- Email: [email protected]
Biosorptive removal of bare-, citrate-, and PVP-coated silver nanoparticles from aqueous solution by activated sludge
Journal of Industrial and Engineering Chemistry 25 (2015) 51–55
Contents lists available at ScienceDirect
Journal of Industrial and Engineering Chemistry journal homepage: www.elsevier.com/locate/jiec
Biosorptive removal of bare-, citrate-, and PVP-coated silver nanoparticles from aqueous solution by activated sludge Seung Yeon Oh, Hwa Kyung Sung, Chulhwan Park *, Younghun Kim * Department of Chemical Engineering, Kwangwoon University, Seoul 139-701, Republic of Korea
A R T I C L E I N F O
Article history: Received 18 August 2014 Received in revised form 2 October 2014 Accepted 13 October 2014 Available online 22 October 2014 Keywords: Silver nanoparticles Activated sludge Adsorption Biological floc Langmuir isotherm
A B S T R A C T
The removal of AgNPs released from nano-consumer products or effluent of wastewater from treatment plants is important to reduce the potential risk of AgNPs. In this work, biosorptive removal of bare-, citrate-, and polyvivnypyrrolidone-AgNPs were tested using biological floc in activated sludge. The results of isotherms and kinetics test showed that zeta potential of AgNPs depended on the hydrodynamic diameter and then finally maximum uptake capacity. Herein, biological floc in activated sludge could be used as effective biosorbents for removal of AgNPs in aqueous solutions. ß 2014 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.
Introduction The rapidly increasing nanotechnology industry is producing nanomaterials that are incorporated into nano-consumer products. In particular, silver nanoparticles (AgNPs) are used in various fields of applications such as disinfecting sprays, cosmetics, fabrics, and household appliances [1–5]. While Ag+ released from AgNPs, AgNP surface-associated ionic Ag and AgNPs are all likely to contribute to observed toxicity, ionic Ag generally exhibits the strongest toxic effects [6]. As a proof-of-Ag+ release from AgNPs contained in nano-consumer products, few studies are available on the leaching of silver from clothing into water [1–5,7,8]. For example, Benn and Westerhoff investigated the silver released from commercial socks into water, and found socks that contained up to 1.36 mg-Ag/g-sock leached as much as 0.65 mg of silver in 0.5 L of distilled water [3]. Nowack et al. [1] also determined the amount and form of silver released during washing from nine fabrics with different methods of silver incorporation into the fibers. Impellitter et al. [7] have shown that bleach has strong effects on the speciation of silver in textiles and that significant transformation of elemental Ag to AgCl occurred in the presence of bleach. Farkas et al. [8] provided evidence that AgNPs of about
* Corresponding authors. Tel.: +82 2 940 5768; fax: +82 2 941 5769. E-mail addresses: [email protected] (C. Park), [email protected] (Y. Kim).
10 nm diameter are present in the effluent of washing machines that release Ag into the washwater. The silver released into sewer systems can be partitioned onto wastewater biomass and be removed at the wastewater treatment plant (WTP). Based on the pilot scaled WTP, about 90–95% spiked AgNPs are efficiently reduced by biological treatment and accumulated in activated sludge or biosolids [9]. Namely, it is possible for 5–10% of silver to remain in the treated water. Although many studies focus on the removal of Ag+ from aqueous solutions [10], few studies have been carried out that investigate the elimination of AgNPs using adsorbents or resins. Khan et al. [11] reported the adsorptive removal of AgNPs by bacteria resistant to AgNP, isolated from the sewage environment. The removal of AgNPs might differ from that of Ag+ in aqueous solutions. In addition, the activity of micoorganisms in active sludge was influenced by AgNPs as well as its ions. In previous report [12], we found that AgNPs inactivated biofilms in a biosorption-dependent manner, although the activity of AgNPs against planktonic cells was ca. 10% that of silver ions. The presence of both a synthetic wastewater component and extracellular polymeric substances (EPS) significantly enhanced the AgNPs removal efficiency [13]. AgNPs appeared disperse within the EPS matrix, while in the absence of EPS, the AgNPs were adsorbed to the surface of the microorganisms. AgNPs entrapment in the EPS matrix was one of the major mechanisms of AgNPs removal, while, regardless of the presence of EPS, the removal of AgNPs are only affected by
http://dx.doi.org/10.1016/j.jiec.2014.10.012 1226-086X/ß 2014 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.
52
S.Y. Oh et al. / Journal of Industrial and Engineering Chemistry 25 (2015) 51–55
AgNPs adsorption onto microorganisms which can be explained by the electrical double layer compression of both AgNPs and microorganisms. In this work the adsorptive features of different type of AgNPs in aqueous phase using activated sludge is investigated. Herein (i) the influence of stabilizer type of nanoparticles for adsorption of AgNPs on activated sludge, (ii) the adsorption isotherms and kinetics of adsorption in de-ionized water (DW) and synthetic sewage water (SSW) and (iii) desorption feature of AgNPs from activated sludge is elucidated. Three different AgNPs, i.e., bare, citrate, and polyvivnypyrrolidone (PVP) coated, are used as target adsorbates.
Adsorption test for isotherm and kinetics The equilibrium concentrations for three different AgNPs were controlled at 5–200 ppm. Approximately 200 mg/L of activated sludge was added to 40 mL of a solution prepared with a predetermined AgNPs concentration in DW and SSW, followed by shaking at room temperature for 24 h. The samples were separated by filtration, and the adsorption capacities were calculated from the difference between the initial and the final concentrations. A kinetic study was carried out at different time intervals according to the above procedure to determine the removal rate of AgNPs by activated sludge at a concentration of 200 ppm. The rate constants were calculated using the conventional rate expression.
Experimental Results and discussion Materials Characterization of AgNPs Expect for citrate-AgNP, all chemicals, and bare-AgNPs and PVP-AgNPs were purchased from Sigma–Aldrich. Citrate-AgNP was prepared using a method previously reported [14]. AgNPs were dispersed in de-ionized water (DW) and synthetic sewage water (SSW) using ultrasonic horn with a frequency of 200 kHz (Ulh700S, ULSSO Hi-Tech). Based on the modified OECD TG (Organization for Economic Co-operation and Development, Test Guidance) 303A [15], SSW stock with 138 mM ionic strength was prepared using a mixture of the following compositions: glucose 200 mg/L, yeast extract 10 mg/L, bactopeptone 10 mg/L, (NH4)2SO4 100 mg/L, K2HPO4 30 mg/L, KH2PO4 30 mg/L, MgSO4 1.8 mg/L, FeCl3 0.04 mg/L, NaCl 1.4 mg/L, CaCl2 0.04 mg/L, CoCl2 0.48 mg/L, and NaHCO3 60 mg/L. Activated sludge for use as biosorbents was obtained from the Joong-Rang sewage treatment plant (STP) in Seoul, Korea. Characterization of AgNPs The morphology of AgNPs dispersed in SSW was observed using transmission electron microscopy (TEM, JEM-200CX Jeol). Hydrodynamic diameter (HDD) and zeta potential was analyzed by dynamic light scattering spectroscopy (DLS, ELS-Z Photal). The adsorption capacities were measured by inductively coupled plasma-atomic emission spectroscopy (ICP-AES, Shimadzu ICPQ 1000).
Because bare-AgNP was powder type but citrate- and PVPAgNPs were colloidal suspensions, the three AgNPs samples have different particle properties in solution. While bare- and citrateAgNPs are stabilized by electrostatic repulsion mechanism, PVP-AgNP is stabilized by the steric hindrance of PVP. Although bare-AgNP was dispersed in DW or SWW using tip-type ultrasonic, TEM images showed this particle in an aggregated form (Fig. 1). Whereas, individually separated particle distribution is observed in colloidal-type AgNPs (citrate- and PVP-), as shown in Fig. 1. Citrate with three carboxyl groups (COO–) immobilized on the surface of AgNP revealed a strong negative charge of the AgNP. Thus, citrate-AgNP was stabilized by electrostatic repulsion, reducing the aggregation between neighboring citrate-AgNPs. PVP with the pyrrolidone group (cyclic organic) immobilized on the surface of AgNPs induced steric hindrance and maintained an appropriate separating distance between neighboring PVP-AgNPs [16]. In addition, carbon and oxygen double bonds in pyrrolidone induced a negative charge, causing PVP-AgNPs to show a weak negative surface charge. As a result, PVP-AgNPs were stabilized by a combination of steric hindrance and electrostatic repulsion. As summarized in Table 1, the primary particle size of bare-, citrate-, and PVP-AgNP was <120, >20, and <150 nm, respectively. When these materials were dispersed in DW and SSW, the zeta potential was affected by the ionic strength of the solution and
Fig. 1. TEM images of bare-, citrate-, and PVP-AgNP.
S.Y. Oh et al. / Journal of Industrial and Engineering Chemistry 25 (2015) 51–55
53
Table 1 Physicochemical information of three different AgNPs. Materials
Particles size (nm)
Bare-AgNP Citrate-AgNP PVP-AgNP
<120 >20 <150
HDD (nm)
Zeta potential (mV)
DW
SSW
DW
SSW
50.6 24.8 23.4 9.2 90.3 13.9
60.9 59.9 82.9 22.9 115.5 1.1
27.1 1.8 33.6 3.5 22.2 1.6
20.2 1.3 15.4 1.6 11.7 0.6
then dispersion stability also changed. The HDD of AgNPs in DW was not the same as that in SSW. Namely, HDD of bare-, citrate-, and PVP-AgNPs in DW was 50, 23, and 90 nm, and HDD in SSW was 61, 83, and 115 nm, respectively. Because of the small zeta potential in SSW compared to that in DW, the dispersion stability of AgNPs was reduced and the AgNPs easily aggregated with adjacent AgNPs, followed increasing HDD. AgNPs with small zeta potential showed generally large HDD, and zeta potential values of AgNPs in DW and SSW were linearly dependent upon HDD of AgNPs, when HDD was fitted with zeta potential in Table 1. Equilibrium study Activated sludge is generally used for treating sewage and industrial wastewaters using air and a biological floc composed of bacteria and protozoa. This biological floc acted as biosorbents to remove AgNPs in aqueous solutions, and thus AgNPs are accumulated and concentrated in sludge in a bioreactor. To optimize the design of a sorption system for the removal of AgNPs from aqueous solutions, it is important to establish the most appropriate correlation for the equilibrium curves. The Langmuir isotherm was examined in the present study. The Langmuir adsorption isotherm has been successfully applied to many other real sorption processes, and has been used to explain the process of sorption of a metal to a porous material. A basic assumption of the Langmuir theory is that sorption takes place at specific homogeneous sites within the adsorbent. The adsorption equilibrium constant (K) and maximum uptake capacity (qm) was obtained using the following equation. qe ¼
Kqm C e 1 þ KC e
(1)
where qe and Ce are the equilibrium concentration in adsorbent and adsorbate, respectively. The resulting isotherms are plotted in Fig. 2, and the parameter values of isotherms are summarized in Table 2. While the maximum uptake capacities for the three AgNPs onto activated sludge in DW were 44.85, 26.54, and 57.59 mmol/g, respectively, those in SSW were 42.23, 19.67, and 89.08, respectively. When AgNPs and activated sludge are exposed in DW, AgNPs with large negative zeta potential showed small particle size in DW and thus its loading capacity in activated sludge was decreased. It is noted that large HDD induced by aggregation between neighboring AgNPs was easily caught by biological floc in activated sludge. Because this sorption feature is analogous to catching a fish with a net, larger particles are more easily caught, compared to smaller particles. When AgNPs and activated sludge are exposed in SSW, PVP-AgNP had largest HDD and showed a large uptake in activated sludge. Therefore, the zeta potential of AgNPs affects by the HDD and finally the uptake capacity. A popular equation used in the analysis of isotherms with a high degree of rectangularity is that proposed by Dubinin and Radushkevich (D–R) [17]: qe ¼ qs expðBe2 Þ
(2)
Fig. 2. Langmuir isotherms of bare-AgNPs in DW and SSW.
where qs is D–R constant and e can be correlated: 1 e ¼ RTln 1 þ Ce
(3)
The constant B gives the mean free energy E of sorption per molecule of adsorbate when it is transferred to the surface of the solid from infinity in the solution and can be computed using the following relationship: 1 E ¼ pffiffiffiffiffiffi 2B
(4)
As summarized in Table 2, the maximum uptake capacity of biological floc to AgNPs in SSW depends linearly on the sorption energy. It is noted that sorption energy between AgNPs and biological floc in activated sludge is essential to adsorb AgNPs on sludge. Adsorption kinetics study The kinetics of sorption that describe the solute uptake rate that governs the residence time of the sorption reaction is one of the
Table 2 Parameters of Langmuir and Dubinin–Radushkevich isotherms for bare-, citrate-, and PVP-AgNPs in DW and SSW. Solutions
Adsorbates
Langmuir
D–R
qm (mg/g)
K (L/mg)
R2
E (kJ/mol)
DW
Bare-AgNP Citrate-AgNP PVP-AgNP
44.85 26.54 57.59
0.048 0.177 0.725
0.974 0.996 0.994
3.12 1.98 2.46
SSW
Bare-AgNP Citrate-AgNP PVP-AgNP
42.23 19.67 89.08
0.189 4.957 0.259
0.999 0.979 0.874
3.33 2.87 3.95
S.Y. Oh et al. / Journal of Industrial and Engineering Chemistry 25 (2015) 51–55
54
Fig. 3. Adsorption kinetic data for PVP-AgNP in DW and SSW.
Fig. 4. Desorption kinetic data for citrate-AgNP in SSW.
important characteristics that defines the efficiency of sorption. In the present study, the kinetics of AgNP removal was carried out to understand the adsorption behavior of AgNPs onto biological floc in activated sludge. The adsorption of AgNPs in DW and SSW to activated sludge was found to be time-dependent (Fig. 3). The adsorption was rapid in the first 1 h and then slowed considerably as the reaction approached equilibrium. The adsorption rate of AgNPs in DW differed slightly to that in SSW. Quantifying the changes in sorption with time requires an appropriate kinetic model be used, and the pseudo-second-order equation [18] is applied to sorption kinetics for describing the adsorption of AgNPs to activated sludge: t 1 1 ¼ þ t qt kad q2e qe
(5)
where kad is the rate constant of adsorption. The initial sorption rate (h) as time approaches zero can be defined as: h ¼ kad q2e
The rate constant for intraparticle diffusion (kid) is given by Weber and Morris (W–M) [17]: q ¼ kid t 1=2
(7) 1/2
When q is plotted versus t , the W–M plot shows an initial steep-sloped portion at intraparticle diffusion, followed by the plateau at equilibrium. The initial steep-sloped portion (from 0 to 60 min) is attributed to surface adsorption and gradual adsorption, where the intraparticle diffusion is rate-controlled and the plateau is at equilibrium. At this region the intraparticle diffusion begins to slow due to the extremely low solute concentration in the solution. The intraparticle diffusion rate was obtained from the slope of the steep-sloped portion. This value of AgNPs in DW is higher than that in SSW, but there is negative correlation between maximum uptake and intraparticle diffusion rate. In addition, the initial sorption rate inversely depended on the intraparticle diffusion rate. It is noted that the intraparticle diffusion rate was not essential factor for fast adsorption and high uptake in the adsorbent.
(6) Desorption of AgNPs from activated sludge
The initial sorption rate (h), the equilibrium sorption capacity (qe), and the pseudo-second-order rate constant (k) can be determined experimentally from the slope and intercept of the plot of t/qt versus t. Good fits were found for all concentrations, indicating that the sorption reaction can be approximated with a pseudo-second-order kinetics model. The rate constants are listed in Table 3. It can be seen that the h value of AgNPs in SSW is higher than that in DW. Although initial sorption of PVP-AgNP in SSW is very fast compared to other case, there is less dependence of initial sorption rate (h) upon the maximum uptake capacity (qm).
Activated sludge adsorbed with 7 mg/g of citrate-AgNP was used to test for the possibility and extent of desorption. As shown in Fig. 4, citrate-AgNPs caught in biological floc does not detach from activated sludge. It is noted that AgNPs contained in influent in WTP and STP were easily removed by activated sludge, but when AgNPs becomes adsorbed to biological floc, there is no desorption. The desorption of AgNPs caught in biological floc was not easy, unless the pH adjustment, namely, the ionization of AgNPs in the low pH condition. Therefore, waste sludge cake used to removal of
Table 3 Pseudo-second-order rate constants and intraparticle diffusion rates for AgNPs in DW and SSW. Solutions
Adsorbates
Pseudo-second-order rate
Intraparticle diffusion rate
h (mg/g min)
kad (g/mg min)
qe (mg/g)
kid (mg/g h1/2)
DW
Bare-AgNP Citrate-AgNP PVP-AgNP
0.124 0.135 0.256
0.0005 0.0009 0.0058
16.051 12.376 6.640
9.17 9.23 6.80
SSW
Bare-AgNP Citrate-AgNP PVP-AgNP
0.244 0.198 0.980
0.0024 0.0017 0.0545
10.173 10.672 4.243
6.87 7.46 5.27
S.Y. Oh et al. / Journal of Industrial and Engineering Chemistry 25 (2015) 51–55
55
NPs in WTP and STP should be carried out post-treatment to reduce the environmental release of AgNPs.
Acknowledgements
Conclusions
This work is supported by the National Research Foundation of Korea (NRF-2013R1A1A2A10004353) and the Korea Environmental Industry & Technology Institute (E314-00014-0402-1).
In the present study, the adsorptive features of bare-, citrate-, and PVP-AgNPs on biological floc in activated sludge was investigated. Because biosorptive removal of AgNPs is generally dependent on the physicochemical properties of AgNPs, particle size, HDD and zeta potential in DW and SSW were analyzed. When HDD was fitted with zeta potential, AgNPs with small zeta potential showed generally large HDD, and zeta potential values of AgNPs in DW and SSW were linearly dependent upon HDD of AgNPs. To determine the adsorption capacity of AgNPs dispersed in DW and SSW, isotherm and kinetics studies were carried out at different concentrations. AgNPs with large negative zeta potentials showed small particle size in DW and thus its loading capacity in activated sludge was decreased. Namely, large HDD induced by aggregation between neighboring AgNPs was easily caught by biological floc in activated sludge. In the kinetics test, while the initial sorption of PVP-AgNP in SSW is very fast compared to other case, there is less dependence of initial sorption rate (h) upon the maximum uptake capacity (qm). In the desorption test, we found that there is no desorption when AgNPs becomes adsorbed to biological floc. Consequently, biological floc in activated sludge could be used as very effective biosorbents for removal of AgNPs in aqueous solutions. However post-treatments, such as recovery of silver or transformation of less-toxic particles (Ag2S) from AgNPs [19], are required to reduce the cytotoxicity of concentrated silver in waste active sludge cake.
References [1] L. Geranio, M. Heuberger, B. Nowack, Environ. Sci. Technol. 43 (2009) 8113. [2] M.E. Quadros, R. Pierson, N.S. Tulve, R. Willis, K. Rogers, T.A. Thomas, L.C. Marr, Environ. Sci. Technol. 47 (2013) 8894. [3] T.M. Benn, P. Westerhoff, Environ. Sci. Technol. 42 (2008) 4133. [4] C. Som, P. Wich, H. Krug, B. Nowack, Environ. Int. 37 (2011) 1131. [5] C. Lorenz, L. Windler, N. Von Goetz, R.P. Lehmann, M. Schuppler, K. Hungerbruhler, M. Heuberger, B. Nowack, Chemosphere 89 (2012) 817. [6] W.C. Hou, B. Stuart, R. Howes, R.G. Zepp, Environ. Sci. Technol. 47 (2013) 7713. [7] C.A. Impellitteri, T.M. Tolaymat, K.G. Scheckel, J. Environ. Qual. 38 (2009) 1528. [8] J. Farkas, H. Peter, P. Christian, J.A.G. Urrea, M. Hassellov, J. Tuoriniemi, S. Gustafsson, E. Olsson, K. Hylland, K.V. Thomas, Environ. Int. 37 (2011) 1057. [9] L. Li, G. Hartmann, M. Doblinger, M. Schuster, Environ. Sci. Technol. 47 (2013) 7317. [10] M. Akgul, A. Karabakan, O. Acar, Y. Yurum, Microporous Mesoporous Mater. 94 (2006) 99. [11] S.S. Khan, A. Mukherjee, N. Chandrasekaran, Colloids Surf. B 92 (2012) 156. [12] H.J. Park, S. Park, J. Roh, S. Kim, K. Choi, K. Yi, Y. Kim, J. Yoon, J. Ind. Eng. Chem. 19 (2013) 614. [13] H.J. Park, H.Y. Kim, S. Cha, C.H. Ahn, J. Roh, S. Park, S. Kim, K. Choi, J. Yi, Y. Kim, J. Yoon, Chemosphere 92 (2013) 524. [14] P.C. Lee, D. Meisel, J. Phys. Chem. 86 (1982) 3391. [15] OECD, Guidelines for the Testing of Chemicals, Test No. 303A: Simulation test – aerobic sewage treatment: activated sludge unit, 2001. [16] J. Roh, H.N. Umh, J. Sim, S. Park, J. Yi, Y. Kim, Korean J. Chem. Eng. 30 (2013) 671. [17] S. Rengaraj, C.K. Joo, Y. Kim, J. Yi, Hazard J. Mater. B102 (2003) 257. [18] Y.S. Ho, D.A.J. Wase, C.F. Forster, Environ. Technol. 17 (1996) 71. [19] J. Liu, K.G. Pennell, R.H. Hurt, Environ. Sci. Technol. 45 (2011) 7345.
Contents lists available at ScienceDirect
Journal of Industrial and Engineering Chemistry journal homepage: www.elsevier.com/locate/jiec
Biosorptive removal of bare-, citrate-, and PVP-coated silver nanoparticles from aqueous solution by activated sludge Seung Yeon Oh, Hwa Kyung Sung, Chulhwan Park *, Younghun Kim * Department of Chemical Engineering, Kwangwoon University, Seoul 139-701, Republic of Korea
A R T I C L E I N F O
Article history: Received 18 August 2014 Received in revised form 2 October 2014 Accepted 13 October 2014 Available online 22 October 2014 Keywords: Silver nanoparticles Activated sludge Adsorption Biological floc Langmuir isotherm
A B S T R A C T
The removal of AgNPs released from nano-consumer products or effluent of wastewater from treatment plants is important to reduce the potential risk of AgNPs. In this work, biosorptive removal of bare-, citrate-, and polyvivnypyrrolidone-AgNPs were tested using biological floc in activated sludge. The results of isotherms and kinetics test showed that zeta potential of AgNPs depended on the hydrodynamic diameter and then finally maximum uptake capacity. Herein, biological floc in activated sludge could be used as effective biosorbents for removal of AgNPs in aqueous solutions. ß 2014 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.
Introduction The rapidly increasing nanotechnology industry is producing nanomaterials that are incorporated into nano-consumer products. In particular, silver nanoparticles (AgNPs) are used in various fields of applications such as disinfecting sprays, cosmetics, fabrics, and household appliances [1–5]. While Ag+ released from AgNPs, AgNP surface-associated ionic Ag and AgNPs are all likely to contribute to observed toxicity, ionic Ag generally exhibits the strongest toxic effects [6]. As a proof-of-Ag+ release from AgNPs contained in nano-consumer products, few studies are available on the leaching of silver from clothing into water [1–5,7,8]. For example, Benn and Westerhoff investigated the silver released from commercial socks into water, and found socks that contained up to 1.36 mg-Ag/g-sock leached as much as 0.65 mg of silver in 0.5 L of distilled water [3]. Nowack et al. [1] also determined the amount and form of silver released during washing from nine fabrics with different methods of silver incorporation into the fibers. Impellitter et al. [7] have shown that bleach has strong effects on the speciation of silver in textiles and that significant transformation of elemental Ag to AgCl occurred in the presence of bleach. Farkas et al. [8] provided evidence that AgNPs of about
* Corresponding authors. Tel.: +82 2 940 5768; fax: +82 2 941 5769. E-mail addresses: [email protected] (C. Park), [email protected] (Y. Kim).
10 nm diameter are present in the effluent of washing machines that release Ag into the washwater. The silver released into sewer systems can be partitioned onto wastewater biomass and be removed at the wastewater treatment plant (WTP). Based on the pilot scaled WTP, about 90–95% spiked AgNPs are efficiently reduced by biological treatment and accumulated in activated sludge or biosolids [9]. Namely, it is possible for 5–10% of silver to remain in the treated water. Although many studies focus on the removal of Ag+ from aqueous solutions [10], few studies have been carried out that investigate the elimination of AgNPs using adsorbents or resins. Khan et al. [11] reported the adsorptive removal of AgNPs by bacteria resistant to AgNP, isolated from the sewage environment. The removal of AgNPs might differ from that of Ag+ in aqueous solutions. In addition, the activity of micoorganisms in active sludge was influenced by AgNPs as well as its ions. In previous report [12], we found that AgNPs inactivated biofilms in a biosorption-dependent manner, although the activity of AgNPs against planktonic cells was ca. 10% that of silver ions. The presence of both a synthetic wastewater component and extracellular polymeric substances (EPS) significantly enhanced the AgNPs removal efficiency [13]. AgNPs appeared disperse within the EPS matrix, while in the absence of EPS, the AgNPs were adsorbed to the surface of the microorganisms. AgNPs entrapment in the EPS matrix was one of the major mechanisms of AgNPs removal, while, regardless of the presence of EPS, the removal of AgNPs are only affected by
http://dx.doi.org/10.1016/j.jiec.2014.10.012 1226-086X/ß 2014 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.
52
S.Y. Oh et al. / Journal of Industrial and Engineering Chemistry 25 (2015) 51–55
AgNPs adsorption onto microorganisms which can be explained by the electrical double layer compression of both AgNPs and microorganisms. In this work the adsorptive features of different type of AgNPs in aqueous phase using activated sludge is investigated. Herein (i) the influence of stabilizer type of nanoparticles for adsorption of AgNPs on activated sludge, (ii) the adsorption isotherms and kinetics of adsorption in de-ionized water (DW) and synthetic sewage water (SSW) and (iii) desorption feature of AgNPs from activated sludge is elucidated. Three different AgNPs, i.e., bare, citrate, and polyvivnypyrrolidone (PVP) coated, are used as target adsorbates.
Adsorption test for isotherm and kinetics The equilibrium concentrations for three different AgNPs were controlled at 5–200 ppm. Approximately 200 mg/L of activated sludge was added to 40 mL of a solution prepared with a predetermined AgNPs concentration in DW and SSW, followed by shaking at room temperature for 24 h. The samples were separated by filtration, and the adsorption capacities were calculated from the difference between the initial and the final concentrations. A kinetic study was carried out at different time intervals according to the above procedure to determine the removal rate of AgNPs by activated sludge at a concentration of 200 ppm. The rate constants were calculated using the conventional rate expression.
Experimental Results and discussion Materials Characterization of AgNPs Expect for citrate-AgNP, all chemicals, and bare-AgNPs and PVP-AgNPs were purchased from Sigma–Aldrich. Citrate-AgNP was prepared using a method previously reported [14]. AgNPs were dispersed in de-ionized water (DW) and synthetic sewage water (SSW) using ultrasonic horn with a frequency of 200 kHz (Ulh700S, ULSSO Hi-Tech). Based on the modified OECD TG (Organization for Economic Co-operation and Development, Test Guidance) 303A [15], SSW stock with 138 mM ionic strength was prepared using a mixture of the following compositions: glucose 200 mg/L, yeast extract 10 mg/L, bactopeptone 10 mg/L, (NH4)2SO4 100 mg/L, K2HPO4 30 mg/L, KH2PO4 30 mg/L, MgSO4 1.8 mg/L, FeCl3 0.04 mg/L, NaCl 1.4 mg/L, CaCl2 0.04 mg/L, CoCl2 0.48 mg/L, and NaHCO3 60 mg/L. Activated sludge for use as biosorbents was obtained from the Joong-Rang sewage treatment plant (STP) in Seoul, Korea. Characterization of AgNPs The morphology of AgNPs dispersed in SSW was observed using transmission electron microscopy (TEM, JEM-200CX Jeol). Hydrodynamic diameter (HDD) and zeta potential was analyzed by dynamic light scattering spectroscopy (DLS, ELS-Z Photal). The adsorption capacities were measured by inductively coupled plasma-atomic emission spectroscopy (ICP-AES, Shimadzu ICPQ 1000).
Because bare-AgNP was powder type but citrate- and PVPAgNPs were colloidal suspensions, the three AgNPs samples have different particle properties in solution. While bare- and citrateAgNPs are stabilized by electrostatic repulsion mechanism, PVP-AgNP is stabilized by the steric hindrance of PVP. Although bare-AgNP was dispersed in DW or SWW using tip-type ultrasonic, TEM images showed this particle in an aggregated form (Fig. 1). Whereas, individually separated particle distribution is observed in colloidal-type AgNPs (citrate- and PVP-), as shown in Fig. 1. Citrate with three carboxyl groups (COO–) immobilized on the surface of AgNP revealed a strong negative charge of the AgNP. Thus, citrate-AgNP was stabilized by electrostatic repulsion, reducing the aggregation between neighboring citrate-AgNPs. PVP with the pyrrolidone group (cyclic organic) immobilized on the surface of AgNPs induced steric hindrance and maintained an appropriate separating distance between neighboring PVP-AgNPs [16]. In addition, carbon and oxygen double bonds in pyrrolidone induced a negative charge, causing PVP-AgNPs to show a weak negative surface charge. As a result, PVP-AgNPs were stabilized by a combination of steric hindrance and electrostatic repulsion. As summarized in Table 1, the primary particle size of bare-, citrate-, and PVP-AgNP was <120, >20, and <150 nm, respectively. When these materials were dispersed in DW and SSW, the zeta potential was affected by the ionic strength of the solution and
Fig. 1. TEM images of bare-, citrate-, and PVP-AgNP.
S.Y. Oh et al. / Journal of Industrial and Engineering Chemistry 25 (2015) 51–55
53
Table 1 Physicochemical information of three different AgNPs. Materials
Particles size (nm)
Bare-AgNP Citrate-AgNP PVP-AgNP
<120 >20 <150
HDD (nm)
Zeta potential (mV)
DW
SSW
DW
SSW
50.6 24.8 23.4 9.2 90.3 13.9
60.9 59.9 82.9 22.9 115.5 1.1
27.1 1.8 33.6 3.5 22.2 1.6
20.2 1.3 15.4 1.6 11.7 0.6
then dispersion stability also changed. The HDD of AgNPs in DW was not the same as that in SSW. Namely, HDD of bare-, citrate-, and PVP-AgNPs in DW was 50, 23, and 90 nm, and HDD in SSW was 61, 83, and 115 nm, respectively. Because of the small zeta potential in SSW compared to that in DW, the dispersion stability of AgNPs was reduced and the AgNPs easily aggregated with adjacent AgNPs, followed increasing HDD. AgNPs with small zeta potential showed generally large HDD, and zeta potential values of AgNPs in DW and SSW were linearly dependent upon HDD of AgNPs, when HDD was fitted with zeta potential in Table 1. Equilibrium study Activated sludge is generally used for treating sewage and industrial wastewaters using air and a biological floc composed of bacteria and protozoa. This biological floc acted as biosorbents to remove AgNPs in aqueous solutions, and thus AgNPs are accumulated and concentrated in sludge in a bioreactor. To optimize the design of a sorption system for the removal of AgNPs from aqueous solutions, it is important to establish the most appropriate correlation for the equilibrium curves. The Langmuir isotherm was examined in the present study. The Langmuir adsorption isotherm has been successfully applied to many other real sorption processes, and has been used to explain the process of sorption of a metal to a porous material. A basic assumption of the Langmuir theory is that sorption takes place at specific homogeneous sites within the adsorbent. The adsorption equilibrium constant (K) and maximum uptake capacity (qm) was obtained using the following equation. qe ¼
Kqm C e 1 þ KC e
(1)
where qe and Ce are the equilibrium concentration in adsorbent and adsorbate, respectively. The resulting isotherms are plotted in Fig. 2, and the parameter values of isotherms are summarized in Table 2. While the maximum uptake capacities for the three AgNPs onto activated sludge in DW were 44.85, 26.54, and 57.59 mmol/g, respectively, those in SSW were 42.23, 19.67, and 89.08, respectively. When AgNPs and activated sludge are exposed in DW, AgNPs with large negative zeta potential showed small particle size in DW and thus its loading capacity in activated sludge was decreased. It is noted that large HDD induced by aggregation between neighboring AgNPs was easily caught by biological floc in activated sludge. Because this sorption feature is analogous to catching a fish with a net, larger particles are more easily caught, compared to smaller particles. When AgNPs and activated sludge are exposed in SSW, PVP-AgNP had largest HDD and showed a large uptake in activated sludge. Therefore, the zeta potential of AgNPs affects by the HDD and finally the uptake capacity. A popular equation used in the analysis of isotherms with a high degree of rectangularity is that proposed by Dubinin and Radushkevich (D–R) [17]: qe ¼ qs expðBe2 Þ
(2)
Fig. 2. Langmuir isotherms of bare-AgNPs in DW and SSW.
where qs is D–R constant and e can be correlated: 1 e ¼ RTln 1 þ Ce
(3)
The constant B gives the mean free energy E of sorption per molecule of adsorbate when it is transferred to the surface of the solid from infinity in the solution and can be computed using the following relationship: 1 E ¼ pffiffiffiffiffiffi 2B
(4)
As summarized in Table 2, the maximum uptake capacity of biological floc to AgNPs in SSW depends linearly on the sorption energy. It is noted that sorption energy between AgNPs and biological floc in activated sludge is essential to adsorb AgNPs on sludge. Adsorption kinetics study The kinetics of sorption that describe the solute uptake rate that governs the residence time of the sorption reaction is one of the
Table 2 Parameters of Langmuir and Dubinin–Radushkevich isotherms for bare-, citrate-, and PVP-AgNPs in DW and SSW. Solutions
Adsorbates
Langmuir
D–R
qm (mg/g)
K (L/mg)
R2
E (kJ/mol)
DW
Bare-AgNP Citrate-AgNP PVP-AgNP
44.85 26.54 57.59
0.048 0.177 0.725
0.974 0.996 0.994
3.12 1.98 2.46
SSW
Bare-AgNP Citrate-AgNP PVP-AgNP
42.23 19.67 89.08
0.189 4.957 0.259
0.999 0.979 0.874
3.33 2.87 3.95
S.Y. Oh et al. / Journal of Industrial and Engineering Chemistry 25 (2015) 51–55
54
Fig. 3. Adsorption kinetic data for PVP-AgNP in DW and SSW.
Fig. 4. Desorption kinetic data for citrate-AgNP in SSW.
important characteristics that defines the efficiency of sorption. In the present study, the kinetics of AgNP removal was carried out to understand the adsorption behavior of AgNPs onto biological floc in activated sludge. The adsorption of AgNPs in DW and SSW to activated sludge was found to be time-dependent (Fig. 3). The adsorption was rapid in the first 1 h and then slowed considerably as the reaction approached equilibrium. The adsorption rate of AgNPs in DW differed slightly to that in SSW. Quantifying the changes in sorption with time requires an appropriate kinetic model be used, and the pseudo-second-order equation [18] is applied to sorption kinetics for describing the adsorption of AgNPs to activated sludge: t 1 1 ¼ þ t qt kad q2e qe
(5)
where kad is the rate constant of adsorption. The initial sorption rate (h) as time approaches zero can be defined as: h ¼ kad q2e
The rate constant for intraparticle diffusion (kid) is given by Weber and Morris (W–M) [17]: q ¼ kid t 1=2
(7) 1/2
When q is plotted versus t , the W–M plot shows an initial steep-sloped portion at intraparticle diffusion, followed by the plateau at equilibrium. The initial steep-sloped portion (from 0 to 60 min) is attributed to surface adsorption and gradual adsorption, where the intraparticle diffusion is rate-controlled and the plateau is at equilibrium. At this region the intraparticle diffusion begins to slow due to the extremely low solute concentration in the solution. The intraparticle diffusion rate was obtained from the slope of the steep-sloped portion. This value of AgNPs in DW is higher than that in SSW, but there is negative correlation between maximum uptake and intraparticle diffusion rate. In addition, the initial sorption rate inversely depended on the intraparticle diffusion rate. It is noted that the intraparticle diffusion rate was not essential factor for fast adsorption and high uptake in the adsorbent.
(6) Desorption of AgNPs from activated sludge
The initial sorption rate (h), the equilibrium sorption capacity (qe), and the pseudo-second-order rate constant (k) can be determined experimentally from the slope and intercept of the plot of t/qt versus t. Good fits were found for all concentrations, indicating that the sorption reaction can be approximated with a pseudo-second-order kinetics model. The rate constants are listed in Table 3. It can be seen that the h value of AgNPs in SSW is higher than that in DW. Although initial sorption of PVP-AgNP in SSW is very fast compared to other case, there is less dependence of initial sorption rate (h) upon the maximum uptake capacity (qm).
Activated sludge adsorbed with 7 mg/g of citrate-AgNP was used to test for the possibility and extent of desorption. As shown in Fig. 4, citrate-AgNPs caught in biological floc does not detach from activated sludge. It is noted that AgNPs contained in influent in WTP and STP were easily removed by activated sludge, but when AgNPs becomes adsorbed to biological floc, there is no desorption. The desorption of AgNPs caught in biological floc was not easy, unless the pH adjustment, namely, the ionization of AgNPs in the low pH condition. Therefore, waste sludge cake used to removal of
Table 3 Pseudo-second-order rate constants and intraparticle diffusion rates for AgNPs in DW and SSW. Solutions
Adsorbates
Pseudo-second-order rate
Intraparticle diffusion rate
h (mg/g min)
kad (g/mg min)
qe (mg/g)
kid (mg/g h1/2)
DW
Bare-AgNP Citrate-AgNP PVP-AgNP
0.124 0.135 0.256
0.0005 0.0009 0.0058
16.051 12.376 6.640
9.17 9.23 6.80
SSW
Bare-AgNP Citrate-AgNP PVP-AgNP
0.244 0.198 0.980
0.0024 0.0017 0.0545
10.173 10.672 4.243
6.87 7.46 5.27
S.Y. Oh et al. / Journal of Industrial and Engineering Chemistry 25 (2015) 51–55
55
NPs in WTP and STP should be carried out post-treatment to reduce the environmental release of AgNPs.
Acknowledgements
Conclusions
This work is supported by the National Research Foundation of Korea (NRF-2013R1A1A2A10004353) and the Korea Environmental Industry & Technology Institute (E314-00014-0402-1).
In the present study, the adsorptive features of bare-, citrate-, and PVP-AgNPs on biological floc in activated sludge was investigated. Because biosorptive removal of AgNPs is generally dependent on the physicochemical properties of AgNPs, particle size, HDD and zeta potential in DW and SSW were analyzed. When HDD was fitted with zeta potential, AgNPs with small zeta potential showed generally large HDD, and zeta potential values of AgNPs in DW and SSW were linearly dependent upon HDD of AgNPs. To determine the adsorption capacity of AgNPs dispersed in DW and SSW, isotherm and kinetics studies were carried out at different concentrations. AgNPs with large negative zeta potentials showed small particle size in DW and thus its loading capacity in activated sludge was decreased. Namely, large HDD induced by aggregation between neighboring AgNPs was easily caught by biological floc in activated sludge. In the kinetics test, while the initial sorption of PVP-AgNP in SSW is very fast compared to other case, there is less dependence of initial sorption rate (h) upon the maximum uptake capacity (qm). In the desorption test, we found that there is no desorption when AgNPs becomes adsorbed to biological floc. Consequently, biological floc in activated sludge could be used as very effective biosorbents for removal of AgNPs in aqueous solutions. However post-treatments, such as recovery of silver or transformation of less-toxic particles (Ag2S) from AgNPs [19], are required to reduce the cytotoxicity of concentrated silver in waste active sludge cake.
References [1] L. Geranio, M. Heuberger, B. Nowack, Environ. Sci. Technol. 43 (2009) 8113. [2] M.E. Quadros, R. Pierson, N.S. Tulve, R. Willis, K. Rogers, T.A. Thomas, L.C. Marr, Environ. Sci. Technol. 47 (2013) 8894. [3] T.M. Benn, P. Westerhoff, Environ. Sci. Technol. 42 (2008) 4133. [4] C. Som, P. Wich, H. Krug, B. Nowack, Environ. Int. 37 (2011) 1131. [5] C. Lorenz, L. Windler, N. Von Goetz, R.P. Lehmann, M. Schuppler, K. Hungerbruhler, M. Heuberger, B. Nowack, Chemosphere 89 (2012) 817. [6] W.C. Hou, B. Stuart, R. Howes, R.G. Zepp, Environ. Sci. Technol. 47 (2013) 7713. [7] C.A. Impellitteri, T.M. Tolaymat, K.G. Scheckel, J. Environ. Qual. 38 (2009) 1528. [8] J. Farkas, H. Peter, P. Christian, J.A.G. Urrea, M. Hassellov, J. Tuoriniemi, S. Gustafsson, E. Olsson, K. Hylland, K.V. Thomas, Environ. Int. 37 (2011) 1057. [9] L. Li, G. Hartmann, M. Doblinger, M. Schuster, Environ. Sci. Technol. 47 (2013) 7317. [10] M. Akgul, A. Karabakan, O. Acar, Y. Yurum, Microporous Mesoporous Mater. 94 (2006) 99. [11] S.S. Khan, A. Mukherjee, N. Chandrasekaran, Colloids Surf. B 92 (2012) 156. [12] H.J. Park, S. Park, J. Roh, S. Kim, K. Choi, K. Yi, Y. Kim, J. Yoon, J. Ind. Eng. Chem. 19 (2013) 614. [13] H.J. Park, H.Y. Kim, S. Cha, C.H. Ahn, J. Roh, S. Park, S. Kim, K. Choi, J. Yi, Y. Kim, J. Yoon, Chemosphere 92 (2013) 524. [14] P.C. Lee, D. Meisel, J. Phys. Chem. 86 (1982) 3391. [15] OECD, Guidelines for the Testing of Chemicals, Test No. 303A: Simulation test – aerobic sewage treatment: activated sludge unit, 2001. [16] J. Roh, H.N. Umh, J. Sim, S. Park, J. Yi, Y. Kim, Korean J. Chem. Eng. 30 (2013) 671. [17] S. Rengaraj, C.K. Joo, Y. Kim, J. Yi, Hazard J. Mater. B102 (2003) 257. [18] Y.S. Ho, D.A.J. Wase, C.F. Forster, Environ. Technol. 17 (1996) 71. [19] J. Liu, K.G. Pennell, R.H. Hurt, Environ. Sci. Technol. 45 (2011) 7345.