weight of raw wastewater was increased from 914 to 1210 Da, and that of the secondary effluent was increased from 660 to 970 Da. More increase in average molecular weight of the raw wastewater compared to that of the secondary effluent indicated that the conventional wastewater treatment did partially remove the organic compounds in pulp and paper wastewater. However, the 28% reduction in average molecular weight between the raw wastewater and the secondary effluent
also demonstrated the recalcitrance of the pulp and paper wastewater. 3.2. COD removal enhancement by laccase polymerization The COD of raw wastewater and secondary effluent was analyzed to be 785 and 142 mg/L, respectively. Figs. 2 and 3 present the COD variations along with time for raw wastewater and secondary effluent treated with 54,000, 30,000, 10,000 and 5000 MWCO membranes, respectively. For raw wastewater, no further decrease in COD concentration was observed after 6 h of filtration operation, indicating that most of the retaining of aqueous contaminants by membrane occurred during the first 6 h. To the contrary, the aqueous COD concentrations for the secondary effluent were found decreasing along with time during the entire 9-h filtration operation. This difference resulted from the molecular size difference of the contaminants in raw wastewater and secondary effluent. In raw wastewater, the aqueous contaminants had large molecular weights which blocked the pores of the membranes easily, resulting in the effective membrane separation only for the beginning few hours of operation. In the secondary effluent, the aqueous contaminants had relatively smaller molecular sizes, so the pore blocking by these contaminants in the membrane separation required longer operation time. The COD removal efficiencies for raw wastewater and secondary effluent using 4 investigated membranes were calculated and are presented in Fig. 4. Without laccase pretreatment, the efficiencies for COD removal from raw wastewater after 9 h filtration were calculated to be 47, 48, 53, and 55% using membranes with MWCO of 54,000, 30,000, 10,000, and 5000 Da, respectively. For the secondary effluent, the efficiencies were 11, 12, 20 and 25%, respec-
C.-H. Ko, C. Fan / Journal of Hazardous Materials 181 (2010) 763–770
767
Fig. 5. Flux reduction of raw wastewater for 54,000, 30,000, 10,000, and 5000 MWCO membranes.
tively. With laccase pretreatment, the COD removal efficiencies for raw wastewater were 58, 64, 68 and 71% using membranes with MWCO of 54,000, 30,000, 10,000, and 5000 Da, respectively. For the secondary effluent, the efficiencies were 32, 32, 36, and 49%, respectively. For the raw wastewater without laccase pretreatment, the higher removal efficiency was observed for the smaller MWCO membranes because they prevented the larger substances from penetrating effectively. The COD removal efficiency decreased as the MWCO of the membrane increased. For the secondary effluent, the similar trends on COD removal were observed, but with different removal efficiencies compared to those for the raw wastewater. Treated with laccase, the analyzed water samples contained more polymerized particles, resulting in the enhancement of COD removal. Using 5000 MWCO membrane as an example, the COD removal efficiencies of the filtration with laccase pretreatment were significantly increased to 71 and 49% for raw wastewater and secondary effluent, respectively. This COD removing enhancement was due to the increase of average molecular weight of the aqueous contaminants after laccase treatment. For comparison, the inactivated laccase was used to treat the water samples prior to membrane filtration and the results are shown in Figs. 2 and 3 as well. The COD removal efficiencies with inactivated laccase pretreatment for the four tested membrane were around 48–55 and 12–26% for raw wastewater and secondary effluent, respectively. No obvious difference in COD permeate concentrations was observed between filtrations with and without pretreatment by inactivated laccase for both raw wastewater and secondary effluent. This is due to the fact that (1) inactivated laccase can only polymerize the organic pollutants in wastewater slightly and somewhat
increase the average molecular weight, and (2) inactivated laccase is a small size protein able to permeate through the membranes to increase COD effluent concentration and reduce the overall COD removal efficiency. These two phenomena cancelled out each other, and the COD concentration in permeate remained fairly the same as that without treatment of inactivated laccase. Among the filtration results obtained in this study, better removal efficiency occurred when the water sample was pretreated with laccase followed by separation process using smaller MWCO membrane. Additionally, the total phenolic compounds in the raw wastewater and secondary effluent were analyzed to be 17.1 and 2 mg/L, respectively. Without laccase pretreatment, the phenolic removal for the raw wastewater by investigated membranes with MWCO of 54,000, 30,000, 10,000, and 5000 Da were 48, 48, 55 and 56%, respectively. For the secondary effluent, the respective removal efficiencies were calculated to be 15, 16, 23 and 24%. These phenolic removal efficiencies are similar to those for COD both for raw wastewater and for secondary effluent. With the laccase treatment prior to membrane filtration, no phenolic compound was observed in the permeates for both investigated wastewaters. This finding indicated that laccase polymerization is very effective in enhancing the removal of phenolic compounds in wastewater by membrane separation. 3.3. Permeate flux Generally, the membrane rejection is a function of the average molecular weight of the compounds in the filtrate. In this study, the average molecular weight of the polymerized substances in the wastewaters became the critical characteristic for COD removal. As the filtration process continued, the laccase polymerization in
768
C.-H. Ko, C. Fan / Journal of Hazardous Materials 181 (2010) 763–770
Fig. 6. Flux reduction of secondary effluent for 54,000, 30,000, 10,000, and 5000 MWCO membranes.
the pretreatment enhanced the pollutant removal but reduced the membrane permeability due to the gel layer formation at the surface of the membrane. In this study, the DI water, laccase solution, raw wastewater, secondary effluent, and the wastewaters treated with laccase and inactivated laccase were introduced into the 4 investigated membranes, and the flux variations along with time were measured and are presented in Figs. 5 and 6. The fluxes were declined in the beginning and slowly reached steady state without further flux decline after 180 min for all the tested aqueous samples, showing 12–41% flux reduction due to membrane fouling. Usually, the membrane with smaller pore size or treated with wastewater containing larger contaminants would be fouled more significantly due to more gel layer formation. In the membrane treatment for raw wastewater (Fig. 5), the flux for DI water treatment required longer time to reach equilibrium compared to the other tested wastewaters, either with or without enzyme pretreatment, implying that the constituents in wastewater would facilitate the membrane fouling process. After the equilibrium was reached, the flux for laccase solution was comparable to that for raw wastewater, while the flux for laccase-treated raw wastewater was comparable to that for raw wastewater with inactivated laccase pretreatment. Also,
the flux for raw wastewater was higher than that for laccasetreated raw wastewater because laccase polymerization in raw wastewater resulted in the membrane pore blocking and flux reduction. After treating with inactivated laccase, the flux for raw wastewater decreased more compared to that treated with laccase. This is because inactivated laccase protein may promote membrane fouling, and reduce the flux. The decline in flux demonstrated the inability of the membrane to remove COD (i.e., sole protein, inactivated laccase) from the raw wastewater by filtration. The concurrence of simultaneous COD removal and flux reduction for laccase-treated raw wastewater validated that enhanced separations were achieved by membrane retention of polymerized organic pollutants. After the equilibrium was reached, the COD mass fluxes for the tested membranes were calculated and the results are presented in Table 1. For a given membrane, the COD mass fluxes both for the wastewater and secondary effluent pretreated with laccase were smaller compared to those without laccase pretreatment. Meanwhile, the mass flux decrease was observed for inactivated laccase treatment as well but with a less extent. This observation also indicated that better removal efficiency occurred for the laccase-treated wastewater and secondary effluent since less COD
Table 1 COD mass flux (g/(h m2 )) for each tested membrane after equilibrium (180 min of operation). 54,000
Wastewater Laccase-treated wastewater Inactivated laccase-treated wastewater
30,000
10,000
5000
Raw
2nd
Raw
2nd
Raw
2nd
Raw
2nd
52.72 36.17 44.58
16.54 11.62 14.66
49.96 29.87 42.33
13.84 9.28 11.86
37.89 23.10 33.83
11.61 8.49 10.53
30.11 18.74 28.89
9.26 6.00 8.66
C.-H. Ko, C. Fan / Journal of Hazardous Materials 181 (2010) 763–770
769
Table 2 Apparent resistances and resistance increases (in parentheses) relative to pure water filtration for tested water samples after 180 min of operation. ×10−12 m−1
54,000 Raw
Water Laccase Wastewater Laccase-treated wastewater Inactivated laccase-treated wastewater
30,000
10,000
2nd
Raw
2nd
Raw
2nd
Raw
2nd
6.16 7.56 (1.40) 7.59 (1.43) 9.36 (3.20)
7.07 (0.91) 8.10 (1.94)
6.45 8.04 (1.59) 7.77 (1.32) 9.93 (3.48)
7.36 (0.91) 9.04 (2.59)
7.24 8.96 (1.72) 9.80 (2.56) 10.98 (3.74)
8.17 (0.93) 10.11 (2.87)
8.78 10.31 (1.53) 10.18 (1.40) 11.60 (2.82)
9.77 (0.99) 12.08 (3.30)
9.30 (3.14)
8.13 (1.97)
9.88 (3.43)
8.93 (2.48)
10.86 (3.62)
10.03 (2.79)
11.49 (2.71)
11.95 (3.17)
penetrated through the membrane. For the tested wastewaters, the COD mass flux decreased as the pore size of the membrane decreased, which, again, supported the previous finding that the membrane with smaller pore size would result in better treatment efficiency. 3.4. Magnitude of apparent membrane resistances Theoretically, the flux loss during membrane filtration occurs though various mechanisms, which may be modeled phenomenologically using a resistance-in-series algorithm. This algorithm has been applied to the development of various models by considering a single hydraulic resistance by the polarized layer or incorporating several different fouling mechanisms [21–23]. Using Eq. (1) as an example, three different resistances were considered, and Jv is flux through the membrane (cm/s), P is trans-membrane pressure (Pa), is dynamic viscosity, rm is membrane hydraulic resistance, rs is substrate resistance and rg is gel layer resistance. Jv =
P (rm + rs + rg )
(1)
In the membrane application, all the different types of resistances may occur simultaneously, and quantification of these resistances requires a series of filtration experiments. Since the overall COD removing performance of the membrane is the major concern of this study, the apparent resistance (rapp ) of the membrane was calculated after 180 min operation, which is considered at steady state, using Eq. (2), Jv =
P rapp
5000
(2)
The calculated apparent resistances for each tested membrane treating various water samples are presented in Table 2. Using the resistance of pure water as the reference, the apparent resistance increases for all the tested membranes treating with laccase solution and various wastewaters are shown in Table 2 as well. For all the tested water and wastewater samples, the apparent resistance increased as the pore size of the membrane decreased. For a given membrane, the apparent resistance for pure water filtration was the least, and the maximum apparent resistance occurred when the raw wastewater was treated with laccase. In general, the existence of organic constituents in the water and wastewater sample increased its respective apparent resistance. For the laccase solution, the apparent resistance increased because the enzyme in the solution also blocks the pores. However, the comparison among apparent resistance increases for laccase solution, raw wastewater, laccase-treated raw wastewater, and inactivated laccase-treated raw wastewater revealed that the laccase enzyme, inactivated enzyme, and contaminants in the raw wastewater have different abilities to block the membrane pores, and the blocking effects were not additive if two or more organic constituents coexisted in the aqueous sample at the same time. Similar trends were also observed for the secondary effluent. For the treatment of raw
wastewater using 54,000 MWCO membrane, the organic contaminants were polymerized with laccase resulting the most increase in apparent resistance. The addition of inactivated laccase to the raw wastewater resulted in a slightly lower increase in apparent resistance compared to that for laccase-treated raw wastewater. As for the filtration for secondary effluent, less increase in apparent resistance was observed compared to that of raw wastewater because most of the organic contaminants were removed during the secondary wastewater treatment process. Among all the tested membranes except for the 5000 MWCO membrane, the apparent resistances for raw wastewater with either laccase or inactivated laccase pretreatment were greater than those for the secondary effluent. The exception of 5000 MWCO was possibly because the smaller membrane pore size did not allow sole protein and organic contaminants to penetrate easily, and addition of laccase or inactivated laccase only resulted in the more contamination of water sample and higher apparent resistance. 4. Conclusions The result from the present study showed the enhanced COD removal using laccase polymerization and membrane filtration combination to treat the pulp and paper wastewater. With 2.98 IU/L of activated laccase applied at room temperature for 60 min, the averaged molecular weight of raw wastewater was increased from 914 to 1210 Da, and that of the secondary effluent was increased from 660 to 970 Da. With laccase polymerization prior to filtration, over 60% COD removal by the four investigated membranes were observed. Moreover, the addition of laccase reduced 4–14% of membrane permeability after 180 min of filtration due to gel layer formation by the polymerized organic pollutants, and no further decline in flux was observed as the filtration continued. The maximum apparent resistance occurred when the sample was treated with laccase prior to membrane filtration. The combination of laccase polymerization and membrane processes was able to treat the pulp and paper wastewater. Pretreatments by inactivated laccase only caused further flux reduction without additional removal of COD. References [1] D. Pokhrel, T. Viraraghavan, Treatment of pulp and paper mill wastewater—a review, Sci. Total Environ. 333 (2004) 37–58. [2] A. Singhal, I.S. Thakur, Decolourization and detoxification of pulp and paper mill effluent by Emericella nidulans var. nidulans, J. Hazard. Mater. 171 (2009) 619–625. [3] M. Xu, Q. Wang, Y. Hao, Removal of organic carbon from wastepaper pulp effluent by lab-scale solar photo-Fenton process, J. Hazard. Mater. 148 (2007) 103–109. [4] R. Borja, A. Martin, R. Maestro, J. Alba, J.A. Fiestas, Enhancement of the anaerobic digestion of olive mill wastewater by the removal of phenolic inhibitors, Process Biochem. 27 (1992) 231–237. [5] T. Hsien, Y. Lin, Biodegradation of phenolic wastewater in a fixed biofilm reactor, Biochem. Eng. J. 27 (2005) 95–103. [6] E.C. Catalkaya, F. Kargi, Color, TOC and AOX removals from pulp mill effluent by advanced oxidation processes: a comparative study, J. Hazard. Mater. B139 (2007) 244–253.
770
C.-H. Ko, C. Fan / Journal of Hazardous Materials 181 (2010) 763–770
[7] S.S. Wong, T.T. Teng, A.L. Ahmada, A. Zuhairi, G. Najafpour, Treatment of pulp and paper mill wastewater by polyacrylamide (PAM) in polymer induced flocculation, J. Hazard. Mater. B135 (2006) 378–388. [8] T.W. Joyce, H. Chang, A.G. Campbell Jr., E.D. Gerrard, T.K. Kirk, A continuous biological process to decolorize bleach plant effluents, Biotechnol. Adv. 2 (1984) 301–308. [9] V.A. Cooper, J.A. Nicell, Removal of phenols from a foundry wastewater using horseradish peroxidase, Water Res. 30 (1996) 954–964. [10] T. Tanaka, M. Nose, A. Endo, T. Fujii, M. Taniguchi, Treatment of nonylphenol with laccase in a rotating reactor, J. Biosci. Bioeng. 96 (2003) 541–546. [11] M. Alppisara, R. Lantto, H. Manner, J. Buchert, Use of enzymes for treatment of internal purification concentrates of pulp and paper industry, in: 223rd ACS National Meeting, American Chemical Society, Orlando, FL, USA, 2002. [12] S. Davis, R.G. Burns, Decolorization of phenolic effluents by soluble and mmobilized phenol oxidases, Appl. Microbiol. Biot. 32 (1990) 721–726. [13] M. Luisa, F.C. Gonlves, W. Steiner, Enzymes for pulp and paper processing, in: ACS Symposium Series, vol. 655, American Chemical Society, Washington, DC, 1996. [14] C.F. Thurston, The structure and function of fungal laccases, Microbiology 140 (1994) 19–26. [15] G.M.B. Soares, M. Costa-Ferreira, M.T. Pessoa de Amorim, Decolorization of an anthraquinone-type dye using a laccase formulation, Bioresource Technol. 79 (2001) 171–177.
[16] N. Durán, E. Esposito, Potential applications of oxidative enzymes and phenoloxidase-like compounds in wastewater and soil treatment: a review, Appl. Catal. B: Environ. 28 (2000) 83–99. [17] N. Durán, M.A. Rosa, A. D’Annibale, L. Gianfreda, Applications of laccases and tyrosinases (phenoloxidases) immobilized on different supports: a review, Enzyme Microb. Tech. 31 (2002) 907–931. [18] C. Ko, S. Chen, Enhanced removal of three phenols by laccase polymerization with MF/UF membranes, Bioresource Technol. 99 (2008) 2293–2298. [19] Standard Methods for the Examination of Water and Wastewater, APHA/AWWA/WEF, Washington, DC, 2005. [20] T.S. Kujala, J.M. Loponen, K.D. Klika, K. Pihlaja, Phenolics and betacyanins in red beetroot (Beta vulgaris) root: distribution and effect of cold storage on the content of total phenolics and three individual compounds, J. Agric. Food Chem. 48 (2000) 5338–5342. [21] J. Cho, G. Amy, Y. Yoon, J. Sohn, Predictive models and factors affecting natural organic matter (NOM) rejection and flux decline in ultrafiltration (UF) membrane, Desalination 142 (2002) 245–255. [22] H. Choi, K. Zhang, D.D. Dionysiou, D.B. Oerther, G.A. Sorial, Influence of cross-flow velocity on membrane performance during filtration of biological suspension, J. Membrane Sci. 248 (2005) 189–199. [23] X. Wang, L. Wang, Y. Liu, W. Duan, Ozonation pretreatment for ultrafiltration of the secondary effluent, J. Membrane Sci. 287 (2007) 187–191.
Contents lists available at ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Enhanced chemical oxygen demand removal and flux reduction in pulp and paper wastewater treatment using laccase-polymerized membrane filtration Chun-Han Ko a , Chihhao Fan b,∗ a b
School of Forest and Resources Conservation, National Taiwan University, No. 1, Sec. 4, Roosevelt Rd, Da-An District, Taipei 10617, Taiwan Department of Safety, Health, and Environmental Engineering, Ming Chi University of Technology, 84 Gung-Juan Rd, Taishan 24301, Taipei County, Taiwan
a r t i c l e
i n f o
Article history: Received 30 January 2010 Received in revised form 16 May 2010 Accepted 17 May 2010 Available online 25 May 2010 Keywords: Chemical oxygen demand Enzymatic degradation Permeate flux Pulp and paper Polymerization
a b s t r a c t The purpose of this present study is to investigate the removal efficiency of chemical oxygen demand (COD) from pulp and paper wastewater using laccase-polymerized membrane filtration process. The membranes with molecular weight cut-off (MWCO) of 5000 and 10,000, 30,000 and 54,000 were used in a cross-flow module to treat the pulp and paper wastewater containing high phenolic constituents and COD. With 2.98 IU/L of activated laccase applied at room temperature for 180 min, the contaminants in raw wastewater and second effluent were polymerized to form larger molecules with average molecular weight of 1300 and 900 Da (Dalton), respectively. With laccase polymerization prior to filtration, over 60% removals of COD by the four investigated membranes were observed, compared with low COD removal without laccase polymerization. Moreover, the addition of laccase resulted in 4–14% reduction of membrane permeability during the first 180 min filtration operation due to gel layer formation by the polymerization. No further flux decline was observed afterwards indicating the steady state was reached and the membranes could be used to remove the polymerized pollutants without significant fouling. The maximum apparent resistance occurrence for raw wastewater treated with laccase also supported the effectiveness for COD removal with laccase polymerization before membrane filtration. Additionally, pretreatment by inactivated laccase only caused further flux reduction without additional removal of COD. © 2010 Elsevier B.V. All rights reserved.
1. Introduction The pulp and paper industry has close relationship to the economic development, and the increasing amount of paper utilization makes the pulp and paper industry even more important in the process of societal modernization. Unfortunately, pulp and paper mills generate a considerable amount of wastewater containing various types of contaminants [1–3]. In the literature, it has been reported that high concentrations of chlorinated phenolic acids and chemical oxygen demand (COD) are a basic characteristic of pulp and paper wastewater [1]. These phenolic compounds are recalcitrant in the environment, difficult to biodegrade and detrimental to the public health. Also, the dark brown colour of the wastewater due to the presence of abundant partially oxidized organic lignins inhibits the natural process of photosynthesis, and many times, results in unacceptable odourous problems [1,4–7]. Without appropriate treatment before discharge, the pulp and paper wastewater becomes a severe threat to the aquatic environment as well as the riverine ecology.
∗ Corresponding author. Tel.: +886 229089899x4656; fax: +886 229080783. E-mail address: [email protected] (C. Fan). 0304-3894/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jhazmat.2010.05.079
In order to protect the aquatic environment from being polluted by pulp and paper wastewater, many contaminant-removing facilities were installed to improve the effluent water quality of pulp and paper mills. The frequent-used conventional treatment units to remove COD include sedimentation and floatation, coagulation and precipitation, adsorption, chemical oxidation, and so many. Although many physical and chemical methods are quite effective in treating pulp and paper wastewater, much effort still has been expended in the innovation of wastewater treatment technology due to the high capital and energy costs and instability in operation of the conventional treatment technologies [8]. Besides, the elevated living standard requests by the public and the increasingly stringent effluent discharge standards have brought on the inevitable advancement in wastewater treatment efficiency for pulp and paper mills. There has been an increasing popularity in membrane technology applied to treating wastewater from hardwood kraft mills over the past two decades because membrane processes offer a high level of contaminant separation and relatively low energy consumption. Pressure-driven membrane processes, including reverse osmosis, nanofiltration, ultrafiltration and microfiltration, are of a great deal of interests because they may prevent a wide range of small particles from penetrating membranes, depending on the
764
C.-H. Ko, C. Fan / Journal of Hazardous Materials 181 (2010) 763–770
Fig. 1. The polystyrene sulfonate standards (straight lines) at 2220, 910, and 220 Da. and the average molecular weights of raw wastewater and the secondary effluent before (dash-dotted line) and after (solid line) laccase addition.
pore size or molecular weight cut-off (MWCO) of the membrane. Recently enzymatic polymerization has been demonstrated to be able to effectively treat the wastewater containing toxic phenolic compounds through physical separation [9,10], although other researches indicated the contaminants from pulp and paper mills might be removed or degraded through different enzymatic reaction mechanisms [11–13]. Enzymes may catalyze the oxidation of phenolic substances and certain amines while reducing oxygen to water [14]. In the literature, it has been reported that adding laccase with 10 IU/mL to wastewater may facilitate the removal of phenolic compounds in pulp mill effluents by inducing cross-linking and precipitation among contaminants [15]. Laccase is a blue copper containing polyphenol oxidase which plays an important role both in lignin biosynthesis and in biodegradation [14]. Compared to other enzymes, laccase is rather unspecific and suitable for use in wide range of applications, such as pulp deligninfication, wastewater detoxification and denim decolourization [16,17]. A successful application of laccase-assisted membrane separation technology to treating various synthetic wastewaters each containing one selected phenol has been reported [18]. In this present study, laccase with the activity of 2.98 IU/L (i.e., 0.3 g protein/L reaction solution) was introduced to polymerize the organic contaminants in the pulp and paper wastewater. The COD removal efficiencies of the resulting polymerized wastewater were evaluated by cross-flow filtration using 4 different membranes with the MWCO of 5000, 10,000, 30,000, and 54,000 Da, respectively. The raw wastewater and secondary effluent of a consolidated kraft pulping and fine grade papermaking mill were used. The flux penetrating each tested membrane at pH value of 7.5 and room temperature was analyzed and the apparent resistance was calculated as well.
the Danish company of Novozyme. The membranes were obtained from GE Osmonics. The membranes of 5000 and 10,000 Da MWCO were made of polyethersulfone. The membranes of 30,000 and 54,000 Da MWCO were made of polysulfone.
2. Materials and methods
2.3. Gel permeation chromatography (GPC)
2.1. Wastewater samples and materials
The molecular weights of the phenolic compounds were analyzed by gel permeation chromatography (GPC). The GPC is operated by the principle of size exclusion, in which target molecules are separated based on their difference in sizes. The UV absorbing organics were eluted in the descending orders of molecular weights, i.e., organics with larger molecular weights were eluted at earlier times since they have difficulties to migrate into the finer micropore structure of column packing materials. The system used in this study consisted of a Jasco 880 HPLC PU pump, a Jasco 875 UV detector, connected to a guard column of Shodex SBG, followed by Shodex SB-802 HQ and SB-803 HQ 8 mm × 300 mm columns (Showa Denko K.K., Japan) in series. The serial columns used were of polyhydroxymethacrylate packing gels with following molecular weight exclusion limits of eluent conditions: 4000 and
The wastewater samples used were the raw wastewater and secondary effluent collected from a kraft pulp and paper mill operating at 47,000 CMD in eastern Taiwan. The wastewater was treated through a series of physical/chemical/biological units of stabilization, coagulation, flocculation, sedimentation, aeration followed by activated sludge, and acidification before being discharged. All the collected wastewater samples were filtered by 0.5 m PE membrane prior to laccase pretreatment and membrane filtration. For each wastewater sample, the pH was adjusted to 7.5 using 5 mM of phosphate buffer. All the chemicals used were of reagent grade purchased from Sigma. The water used was deionized (DI) water. The enzyme used in this study was Laccase 51003 obtained from
2.2. Enzymatic polymerization In pretreatment process, the laccase was added into the water sample and stirred for 1 h, resulting in the activity of 2.98 IU/L prior to filtration tests. The activity of laccase was measured by the rate of oxidation of syringaldazine. One international unit (IU) is defined as the amount of enzyme which oxidizes 1 milli molar (mM) syringaldazine per minute under the condition of pH 7.5 and 30 ◦ C. After 60 min of reaction, 4% (w/v) sodium azide solution, resulting in the total concentration of 0.004%, was added to stop the enzymatic reaction. The average molecular weights of the constituents in treated samples were monitored at the reaction times 1 and 3 h. The monitoring result showed that most of the polymerization occurred during the first 1 h. Therefore, the samples for subsequent molecular weight analysis were collected after 1 h polymerization reaction. For additional comparison, the laccase was inactivated by sterilization at 120 ◦ C for 2 h, and the resulting inactivated enzyme solution was added to water samples to evaluate its impact on COD removal from pulp and paper wastewater as well. The overall laccase activity in the stock solution was 42,973 IU/mL, with the COD concentration of 30,800 mg/L. The protein concentration of laccase stock solution employed in the present work was 4320 mg/L. Since laccase was added in reaction mixture at the dosage of 2.98 IU/L, the contributions for both COD (i.e., 0.0021 mg/L) and protein mass loadings (i.e., 0.0003 mg/L) from laccase addition were assumed negligible.
C.-H. Ko, C. Fan / Journal of Hazardous Materials 181 (2010) 763–770
Fig. 2. COD variations for raw wastewater treated with 54,000, 30,000, 10,000, and 5000 MWCO membranes.
Fig. 3. COD variations for secondary effluent treated with 54,000, 30,000, 10,000, and 5000 MWCO membranes.
765
766
C.-H. Ko, C. Fan / Journal of Hazardous Materials 181 (2010) 763–770
100,000 Da, respectively. The 0.005 M K2 HPO4 and DI water were used as the mobile phases for the raw wastewater and secondary effluent, respectively. The flow rate was controlled at 0.8 mL/min. The molecular weight distribution was monitored by a UV detector at the wavelength of 280 nm. For system calibration, the aqueous polystyrene sulfonate standards purchased from Polymer Standards Service GmbH, Mainz, Germany were used. These standards with molecular weights of 2220, 910 and 220 Da were eluted at 18.58, 22.42 and 25.82 min, respectively. The molecular weights were regressed with the obtained elution times, and the resulting calibration was employed as the reference for molecular weight estimation in the subsequent analysis using chromatography manager software [18]. The COD removal efficiency was calculated taking the COD concentration difference between influent and effluent divided by influent COD concentration. 2.4. Membrane filtration The cross-flow membrane filtration module used in this study consisted of a liquid reservoir, a gear pump (Cole-Parmer Model 75211-10) and a flat sheet plate with filtration area of 0.0084 m2 (Chi-Guang, Taiwan). For each tested wastewater, two samples were pretreated with laccase and inactivated laccase, respectively, prior to the filtration experiment. Additionally, another wastewater sample was used for filtration experiments without further pretreatment. Each filtration experiment was conducted at the operation pressure of 45 psi (3 kg/cm2 ) under room temperature (22 ± 2 ◦ C) for 9 h. Aqueous samples were collected in duplicate and the COD concentration was determined according to the standard methods [19]. The total phenols contents were estimated with a spectrophotometer method based on the Folin and Ciocalteu reagent [20].
Fig. 4. COD removal efficiencies of (a) raw wastewater and (b) the secondary effluent for the membrane MWCO at 5000, 10,000, 30,000, and 54,000.
3. Results and discussion 3.1. Molecular weights increase by laccase polymerization Fig. 1 shows the chromatograms of eluted average molecular weights of the raw wastewater and secondary effluent before and after the addition of laccase. The empty boxes are the responses of selected standards at their respective elution times. The dotted and solid curves are the chromatographic graphs for tested water samples without and after laccase pretreatment, respectively. For raw wastewater, four major peaks were observed at 1464, 1107, 590 and 304 Da, and three peaks were observed at 1711, 1344 and 795 Da after laccase addition. For the secondary effluent, four major peaks at 1292, 824, 558 and 251 Da and three peaks at 1280, 931 and 777 Da were observed prior to and after laccase addition, respectively. The above observation indicated that the laccase polymerization shifted the molecular distribution to a higher range for both investigated wastewaters. The disappearance of constituents with smaller molecular weights would contribute the enhanced removal of COD by following membrane separation. Further integrated calculation of these chromatograms revealed that the raw wastewater and secondary effluent had the averaged molecular weights of 914 and 660 Da prior to laccase polymerization. The average molecular weight of raw wastewater was increased from 914 to 1210 Da, and that of the secondary effluent was increased from 660 to 970 Da. More increase in average molecular weight of the raw wastewater compared to that of the secondary effluent indicated that the conventional wastewater treatment did partially remove the organic compounds in pulp and paper wastewater. However, the 28% reduction in average molecular weight between the raw wastewater and the secondary effluent
also demonstrated the recalcitrance of the pulp and paper wastewater. 3.2. COD removal enhancement by laccase polymerization The COD of raw wastewater and secondary effluent was analyzed to be 785 and 142 mg/L, respectively. Figs. 2 and 3 present the COD variations along with time for raw wastewater and secondary effluent treated with 54,000, 30,000, 10,000 and 5000 MWCO membranes, respectively. For raw wastewater, no further decrease in COD concentration was observed after 6 h of filtration operation, indicating that most of the retaining of aqueous contaminants by membrane occurred during the first 6 h. To the contrary, the aqueous COD concentrations for the secondary effluent were found decreasing along with time during the entire 9-h filtration operation. This difference resulted from the molecular size difference of the contaminants in raw wastewater and secondary effluent. In raw wastewater, the aqueous contaminants had large molecular weights which blocked the pores of the membranes easily, resulting in the effective membrane separation only for the beginning few hours of operation. In the secondary effluent, the aqueous contaminants had relatively smaller molecular sizes, so the pore blocking by these contaminants in the membrane separation required longer operation time. The COD removal efficiencies for raw wastewater and secondary effluent using 4 investigated membranes were calculated and are presented in Fig. 4. Without laccase pretreatment, the efficiencies for COD removal from raw wastewater after 9 h filtration were calculated to be 47, 48, 53, and 55% using membranes with MWCO of 54,000, 30,000, 10,000, and 5000 Da, respectively. For the secondary effluent, the efficiencies were 11, 12, 20 and 25%, respec-
C.-H. Ko, C. Fan / Journal of Hazardous Materials 181 (2010) 763–770
767
Fig. 5. Flux reduction of raw wastewater for 54,000, 30,000, 10,000, and 5000 MWCO membranes.
tively. With laccase pretreatment, the COD removal efficiencies for raw wastewater were 58, 64, 68 and 71% using membranes with MWCO of 54,000, 30,000, 10,000, and 5000 Da, respectively. For the secondary effluent, the efficiencies were 32, 32, 36, and 49%, respectively. For the raw wastewater without laccase pretreatment, the higher removal efficiency was observed for the smaller MWCO membranes because they prevented the larger substances from penetrating effectively. The COD removal efficiency decreased as the MWCO of the membrane increased. For the secondary effluent, the similar trends on COD removal were observed, but with different removal efficiencies compared to those for the raw wastewater. Treated with laccase, the analyzed water samples contained more polymerized particles, resulting in the enhancement of COD removal. Using 5000 MWCO membrane as an example, the COD removal efficiencies of the filtration with laccase pretreatment were significantly increased to 71 and 49% for raw wastewater and secondary effluent, respectively. This COD removing enhancement was due to the increase of average molecular weight of the aqueous contaminants after laccase treatment. For comparison, the inactivated laccase was used to treat the water samples prior to membrane filtration and the results are shown in Figs. 2 and 3 as well. The COD removal efficiencies with inactivated laccase pretreatment for the four tested membrane were around 48–55 and 12–26% for raw wastewater and secondary effluent, respectively. No obvious difference in COD permeate concentrations was observed between filtrations with and without pretreatment by inactivated laccase for both raw wastewater and secondary effluent. This is due to the fact that (1) inactivated laccase can only polymerize the organic pollutants in wastewater slightly and somewhat
increase the average molecular weight, and (2) inactivated laccase is a small size protein able to permeate through the membranes to increase COD effluent concentration and reduce the overall COD removal efficiency. These two phenomena cancelled out each other, and the COD concentration in permeate remained fairly the same as that without treatment of inactivated laccase. Among the filtration results obtained in this study, better removal efficiency occurred when the water sample was pretreated with laccase followed by separation process using smaller MWCO membrane. Additionally, the total phenolic compounds in the raw wastewater and secondary effluent were analyzed to be 17.1 and 2 mg/L, respectively. Without laccase pretreatment, the phenolic removal for the raw wastewater by investigated membranes with MWCO of 54,000, 30,000, 10,000, and 5000 Da were 48, 48, 55 and 56%, respectively. For the secondary effluent, the respective removal efficiencies were calculated to be 15, 16, 23 and 24%. These phenolic removal efficiencies are similar to those for COD both for raw wastewater and for secondary effluent. With the laccase treatment prior to membrane filtration, no phenolic compound was observed in the permeates for both investigated wastewaters. This finding indicated that laccase polymerization is very effective in enhancing the removal of phenolic compounds in wastewater by membrane separation. 3.3. Permeate flux Generally, the membrane rejection is a function of the average molecular weight of the compounds in the filtrate. In this study, the average molecular weight of the polymerized substances in the wastewaters became the critical characteristic for COD removal. As the filtration process continued, the laccase polymerization in
768
C.-H. Ko, C. Fan / Journal of Hazardous Materials 181 (2010) 763–770
Fig. 6. Flux reduction of secondary effluent for 54,000, 30,000, 10,000, and 5000 MWCO membranes.
the pretreatment enhanced the pollutant removal but reduced the membrane permeability due to the gel layer formation at the surface of the membrane. In this study, the DI water, laccase solution, raw wastewater, secondary effluent, and the wastewaters treated with laccase and inactivated laccase were introduced into the 4 investigated membranes, and the flux variations along with time were measured and are presented in Figs. 5 and 6. The fluxes were declined in the beginning and slowly reached steady state without further flux decline after 180 min for all the tested aqueous samples, showing 12–41% flux reduction due to membrane fouling. Usually, the membrane with smaller pore size or treated with wastewater containing larger contaminants would be fouled more significantly due to more gel layer formation. In the membrane treatment for raw wastewater (Fig. 5), the flux for DI water treatment required longer time to reach equilibrium compared to the other tested wastewaters, either with or without enzyme pretreatment, implying that the constituents in wastewater would facilitate the membrane fouling process. After the equilibrium was reached, the flux for laccase solution was comparable to that for raw wastewater, while the flux for laccase-treated raw wastewater was comparable to that for raw wastewater with inactivated laccase pretreatment. Also,
the flux for raw wastewater was higher than that for laccasetreated raw wastewater because laccase polymerization in raw wastewater resulted in the membrane pore blocking and flux reduction. After treating with inactivated laccase, the flux for raw wastewater decreased more compared to that treated with laccase. This is because inactivated laccase protein may promote membrane fouling, and reduce the flux. The decline in flux demonstrated the inability of the membrane to remove COD (i.e., sole protein, inactivated laccase) from the raw wastewater by filtration. The concurrence of simultaneous COD removal and flux reduction for laccase-treated raw wastewater validated that enhanced separations were achieved by membrane retention of polymerized organic pollutants. After the equilibrium was reached, the COD mass fluxes for the tested membranes were calculated and the results are presented in Table 1. For a given membrane, the COD mass fluxes both for the wastewater and secondary effluent pretreated with laccase were smaller compared to those without laccase pretreatment. Meanwhile, the mass flux decrease was observed for inactivated laccase treatment as well but with a less extent. This observation also indicated that better removal efficiency occurred for the laccase-treated wastewater and secondary effluent since less COD
Table 1 COD mass flux (g/(h m2 )) for each tested membrane after equilibrium (180 min of operation). 54,000
Wastewater Laccase-treated wastewater Inactivated laccase-treated wastewater
30,000
10,000
5000
Raw
2nd
Raw
2nd
Raw
2nd
Raw
2nd
52.72 36.17 44.58
16.54 11.62 14.66
49.96 29.87 42.33
13.84 9.28 11.86
37.89 23.10 33.83
11.61 8.49 10.53
30.11 18.74 28.89
9.26 6.00 8.66
C.-H. Ko, C. Fan / Journal of Hazardous Materials 181 (2010) 763–770
769
Table 2 Apparent resistances and resistance increases (in parentheses) relative to pure water filtration for tested water samples after 180 min of operation. ×10−12 m−1
54,000 Raw
Water Laccase Wastewater Laccase-treated wastewater Inactivated laccase-treated wastewater
30,000
10,000
2nd
Raw
2nd
Raw
2nd
Raw
2nd
6.16 7.56 (1.40) 7.59 (1.43) 9.36 (3.20)
7.07 (0.91) 8.10 (1.94)
6.45 8.04 (1.59) 7.77 (1.32) 9.93 (3.48)
7.36 (0.91) 9.04 (2.59)
7.24 8.96 (1.72) 9.80 (2.56) 10.98 (3.74)
8.17 (0.93) 10.11 (2.87)
8.78 10.31 (1.53) 10.18 (1.40) 11.60 (2.82)
9.77 (0.99) 12.08 (3.30)
9.30 (3.14)
8.13 (1.97)
9.88 (3.43)
8.93 (2.48)
10.86 (3.62)
10.03 (2.79)
11.49 (2.71)
11.95 (3.17)
penetrated through the membrane. For the tested wastewaters, the COD mass flux decreased as the pore size of the membrane decreased, which, again, supported the previous finding that the membrane with smaller pore size would result in better treatment efficiency. 3.4. Magnitude of apparent membrane resistances Theoretically, the flux loss during membrane filtration occurs though various mechanisms, which may be modeled phenomenologically using a resistance-in-series algorithm. This algorithm has been applied to the development of various models by considering a single hydraulic resistance by the polarized layer or incorporating several different fouling mechanisms [21–23]. Using Eq. (1) as an example, three different resistances were considered, and Jv is flux through the membrane (cm/s), P is trans-membrane pressure (Pa), is dynamic viscosity, rm is membrane hydraulic resistance, rs is substrate resistance and rg is gel layer resistance. Jv =
P (rm + rs + rg )
(1)
In the membrane application, all the different types of resistances may occur simultaneously, and quantification of these resistances requires a series of filtration experiments. Since the overall COD removing performance of the membrane is the major concern of this study, the apparent resistance (rapp ) of the membrane was calculated after 180 min operation, which is considered at steady state, using Eq. (2), Jv =
P rapp
5000
(2)
The calculated apparent resistances for each tested membrane treating various water samples are presented in Table 2. Using the resistance of pure water as the reference, the apparent resistance increases for all the tested membranes treating with laccase solution and various wastewaters are shown in Table 2 as well. For all the tested water and wastewater samples, the apparent resistance increased as the pore size of the membrane decreased. For a given membrane, the apparent resistance for pure water filtration was the least, and the maximum apparent resistance occurred when the raw wastewater was treated with laccase. In general, the existence of organic constituents in the water and wastewater sample increased its respective apparent resistance. For the laccase solution, the apparent resistance increased because the enzyme in the solution also blocks the pores. However, the comparison among apparent resistance increases for laccase solution, raw wastewater, laccase-treated raw wastewater, and inactivated laccase-treated raw wastewater revealed that the laccase enzyme, inactivated enzyme, and contaminants in the raw wastewater have different abilities to block the membrane pores, and the blocking effects were not additive if two or more organic constituents coexisted in the aqueous sample at the same time. Similar trends were also observed for the secondary effluent. For the treatment of raw
wastewater using 54,000 MWCO membrane, the organic contaminants were polymerized with laccase resulting the most increase in apparent resistance. The addition of inactivated laccase to the raw wastewater resulted in a slightly lower increase in apparent resistance compared to that for laccase-treated raw wastewater. As for the filtration for secondary effluent, less increase in apparent resistance was observed compared to that of raw wastewater because most of the organic contaminants were removed during the secondary wastewater treatment process. Among all the tested membranes except for the 5000 MWCO membrane, the apparent resistances for raw wastewater with either laccase or inactivated laccase pretreatment were greater than those for the secondary effluent. The exception of 5000 MWCO was possibly because the smaller membrane pore size did not allow sole protein and organic contaminants to penetrate easily, and addition of laccase or inactivated laccase only resulted in the more contamination of water sample and higher apparent resistance. 4. Conclusions The result from the present study showed the enhanced COD removal using laccase polymerization and membrane filtration combination to treat the pulp and paper wastewater. With 2.98 IU/L of activated laccase applied at room temperature for 60 min, the averaged molecular weight of raw wastewater was increased from 914 to 1210 Da, and that of the secondary effluent was increased from 660 to 970 Da. With laccase polymerization prior to filtration, over 60% COD removal by the four investigated membranes were observed. Moreover, the addition of laccase reduced 4–14% of membrane permeability after 180 min of filtration due to gel layer formation by the polymerized organic pollutants, and no further decline in flux was observed as the filtration continued. The maximum apparent resistance occurred when the sample was treated with laccase prior to membrane filtration. The combination of laccase polymerization and membrane processes was able to treat the pulp and paper wastewater. Pretreatments by inactivated laccase only caused further flux reduction without additional removal of COD. References [1] D. Pokhrel, T. Viraraghavan, Treatment of pulp and paper mill wastewater—a review, Sci. Total Environ. 333 (2004) 37–58. [2] A. Singhal, I.S. Thakur, Decolourization and detoxification of pulp and paper mill effluent by Emericella nidulans var. nidulans, J. Hazard. Mater. 171 (2009) 619–625. [3] M. Xu, Q. Wang, Y. Hao, Removal of organic carbon from wastepaper pulp effluent by lab-scale solar photo-Fenton process, J. Hazard. Mater. 148 (2007) 103–109. [4] R. Borja, A. Martin, R. Maestro, J. Alba, J.A. Fiestas, Enhancement of the anaerobic digestion of olive mill wastewater by the removal of phenolic inhibitors, Process Biochem. 27 (1992) 231–237. [5] T. Hsien, Y. Lin, Biodegradation of phenolic wastewater in a fixed biofilm reactor, Biochem. Eng. J. 27 (2005) 95–103. [6] E.C. Catalkaya, F. Kargi, Color, TOC and AOX removals from pulp mill effluent by advanced oxidation processes: a comparative study, J. Hazard. Mater. B139 (2007) 244–253.
770
C.-H. Ko, C. Fan / Journal of Hazardous Materials 181 (2010) 763–770
[7] S.S. Wong, T.T. Teng, A.L. Ahmada, A. Zuhairi, G. Najafpour, Treatment of pulp and paper mill wastewater by polyacrylamide (PAM) in polymer induced flocculation, J. Hazard. Mater. B135 (2006) 378–388. [8] T.W. Joyce, H. Chang, A.G. Campbell Jr., E.D. Gerrard, T.K. Kirk, A continuous biological process to decolorize bleach plant effluents, Biotechnol. Adv. 2 (1984) 301–308. [9] V.A. Cooper, J.A. Nicell, Removal of phenols from a foundry wastewater using horseradish peroxidase, Water Res. 30 (1996) 954–964. [10] T. Tanaka, M. Nose, A. Endo, T. Fujii, M. Taniguchi, Treatment of nonylphenol with laccase in a rotating reactor, J. Biosci. Bioeng. 96 (2003) 541–546. [11] M. Alppisara, R. Lantto, H. Manner, J. Buchert, Use of enzymes for treatment of internal purification concentrates of pulp and paper industry, in: 223rd ACS National Meeting, American Chemical Society, Orlando, FL, USA, 2002. [12] S. Davis, R.G. Burns, Decolorization of phenolic effluents by soluble and mmobilized phenol oxidases, Appl. Microbiol. Biot. 32 (1990) 721–726. [13] M. Luisa, F.C. Gonlves, W. Steiner, Enzymes for pulp and paper processing, in: ACS Symposium Series, vol. 655, American Chemical Society, Washington, DC, 1996. [14] C.F. Thurston, The structure and function of fungal laccases, Microbiology 140 (1994) 19–26. [15] G.M.B. Soares, M. Costa-Ferreira, M.T. Pessoa de Amorim, Decolorization of an anthraquinone-type dye using a laccase formulation, Bioresource Technol. 79 (2001) 171–177.
[16] N. Durán, E. Esposito, Potential applications of oxidative enzymes and phenoloxidase-like compounds in wastewater and soil treatment: a review, Appl. Catal. B: Environ. 28 (2000) 83–99. [17] N. Durán, M.A. Rosa, A. D’Annibale, L. Gianfreda, Applications of laccases and tyrosinases (phenoloxidases) immobilized on different supports: a review, Enzyme Microb. Tech. 31 (2002) 907–931. [18] C. Ko, S. Chen, Enhanced removal of three phenols by laccase polymerization with MF/UF membranes, Bioresource Technol. 99 (2008) 2293–2298. [19] Standard Methods for the Examination of Water and Wastewater, APHA/AWWA/WEF, Washington, DC, 2005. [20] T.S. Kujala, J.M. Loponen, K.D. Klika, K. Pihlaja, Phenolics and betacyanins in red beetroot (Beta vulgaris) root: distribution and effect of cold storage on the content of total phenolics and three individual compounds, J. Agric. Food Chem. 48 (2000) 5338–5342. [21] J. Cho, G. Amy, Y. Yoon, J. Sohn, Predictive models and factors affecting natural organic matter (NOM) rejection and flux decline in ultrafiltration (UF) membrane, Desalination 142 (2002) 245–255. [22] H. Choi, K. Zhang, D.D. Dionysiou, D.B. Oerther, G.A. Sorial, Influence of cross-flow velocity on membrane performance during filtration of biological suspension, J. Membrane Sci. 248 (2005) 189–199. [23] X. Wang, L. Wang, Y. Liu, W. Duan, Ozonation pretreatment for ultrafiltration of the secondary effluent, J. Membrane Sci. 287 (2007) 187–191.