Removal of some polycyclic aromatic hydrocarbons from petrochemical wastewater using low-cost adsorbents of natural origin

Removal of some polycyclic aromatic hydrocarbons from petrochemical wastewater using low-cost adsorbents of natural origin

Available online at www.sciencedirect.com Bioresource Technology 99 (2008) 4515–4519 Short Communication Removal of some polycyclic aromatic hydroc...

133KB Sizes 0 Downloads 0 Views

Available online at www.sciencedirect.com

Bioresource Technology 99 (2008) 4515–4519

Short Communication

Removal of some polycyclic aromatic hydrocarbons from petrochemical wastewater using low-cost adsorbents of natural origin Rudy Crisafully a, Maria Aparecida L. Milhome d, Rivelino M. Cavalcante a,b, Edilberto R. Silveira b, Denis De Keukeleire c, Ronaldo F. Nascimento a,* a

d

Departamento de Quı´mica Analı´tica e Fı´sico Quı´mica, Universidade Federal do Ceara´, Campus do Pici, Centro de Cieˆncias, Bloco 940, CEP: 60451-970 Fortaleza, CE, Brazil b Departamento de Quı´mica Orgaˆnica e Inorgaˆnica, Universidade Federal do Ceara´, Campus do Pici, Centro de Cieˆncias, Bloco 940, CEP: 60451-970 Fortaleza, CE, Brazil c Ghent University, Faculty of Pharmaceutical Sciences, Harelbekestraat 72, B-9000 Ghent, Belgium Departamento de Engenharia Hidra´ulica e Ambiental, Universidade Federal do Ceara´, Campus do Pici, Centro de Tecnologia, Bloco 713, CEP: 60455-760 Fortaleza, CE, Brazil Received 13 September 2006; received in revised form 20 August 2007; accepted 23 August 2007 Available online 25 October 2007

Abstract Removal of polycyclic aromatic hydrocarbons (PAHs) from petrochemical wastewater was investigated using various low-cost adsorbents of natural origin including sugar cane bagasse, green coconut shells, chitin, and chitosan. Adsorption experiments of mixtures of PAHs (5.0–15.0 mg/L) have been carried out at ambient temperature (28 ± 2 C) and pH 7.5. The adsorption isotherms of PAHs were in agreement with a Freundlich model, while the uptake capacity of PAHs followed the order: green coconut shells > sugar cane bagasse > chitin > chitosan. The adsorption properties of green coconut shells were comparable to those of some conventional adsorbents such as Amberlite T. The partition coefficients in acetone:water, the adsorption constants at equilibrium, and the molecular masses of the PAHs could be linearly correlated with octanol–water partition coefficients.  2007 Elsevier Ltd. All rights reserved. Keywords: Polycyclic aromatic hydrocarbons; Natural adsorbents; Wastewater; Chitin; Coconut

1. Introduction Polycyclic aromatic hydrocarbons (PAHs) are priority pollutants due to their carcinogenic, mutagenic, and toxic properties (Manoli and Samara, 1999; Pereira Netto et al., 2000). Natural sources of PAHs in the environment are insignificant compared to anthropogenic sources (Witt, 1995; Charlesworth et al., 2002) that originate in combustion of coal and oil (Nielsen, 1996), exhaust from motor

*

Corresponding author. Address: Laborato´rio de Ana´lise Trac¸os – LAT, Departamento de Quı´mica Analı´tica e Fı´sico-Quı´mica, Universidade Federal do Ceara´, Bloco 939, Campus do Pici, CEP 60455-760 Fortaleza, CE, Brazil. Tel.: +55 85 4008 99 58; fax: +55 85 4008 99 82. E-mail address: [email protected] (R.F. Nascimento). 0960-8524/$ - see front matter  2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2007.08.041

vehicles (Halsall et al., 1994; Harrison et al., 1996), and wastewater from petrochemical plants (Domen˜o and Ner´ın, 2003). PAHs are often resistant to biological degradation and are not efficiently removed by conventional physicochemical methods such as coagulation, flocculation, sedimentation, filtration or ozonation (Hinchee and Alleman, 1994). However, adsorption processes are effective in removal of persistent organic pollutants and, in particular, activated carbon is widely used (Cooney, 1999), but high costs and difficult regeneration are clearly disadvantages (Lalvani et al., 1998). Interesting alternatives for removal of organic pollutants (Mckay, 1995; Mackay and Gschwend, 2000; Grupta et al., 2002; Boving and Zhang, 2004; Zheng et al., 2004) from wastewaters are presented by the use of low-cost solid residues from agricultural activities.

4516

R. Crisafully et al. / Bioresource Technology 99 (2008) 4515–4519

In Brazil, huge amounts of waste are produced by largescale agriculture, mainly sugar cane bagasse, which is composed of lignin (ca. 22%), cellulose (ca. 44%), and hemicellulose (ca. 27%) (Teixeira et al, 1997; Pandey et al, 2000), and green coconut shells composed of lignin (35–45%), cellulose (23–43%), and hemicellulose (3–12%) (Noguera et al., 2000; Carrijo et al., 2002) for which applications are only sparingly available. The purpose of this study was to investigate the adsorption features of adsorbents from agricultural waste materials with respect to removal of PAHs from petrochemical wastewater. Results were compared with performances of some adsorbents used in practice. 2. Methods 2.1. Reagents and solutions Acetone was supplied by Merck, Sa˜o Paulo, Brazil. PAHs (>98% purity) were obtained from Sigma–Aldrich (St. Louis, Missouri, USA): naphthalene (Nap), acenaphthene (Acen), anthracene (Ant), pyrene (Pyr). Compounds (1 g/L) were dissolved in acetone and stock solutions in water–acetone 70:30% v/v (5.0–15.0 mg/L) were prepared by serial dilutions in water. A pH meter (Micronal, Sa˜o Paulo, Brazil) was used for pH measurements. 2.2. Adsorbents Sugar cane bagasse was collected from a factory of spirits based on sugar cane (Fortaleza, Brazil) and green coconut shells (40–60 mesh) were obtained from Embrapa (Fortaleza, Brazil). Chitosan (60–80 mesh, 85%) and chitin (60–80 mesh) were delivered by Polymer (Fortaleza, Brazil). These materials were washed with de-ionized water, dried by lyophilization, and stored in sealed flasks. Activated carbon (100 mesh, Merck Sa˜o Paulo, Brazil), silica (80–100 mesh, Merck), Amberlite T (80–100 mesh, Carlo Erba, Sa˜o Paulo, Brazil), and cellulose (20 mesh, Carlo Erba, Sa˜o Paulo, Brazil) were activated according to the literature (Mckay, 1995). 2.3. Analyses of PAHs A Shimadzu model GC-17A gas chromatograph equipped with a flame ionization detector (FID) and with a capillary DB-5 column (30 m · 0.25 mm, 0.25 lm film thickness) was used. Wastewater samples from a local petrochemical plant were spiked with PAHs and filtered through a 0.2 mm membrane (Millipore, Sa˜o Paulo, Brazil). The first 5 mL of the filtrate were removed and the next 5 mL were collected into a sample tube and stored at 4 C until analysis. A volume of 1.0 lL was injected in the split mode (1:30). The analysis started at 120 C and the temperature was increased to 250 C at a rate of 10 C/min. The temperature of the injector and of the detector was set at 250 C. The carrier gas was H2 at a flow rate of 1 mL/

min. PAHs were quantified using calibration curves by direct injection of standard mixtures with know concentrations (correlation coefficients (R2) were greater than 0.99). 2.4. Adsorption studies Wastewaters were diluted in acetone (30% v/v) prior to spiking with known concentrations of PAHs (5.0–15.0 mg/ L). Closed flasks containing 10 mL of these solutions and 0.20 g dry adsorbent were shaken (150 rpm) for 10 h at ambient temperature (28 ± 2 C) (pH 7.5). After filtration, PAHs were quantified by GC–FID. Comparison between the adsorbents of natural origin and some conventional adsorbents was done by mixing the PAHs (10.0 mL, 15.0 mg/L) with the adsorbent (0.30 g) in water–acetone (70:30 v/v) under the same conditions as described above. The maximum concentrations of PAHs as measured in control samples (no PAHs added) were lower than 0.1% of the total contents of PAHs in the spiked samples. All experiments using wastewaters were performed in triplicate. The adsorption capacities were calculated based on the differences of the concentrations of solutes before and after the experiment according to the following Eq. (1): qe ¼

ðC 0  C e ÞV W

ð1Þ

where qe is the concentration of the adsorbed solute (mg/g); C0 and Ce are the initial and final concentrations of the solute in solution (mg/L); V (mL) is the volume of the solution and W (g) is the mass of the adsorbent. Adsorption isotherms (relationship between the adsorption capacities and the concentrations of PAHs) were investigated using the linearized form of the Freundlich sorption isotherm equation (Cooney, 1999). 3. Results and discussion 3.1. Adsorption isotherms The nature of the solvent evidently influences the adsorption of solutes. As the experiments have been carried out in wastewaters containing 30% (v/v) acetone as a cosolvent to improve solubilities and hamper adsorption at the walls of the flasks, the results reported here only mimic adsorption features in pure water. Therefore, quantitative data referring to PAHs are undoubtedly underestimated. Although this restriction should be kept in mind, it remains true that interactions between an adsorbent and an organic adsorbate can be governed by Langmuir and Freundlich isotherms (Cooney, 1999; Deosarkar and Pangarkar, 2004; Ayranci, 2005). The Langmuir model is based on monolayer adsorption on equi-energetic active surface sites, while the Freundlich model relies on heterogeneous adsorption. However, to evaluate the linearity, the experimental data were only fitted to the linearized form of the

R. Crisafully et al. / Bioresource Technology 99 (2008) 4515–4519

Freundlich sorption isotherm equation (Cooney, 1999), Eq. (2): log qe ¼ log K f þ

1 log C e n

ð2Þ

where qe is the concentration of the adsorbed solute (mg/g); Ce is the concentration of the solute in solution (mg/L); Kf (mg g1(1/n) l1/n g1) is related to the adsorption capacity adsorbent and 1/n is related to the surface heterogeneity. After fitting the equilibrium adsorption data for multisolute adsorption, Freundlich parameters (1/n and Kf) were obtained from straight lines when log qe was plotted against log Ce (Table 1). Satisfactory correlation coefficients were noted (r > 0.90). In order to more clearly define the model which represented the experimental data most correctly, parameter known as ‘normalized percent deviation’ (P) (Akzu and Yener, 2001; Ayranci, 2005) was applied, according to the following Eq. (3):   100 X jqe  qp j P¼ ð3Þ N qe where qe and qp are the experimental and predicted adsorption capacities (mg/g); N is the number of observations. The lower the P value is, the better the fit is. The P values calculated for the PAHs according to the Freundlich equation are given in Table 1. P deviations are in general lowest when the experimental data were fitted to the Freundlich equation, however, with some exceptions (Acen and Ant for sugar cane bagasse and Pyr for chitin).

Table 1 Parameters derived from Freundlich isotherm equation for the adsorption of polycyclic aromatic hydrocarbons on adsorbents of natural origin (ambient temperature at 28 ± 2 C, pH 7.5) Freundlich parameters Kf

1/n

P

Naphthalene Sugar cane bagasse Green coconut shells Chitin Chitosan

0.013 0.023 0.138 0.002

0.98 1.34 0.05 1.39

15.1 3.4 6.4 13.6

Acenaphtene Sugar cane bagasse Green coconut shells Chitin Chitosan

0.014 0.018 0.003 0.001

1.23 2.39 1.99 2.71

7.1 6.8 3.8 9.2

Anthracene Sugar cane bagasse Green coconut shells Chitin Chitosan

0.025 0.050 0.015 0.019

0.64 0.77 0.26 0.77

9.9 6.1 19.5 10.8

Pyrene Sugar cane bagasse Green coconut shells Chitin Chitosan

0.173 0.138 0.125 0.047

0.15 0.99 0.14 0.21

1.9 12.4 5.5 16.0

4517

The adsorption process of organic solutes is determined by various interactions that govern the association between solute and adsorbent, such as Van der Waals and dipole– dipole interactions, electrostatic forces, and weak intermolecular associations (Mckay, 1995; Cooney, 1999). Since PAHs are nonpolar compounds, adsorption must be governed mainly by hydrophobic interactions (Schwarzenbach et al, 1993; Boving and Zang, 2004). In order to reveal a potential relationship between adsorption efficiency and lignin content, experimental partition coefficients, Kads, were correlated to the published n-octanol–water partition coefficients, Kow (Schwarzenbach et al, 1993). Bagasse partition coefficients (Kads), defined according to Eq. (4), were calculated from the linear variation of the sorbed PAHs concentrations, qe, with aqueous PAHs solutions, Ce (Mackay and Gschwend, 2000; Boving and Zang, 2004). For each sample event, Kads was determined from the regression qe versus Ce. The slope of the graph yields the partition coefficient (Kads), Eq. (4): q ð4Þ K ads ¼ e Ce 1 where K ads ðmg kg1 adsorbent =mg L Þ is the bagasse partition coefficients, Ce is the dissolved PAHs concentration (mg L1) and qe is the mass of sorbed PAHs per dry unit weight of bagasse (mg g1). The natural polymers lignin fraction has been identified as the principal factor determining the degree of sorption of nonpolar compounds (Mackay and Gschwend, 2000). Thus, the lignin–water partition coefficients (Klignin) for PAHs interacting with intact bagasse (i.e., sugar cane, coconut shells) were calculated from bagasse partition coefficients (Kads) and lignin mass (flignin) according to the following Eq. (5):

Table 2 Experimental partition coefficients Kads (L Kg1) and Klignin (L Kg1), and known values of log Kow (mol L1)a for PAHs using adsorbents of natural origin (water:acetone 70:30 v/v) Partition coefficient

Naphthalene

Acenaphtene

Anthracene

Pyrene

Sugar cane bagasse Kads Klignin Log Kow

16.80 56.00 3.36

17.30 57.67 3.76

18.60 62.00 4.54

21.50 71.67 5.13

Green coconut shells 18.20 Kads 50.00 Klignin Log Kow 3.36

24.30 60.75 3.76

26.80 67.00 4.54

38.30 95.75 5.13

Chitin Kads Klignin Log Kow

14.20 – 3.36

15.30 – 3.76

17.30 – 4.54

19.60 – 5.13

Chitosan Kads Klignin Log Kow

5.10 – 3.36

5.50 –

8.80 – 4.54

10.10 – 5.13

a

3.76

From Schwarzenbach et al. (1993). Klignin was calculated based on lignin contents of 20% for sugar cane bagasse and 40% for green coconut shells.

4518

R. Crisafully et al. / Bioresource Technology 99 (2008) 4515–4519

K lignin ¼ K ads =flignin

ð5Þ

log Klignin and log Kow, and for Klignin and the molecular masses of the adsorbates. The adsorption of PAHs related to the lignin content increases with increasing molecular mass, hence Pyr (molecular mass of 202) is most efficiently adsorbed and Nap (molecular mass of 128) is least efficiently adsorbed. Thus, the partition coefficients of PAHs by sugar cane bagasse and green coconut shells based on the lignin content can be estimated from the known values of Kow and from the molecular masses. Our results agree with those reported by Boving and Zhang (2004) in an investigation on removal of PAHs from aqueous solutions using aspen wood fibers. These results confirm previous observations on adsorption of aromatic hydrocarbons to wood which appears to be controlled mainly by the lignin content (Mackay and Gschwend, 2000; Boving and Zang, 2004). The higher lignin content in green coconut shells relative to sugar cane bagasse, 35–45% (Teixeira et al., 1997; Pandey et al., 2000) and ca. 22% (Nogueira et al., 2000; Carrijo et al.,

1 where K lignin ðmg kg1 lignin =mg L Þ is the partition coefficient between lignin and the solution; flignin ðglignin g1 adsorbent Þ: the amount of lignin present. For this study, flignin was considered to be equivalent to 20% (flignin = 0.20) for sugar cane bagasse (Teixeira et al., 1997; Pandey et al., 2000) and to 40% (flignin = 0.40) for green coconut shells (Noguera et al., 2000; Carrijo et al., 2002). To evaluate linearity, the experimental data were also fitted to the linearized form of the Freundlich sorption isotherm equation. The calculated equilibrium Kads values and published Kow values (Schwarzenbach et al, 1993) for each compound are summarized in Table 2. The parameter Kow is the n-octanol–water partition coefficient and is a measure of a compound’s hydrophobicity. When the Klignin value is less than Kow it indicates that lignin is a less favorable partition medium than octanol for the PAHs studied. Correlations for the two adsorbents are also visualized in Fig. 1, both for

a

2.2 Sugar cane bagasse Green coconut shells

lignin

2.0

logK

logK lignin= 0.058logK ow + 1.72 R 2 = 0.93, n= 4

N=4

1.8 logK lignin= 0.161logK ow + 1.13 R 2 = 0.92, n=4 Nap

Acen

Pyr

Ant

1.6 3.2

3.7

4.2

4.7

5.2

logK ow

b

105.0

Sugar cane bagasse Green coconut shells

Klignin = 0.310 MM + 41.40

80.0

Klignin

R2 = 0.88, n=4

55.0

Klignin = 0.633MM - 37.49 R2 = 0.92, n=4 Acen

Nap 30.0 120

130

140

150

Ant 160

170

180

Pyr 190

200

210

Molecular mass (MM) Fig. 1. Correlation of the adsorption coefficients Klignin with n-octanol–water partition coefficients (Kow) (A) and with the molecular masses of naphthalene (Nap), acenaphthene (Acen), anthracene (Ant), and pyrene (Pyr) (B).

R. Crisafully et al. / Bioresource Technology 99 (2008) 4515–4519

2002), respectively, agrees with the extent of adsorption of PAHs followed the order: green coconut shells > sugar cane bagasse > chitin > chitosan. 3.2. Comparison with conventional adsorbents It was of interest to compare the results to applications of conventional adsorbents such as activated carbon, cellulose, silica, and Amberlite T in a model system. The concentrations of total PAHs, expressed in mg/g, were as follows: 0.345 for sugar cane bagasse, 0.553 for green coconut shells, 0.215 for chitin, and 0.112 for chitosan; 0.782 for activated carbon, 0.376 for Amberlite T, 0.123 for silica, and 0.101 for cellulose. Activated carbon proved to be superior, but green coconut shells and sugar cane bagasse were as efficient in removal of PAHs as Amberlite T and performed even better than silica. 4. Conclusions This study showed that adsorption of polycyclic aromatic hydrocarbons (pyrene, anthracene, acenaphthene, naphthalene) from petrochemical wastewater can be effected using low-cost adsorbents of natural origin. Widely available agricultural residues such as sugar cane bagasse and green coconut shells could be useful for the purpose. The adsorption efficiencies followed the order: green coconut shells > sugar cane bagasse > chitin > chitosan, and a correlation could be found with the lignin contents. The adsorption features of green coconut shells proved well comparable to those of conventional adsorbents such as Amberlite T. The adsorption isotherms followed the Freundlich model. Acknowledgements The authors thank to Brazilian agencies CNPq and FUNCAP for financial support of this project and to Federal University of Ceara´ (Department of Analytical Chemistry and Physical Chemistry) and PADETEC for the use of their facilities. References Akzu, Z., Yener, J., 2001. A comparative adsorption/biosorption study of mono-chlorinated phenols onto various sorbents. Waste Manage. 21, 695–702. Ayranci, E., 2005. Adsorption kinetics and isotherms of pesticides onto actived carbon-cloth. Chemosphere 60, 1600–1607.

4519

Boving, T.B., Zhang, W., 2004. Removal of aqueous-phase polynuclear aromatic hydrocarbons using aspen wood fibers. Chemosphere 54, 839–881. Carrijo, O.A., De Liz, R.S., Makishima, N., 2002. Fiber of green coconut shell as an agricultural substrate. Horticult. Brasileira 20, 533–535. Charlesworth, M., Service, M., Gibson, C.E., 2002. PAHs contamination of Irish sediments. Mar. Pollut. Bull. 44, 1421–1424. Cooney, D.O., 1999. Adsorption Designer for Wastewater Treatment. Lewis Publishers, London, England, UK, pp. 45–190. Deosarkar, S.P., Pangarkar, V.G., 2004. Adsorptive separation and recovery of organics from PHA and SA plant effluents. Sep. Purif. Technol. 38, 241–254. Domen˜o, C., Nerı´n, C., 2003. Fate of polyaromatic hydrocarbons in the pyrolysis of industrial waste oils. J. Anal. Appl. Pyrol. 67, 237–246. Grupta, V.K., Jain, C.K., Ali, I., Chandra, S., Agarwal, S., 2002. Removal of lindane and malathion from wastewater using bagasse fly ash – a sugar industry waste. Water Res. 36, 2483–2490. Halsall, C.J., Coleman, P.J., Davis, B.J., Burnett, V., Waterhouse, K.S., Harding-Jones, P., Jones, K.C., 1994. Polycyclic aromatic hydrocarbons in UK urban air. Environ. Sci. Technol. 28, 2380–2386. Harrison, R.M., Smith, D.J.T., Luhana, L., 1996. Source apportionment of atmospheric polycyclic aromatics collected from an urban location in Birmingham, UK. Environ. Sci. Technol. 30, 825–832. Hinchee, R.E., Alleman, B.C., 1994. In: Hoeppel, R.E., Miller, R.N. (Eds.), Hydrocarbon Bioremediation. CRC Press, Boca Raton, FL, USA. Lalvani, S.B., Wiltoski, T., Hubner, A., Weston, A., Mandich, N., 1998. Removal of hexavalent chromium and metal cations by a selective and novel carbon adsorbent. Carbon 36, 1219–1226. Mackay, A.A., Gschwend, P.M., 2000. Sorption of monoaromatic hydrocarbons to wood. Environ. Sci. Technol. 34, 839–845. Manoli, E., Samara, C., 1999. Polycyclic hydrocarbons in natural waters: sources, occurrence and analysis. Trends Anal. Chem. 18, 417–428. Mckay, G., 1995. Use of Adsorbents for the Removal of Pollutants from Wastewater. CRC Press, Boca Raton, FL, USA. Nielsen, T., 1996. Traffic contribution of polycyclic aromatic hydrocarbons in the center of a large city. Atmos. Environ. 30, 3481–3490. Noguera, P., Abad, M., Noguera, V., Purchades, R., Maquiera, A., 2000. Coconut coir waste, a new and viable ecologically-friendly peat substitute. Acta Horticult. 517, 279–286. Pandey, A., Soccol, C.R., Nigam, P., Soccol, V.T., 2000. Biotechnological potential of agro-industrial residues. I: sugar cane bagasse. Biores. Technol. 74, 69–80. Pereira Netto, A.D., Moreira, J.C., Xavier, A.E.O.D., Arbilla, G., Ferreira, L.F.V., Oliveira, A.S., Barek, J., 2000. Avaliac¸a˜o da contaminac¸a˜o humana por hidrocarbonetos policı´clicos aroma´ticos (PAHs) e seus derivados nitrados (NPAHs): uma revisa˜o metodolo´gica. Quı´m. Nova. 23, 765–773. Schwarzenbach, R.P., Gschwend, P.M., Imboden, D.M., 1993. Environmental Organic Chemistry. Wiley Interscience, New York, USA. Teixeira, D.E., Florian, A.C., Santana, M.A.E., 1997. Test for natural decay resistance of the sugar cane bagasse particle board. Scientia Forestalis. 52, 29–34. Witt, G., 1995. Polycyclic aromatic hydrocarbons in water and sediment of the Baltic Sea. Mar. Pollut. Bull. 31, 237–248. Zheng, S., Yang, Z., Jo Hyun, D., Park, Y.H., 2004. Removal of chlorophenols from groundwater by chitosan sorption. Water Res. 38, 2315–2322.