Environ Sci Pollut Res DOI 10.1007/s11356-015-4451-5

RESEARCH ARTICLE

The characteristics of phenanthrene biosorption by chemically modified biomass of Phanerochaete chrysosporium Haiping Gu 1 & Xiaoyan Luo 1 & Haizhen Wang 1 & Laosheng Wu 1 & Jianjun Wu 1 & Jianming Xu 1

Received: 9 October 2014 / Accepted: 27 March 2015 # Springer-Verlag Berlin Heidelberg 2015

Abstract The natural (S 0 ) and chemically modified Phanerochaete chrysosporium including the methylation of amino groups (S1), acetylation of hydroxyl groups (S2), lipid removal (S3), esterification of carboxyl groups (S4), and base hydrolysis (S5) were characterized, and their sorption for phenanthrene (PHE) was investigated. The sorption isotherm of PHE on natural biomasses was apparently linear, while it was nonlinear for the modified ones. The partition coefficient (Kd) describing the sorption affinity of PHE by biomasses followed the order of S0 (9.24 L g−1)>S5 (8.94 L g−1)>S1 (7.13 L g − 1 ) > S 2 (6.97 L g −1 ) > S 3 (6.38 L g −1 ) > S 4 (3.51 L g−1) and decreased as temperature increased. The PHE adsorption fitted well to the pseudo-second-order kinetic model, and the sorption capacity was in the order of S5 (2041.5 μg g−1)>S0 (1768.8 μg g−1)>S2 (1570.9 μg g−1)> S1 (1552.9 μg g−1)>S3 (1346.4 μg g−1)>S4 (991.0 μg g−1). Moreover, the π–π and electron donor–acceptor interactions may govern PHE sorption which processed spontaneously and exothermally. The natural and modified biomasses, especially the base hydrolysis treated ones, were economical and effective biosorbents for PHE removal. Keywords Phenanthrene . Phanerochaete chrysosporium . Modification . Biosorption Responsible editor: Philippe Garrigues * Haizhen Wang [email protected] * Jianming Xu [email protected] 1

Institute of Soil and Water Resources and Environmental Science, Zhejiang Provincial Key Laboratory of Subtropical Soil and Plant Nutrition, Zhejiang University, Hangzhou 310058, China

Introduction Polycyclic aromatic hydrocarbons (PAHs) have aroused considerable public concerns with respect to their high bioaccumulation, potential mutagenic, carcinogenic, teratogenic properties, and adverse impacts on ecosystems and human health (Chen et al. 2010; Liu et al. 2010; Zhou et al. 2012). They have been widely detected at elevated concentrations in storm water runoff and wastewater streams (Huang et al. 2006; Li et al. 2010; Yuan et al. 2010). Sorption is regarded to be a useful and economical method for removing organic pollutants (OPs) including PAHs from water (Huang et al. 2006; Li et al. 2010; Tang et al. 2010; Yuan et al. 2010; Xi and Chen 2014a). Active carbon was widely used as a conventional sorbent and showed good performance in PAHs sorption (Li et al. 2010; Tang et al. 2010; Xi and Chen 2014a). However, its notably high cost and complex manufacturing procedure hinder their application in large-scale wastewater treatment (Li et al. 2010; Tang et al. 2010; Xi and Chen 2014a). Hence, innovative alternatives need to be developed by using low-cost, easily available, and highly effective sorbents. In recent years, research attention has focused on natural biosorbents, such as plant products and microorganisms (algae, fungi, and bacteria), which showed broad application in adsorption of heavy metals, dyes, pesticides, and other OPs due to their inherent adsorption capability and costeffectiveness (Aksu 2005; Aksu and Tunc 2005; Huang et al. 2006; Chen et al. 2010; Li et al. 2010; Tang et al. 2010; Zhou et al. 2012; Valili et al. 2013; Xi and Chen 2014a). Meanwhile, researchers demonstrated that the biosorption performance of natural biosorbents can be significantly improved by surface modification (Aksu 2005; Huang et al. 2006; Li et al. 2010; Tang et al. 2010; Zhou et al. 2012; Valili et al. 2013). For example, Li et al. found that the PAHs sorption capacity of pine barks increased after treatments of

Environ Sci Pollut Res

Soxhlet extraction, alkaline saponification, and acid hydrolysis (Li et al. 2010). The treatments of malt spent rootlets with methanol and sodium hydroxide resulted in better sorption efficiency for phenanthrene (PHE) than the natural malt spent rootlets (Valili et al. 2013). However, few studies have been performed on the sorption of PAHs by modified microorganisms. Phanerochaete chrysosporium, a PAH-degrading model fungus, can be easily cultivated, and it produces high biomass yield (Chen and Ding 2012). Previous studies showed that biosorption of PAHs by P. chrysosporium could play a vital role in PAHs removal due to its high carbon content, aliphatic and aromatic components, and numerous chemical groups of the fungal mycelium (Chen et al. 2010; Chen and Ding 2012). Compared to other adsorbents, such as the high-cost active carbon (Li et al. 2010; Tang et al. 2010; Xi and Chen 2014a), P. chrysosporium is a more economical and effective biosorbent to remove PAHs. Nonetheless, the sorption of PAHs by modified P. chrysosporium, the sorption characteristics, and the mechanisms have received considerably less attention. In this study, phenanthrene, a PAH with three benzene rings, was selected as adsorbate because it is widespread in wastewater and surface water (Xi and Chen 2014a; Zeng et al. 2014). The natural (S0) and chemically modified fungal biomasses of P. chrysosporium (S1–S5) were used as biosorbents. The chemical modifications include the methylation of amino groups (S1), acetylation of hydroxyl groups (S2), lipid removal (S3), esterification of carboxyl groups (S4), and base hydrolysis (S5). All biosorbents were characterized by elemental analysis, Fourier transform infrared spectroscopy (FTIR), and Xray photoelectron spectroscopy (XPS). The sorption isotherms, sorption kinetics, and thermodynamic experiments of PHE by the natural and modified biomasses were further investigated.

2014). The volumetric fraction of methanol in each initial solution was less than 0.1 % to avoid cosolvent effect (Wang et al. 2011; Zeng et al. 2014). All the organic solvents used were of HPLC grade, and inorganic chemicals were of analytical grade or better. Preparation of biosorbents P. chrysosporium (collection number 5.776) was purchased from China General Microbiological Culture Collection Center (Beijing). After 3 days of incubation in potato dextrose broth (6 g L−1 potato extract powder, 20 g L−1 dextrose) on a rotary shaker (130 rpm) at 28 °C, mycelial pellets of P. chrysosporium were harvested and washed several times with deionized water. After being freeze-dried, the mycelial pellets were finally ground into particles less than 0.15 mm in diameter. These powdered biomasses (referred to as natural biomasses, S0) were subjected to various chemical treatments for modifications. Methylation of amine groups (S1) was by mixing 1.0 g powdered natural biomasses of P. chrysosporium with 20 mL of formaldehyde and 40 mL of formic acid (Park et al. 2005; Das et al. 2008). The mixture was shaken on a rotary shaker for 6 h (125 rpm, 25 °C). The general reaction for the mixture is described by Eq. 1: RCH2 NH2

HCHOþHCOOH



RCH2 NðCH3 Þ2 þ CO2 þ H2

ð1Þ

where R represents all of the components in the powdered biomasses of P. chrysosporium. Acetylation of hydroxyl groups (S2) was by reflux of 1.0 g powdered biomasses of P. chrysosporium for 10 h with 60 mL acetic anhydride at 80 °C (Das et al. 2008). The reaction that occurred in the acetylation hydroxyl groups can be described by Eq. 2: R‐CH2 OH þ ðCH3 COÞ2 O→R‐CH2 OCOCH3

Materials and methods Chemicals The stock solution of PHE (purity>99 %, Sigma-Aldrich, Shanghai, China) was dissolved in methanol (4.0 mg mL−1). A desired volume of PHE stock solution was mixed with a background solution (0.01 mol L−1 CaCl2, 200 mg L−1 NaN3, pH 6.6) in a volumetric flask to make an initial aqueous solution of PHE (ca. 1000.0 μg L−1). It was diluted with the background solution to make a series of test solutions at various PHE concentrations (ca. 5.0–1000.0 μg L−1) for sorption experiments. The 0.01 mol L−1 CaCl2 was used to simulate the ionic strength of environmental water, and 200 mg L−1 NaN3 was added to inhibit the microbial activity for preventing PHE biodegradation (Chen et al. 2008; Wang et al. 2011; Zeng et al.

þ CH3 COOH

ð2Þ

Lipid removal treatment (S3) was by reflux of 1.0 g powdered biomasses of P. chrysosporium for 6 h with 75 mL acetone at 80 °C (Fu and Viraraghavan 2002). Esterification of carboxyl groups (S4) was by stirring 1.0 g natural biomasses of P. chrysosporium for 48 h (150 rpm, 60 °C) with 250 mL of 99.9 % acidic methanol solution and 3 mL of concentrated hydrochloric acid (Fang et al. 2011). The chemical esterification reaction is shown by Eq. 3: Hþ

RCOOH þ CH3 OH → RCOOCH3 þ H2 O

ð3Þ

Base hydrolysis (S5) was done by reaction of 1.0 g of powdered natural biomasses of P. chrysosporium with 35 mL of 0.1 mol L−1 NaOH for 2 h (130 rpm, 30 °C) by

Environ Sci Pollut Res

the following reaction (Eq. 4) (Fang et al. 2011; Valili et al. 2013): RCOOCH3 þ NaOH→RCOO− þ CH3 OH þ Naþ

ð4Þ

After these reactions, the biomasses (S0–S5) were washed thoroughly, lyophilized, ground again to pass a 0.15-mm sieve, and stored in a desiccator for further experiments. Characterization of the modified biomasses Carbon (C), hydrogen (H), and nitrogen (N) contents of the natural and modified biomasses were determined using a CHN elemental analyzer (Vario MICRO cube, Elementar, Germany). Ash contents of all samples were measured by heating at 750 °C for 6 h (Wang et al. 2006), and the oxygen (O) content was calculated by mass difference. Fourier transform infrared spectroscopy (FTIR) was recorded over a wavenumber range of 4000 to 400 cm−1 by FTIR spectrophotometer (Avatar 370, Thermo Nicolet, USA) with a resolution of 4.0 cm−1 and scanning for 32 times. Samples (1.5 mg) were mixed with 100 mg KBr (to ensure 20–80 % transmittance rate) and compressed into pellets for FTIR analysis. The surface characterizations of the natural and modified biomasses (S0–S5) were also determined using X-ray photoelectron spectroscopy (XPS, ESCALAB 250 Xi, Thermo Fisher, USA) with an Al Kα X-ray source (hν=1486.6 eV). The X-ray source was run at a reduced power of 150 W, and beam spot was 500 μm during each measurement. All binding energies were calibrated by setting neutral C 1s peak at 284.6 eV. The software package, XPS Peak 4.1, was used to fit the XPS spectra peaks.

liquid chromatography (HPLC). The adsorbed PHE by each biosorbent was determined by mass difference in the control and treatment. In addition, sorption kinetics experiments were carried out at 1.0 mg L−1 of PHE during 36 h (2160 min). In order to evaluate the thermodynamic properties of PHE sorption on the natural and modified biomass samples, equilibrium sorption of PHE was conducted at temperatures of 15 °C (288.15 K), 25 °C (298.15 K), and 35 °C (308.15 K). Detailed processes of sorption experiments were the same as the sorption isotherm experiments. Phenanthrene analysis The PHE concentrations were analyzed by Waters Alliance 2695-2475 HPLC system fitted with a Symmetry® C18 column (5 μm, 3.9 × 150 mm) and a fluorescence detector (Waters, Milford, MA, USA). The mobile phase was methanol and water mixture (90:10, v:v) with a flow rate of 1 mL min−1, the column temperature was 30 °C, and the injection volume was 50 μL. The excitation and emission wavelengths of PHE were 254 and 375 nm, respectively. The minimum detectable concentration for PHE in this study was 3.17 μg L−1, and the relative standard deviation (RSD) was 0.56 % (n=5). Model fitting The sorption isotherms of PHE were fit to the following linear and Freundlich models (Xi and Chen 2014b; Zeng et al. 2014). Linear model: qe ¼ K d C e

Sorption experiments Sorption isotherms of PHE to the natural and modified biomasses (S0–S5) were obtained using a batch equilibration technique. The ratio of sorbent to solution was determined by preliminary experiments to achieve 30–70 % uptake of initial PHE at apparent equilibrium to ensure the reliability of experimental results (Kang and Xing 2005). The natural and modified biosorbents (0.003 g) were placed in 25-mL Teflon-lined screw cap brown glass tubes, along with 10 mL of test solutions (5–1000 μg L−1 PHE, pH 6.6). The treatments of each PHE concentration, including the control (without biomass), were run in triplicate. Tubes were shaken at 25 °C and 200 rpm for 12 h, which was long enough for equilibrium according to our pre-experiments. After centrifugation at 4000×g for 10 min, the supernatants were mixed with metha n ol ( 1 : 1 , v: v ) a n d f i l t r at e d t hr o u gh a 0 . 2 2- μ m polytrifluoroethylene (PTFE) membrane. PHE concentrations in the mixed solutions were detected by high performance

ð5Þ

Freundlich model: qe ¼ K f C Ne

ð6Þ

where qe (μg g−1) and Ce (μg L−1) are the solid- and aqueous-phase equilibrium concentrations of PHE, respectively; Kd (L g−1) is the linear sorption coefficient, and Kf [(μg g −1 ) (g L −1 ) −N ] and N (dimensionless) are the Freundlich sorption parameters. Lagergren’s pseudo-first- and pseudo-second-order kinetic models were used to fit the sorption kinetics experimental data (Kaya et al. 2013; Xi and Chen 2014a, b). The pseudo-firstorder kinetic model: 1nðqe −qt Þ ¼ 1n qe −k 1 t

ð7Þ

and the pseudo-second-order kinetic model: . .  t qt ¼ 1 k 2 qe 2 þ t qe

ð8Þ

.

Environ Sci Pollut Res

where qe (μg g−1) and qt (μg g−1) represent the amounts of PHE adsorbed by the sorbents at equilibrium and at time t (min), respectively; k1 and k2 (g μg−1 min−1) are the rate constants of the pseudo-first- and pseudo-second-order sorption, respectively. Van’t Hoff equation (Aksu and Tunc 2005) was employed to describe the thermodynamic sorption properties of PHE: . . ln K d ¼ ‐ΔH o ðR T Þ þ ΔS 0 R ð9Þ where Kd is the sorption coefficient which can be obtained from the linear regression of sorption isotherms (Eq. 5); R is the gas constant (8.314×10−3 kJ mol−1 K−1); △H0 and △S0 are the standard enthalpy and entropy changes in sorption processes at a different temperature T. The change of standard Gibbs free energy (△G0) was calculated by Eq. 10: ΔGo ¼ ‐RT ln K d

ð10Þ

Results and discussion Characterization of the natural and modified biomasses The elemental compositions of the natural and modified biomasses are shown in Table 1. The carbon contents and polarity index [(N+O)/C] of the modified biomasses were higher than that of the natural biomass. It indicated an increase of the surface polar functional groups in the modified biomasses, which was in good agreement with earlier reports (Chen et al. 2008; Fang et al. 2014). The aromaticity (H/C) of the biomasses ranged from 1.837 to 2.103. The high H/C ratio (>1.0) suggested that the biomasses contained a high amount of original organic residues, such as polymeric association, fatty acid, polysaccharides, etc. (Chen et al. 2008). Thus, it can be concluded that there existed abundant functional groups on the surface of the natural and modified biomasses. Results of FTIR revealed the functional group information of the biomasses (Fig. 1). The broad band near 3410 cm−1 was presented in all the biomasses, which represented the Table 1 The elemental analysis and atomic ratios of the natural and modified biomasses

stretching vibration of intermolecular hydrogen-bonded hydroxyl functional groups (Chen et al. 2005). The bands at 2924, 2853, and 1315 cm−1 were assigned mainly to CH2 units in biopolymers (Chen et al. 2008, 2010). The band at 1652 cm−1 was assigned to C=C and C=O stretching in the aromatic ring, and 565 cm−1 was also assigned to aromatic components (Chen et al. 2010). A small band at 1406 cm−1, corresponding to the symmetric stretching of –COO− groups derived from proteins and carboxylated polysaccharides (Chen et al. 2010; Fang et al. 2011), was observed in some of the biomasses. The bands at 1236, 1151, and 1036 cm−1 were related to the vibration of C–O–C and –OH of polysaccharides (Chen et al. 2010; Song et al. 2014; Xi and Chen 2014b). The major functional groups of protein, including the amide I band (C=O stretching), the amide II band (combination of N–H bending and C–N stretching), and the more complex amide III, were also found near 1652, 1545, and 1377 cm−1, respectively (Das et al. 2008; Song et al. 2014). Compared with the natural biomasses (S0), methylation of amino groups (S1) caused reinforcing and sharping vibration of the band at 1377 cm−1. It can be explained by the formatting of CH2 units and nitrogen-containing compounds during the methylation reaction. The band at 1236 cm−1 in the natural biomasses disappeared in the acetylated biomasses (S2), indicating that hydroxyl groups were blocked after acetylation. Moreover, except for the natural (S0) and lipid removal biomasses (S3), the bands at 1718, 1722, 1734, and 1736 cm−1 of the other treatments were related to the stretching vibration of C=O in ester groups (Chen et al. 2010; Fang et al. 2011; Xi and Chen 2014b). In contrast to the natural biomasses, the band near 1406 cm−1 (–COO−) disappeared in the esterified biomasses (S4), which indicated that the carboxyl groups were transformed through esterification reaction. In contrast, the band near 1406 cm−1 (–COO−) vibrated stronger in the basehydrolyzed biomasses (S5). XPS spectra further verified the chemical reactions that have taken place during the modifications. XPS highresolution scans of C 1s, N 1s, and O 1s were expected to obtain the corresponding functional groups. Results clearly showed the presence of C 1s (Fig. 2a), N 1s (Fig. 2b), and

Sorbent

C (%)

H (%)

N (%)

O (%)

(O+N)/C

O/C

H/C

Ash (%)

S0 S1

43.34 41.56

7.03 6.90

5.18 2.41

39.86 47.98

0.792 0.916

0.690 0.867

1.932 1.980

4.60 1.16

S2 S3 S4 S5

43.26 42.71 42.04 40.86

6.67 7.23 6.95 7.21

5.45 6.21 4.45 2.20

42.83 41.02 45.74 46.00

0.850 0.845 0.907 0.891

0.743 0.721 0.817 0.845

1.837 2.016 1.970 2.103

1.80 2.83 0.82 3.74

S0 natural biomass with no treatment, S1 methylation of amine groups, S2 acetylation of hydroxyl groups, S3 lipid removal treatment, S4 esterification of carboxyl groups, S5 base hydrolysis treatment

Environ Sci Pollut Res Fig. 1 Fourier transform infrared spectra of the natural and modified biomasses. The treatments of S0–S5 are the same as shown in Table 1

O 1s (Fig. 2c) peaks in the natural biomasses (S0). The C 1s peak was resolved into the following three component peaks by the software package XPS Peak 4.1 (Yuan et al. 2011; Song et al. 2014): (1) the peaks at 284.4 eV was associated with aliphatic/aromatic carbon C–(C, H) of lipids or amino acid side chains, which was the largest percentage in the spectral band (37.20 %); (2) the peak (34.18 %) at 285.4 eV was attributed to C–(O, N) from alcohol, ether amine, or amide; and (3) the peak of C=O or O–C–O (286.2 eV), as in carboxylate, carbonyl, amide, acetal, or hemiacetal, respectively, accounted for 28.61 % in the spectral band. The N 1s peak at 399.5 eV was related to the nonprotonated nitrogen from amines and amides (Yuan et al.

2011; Song et al. 2014). The O 1s peak at 531.6 eV (54.98 %) was mainly attributed to the O double bonded to C (O=C), as in carboxylate, carbonyl, ester, or amide (Yuan et al. 2011). Another O 1s peak at 532.5 eV (45.02 %) was related to alcohols, hemiacetal, or acetal groups (Yuan et al. 2011; Song et al. 2014). The changes of peaks and quantifications of C 1s, O 1s, and N 1s after the biomass modification are shown in Table 2. Compared with the natural biomasses (S0), the binding energies of C 1s at 284.4 eV was transformed into 284.0 and 284.9 eV in the methylated biomasses (S1) under the influence of electron cloud along with the increasing C–H. In addition, a new N 1s peak (401.7 eV) appeared after methylation, which

was attributed to protonated amines. These changes confirmed the methylation of amino groups in this research. A new C 1s peak (288.6 eV) represented as RO–C=O appeared in the acetylated biomasses (S2), indicating the modification of

0.3720 0.3418 0.2861

1.0000

0.5498 0.4502

284.4 285.4 286.2

399.5

531.6 532.5

399.3 401.7 531.6 532.5

284.0, 284.9 285.8 287.2

Peak (eV)

Peak (eV)

Atomic (%)

S1

S0

0.7777 0.2223 0.4459 0.5541

0.6518 0.2274 0.1208

Atomic (%)

The treatments of S0–S5 are the same as shown in Table 1

C 1s C 1s C 1s C 1s N 1s N 1s O 1s O 1s O 1s

Element

1.0000

0.1503 1.0000

288.6 399.4 532.2

0.2754 0.5743

Atomic (%)

284.6 286.0

Peak (eV)

S2

531.2 532.3

399.4

284.3 285.7 287.3

Peak (eV)

S3

0.2550 0.7450

1.0000

0.4959 0.3723 0.1318

Atomic (%)

399.4 400.4 532.0 533.0

284.4 285.7 287.0

Peak (eV)

S4

0.7020 0.2980 0.5588 0.4412

0.4220 0.4008 0.1772

Atomic (%)

Binding energies (eV), quantization (%) and assignments of X-ray photoelectron spectra of the natural and modified biomasses

Fig. 2 X-ray photoelectron spectra of the natural biomass: a C 1 s, b N 1 s, c O 1 s

Table 2

532.5 536.7

399.6 403.3

284.4 285.9 287.8

Peak (eV)

S5

0.7083 0.2917

0.9067 0.0933

0.4533 0.3128 0.2339

Atomic (%) C–(C,H) C–(O,N) C=O, O–C–O HO–C=O, RO–C=O Nonprotonated nitrogen Protonated nitrogen C=O C–O–C, C–O–H C=O in micromolecule

Assignments

Environ Sci Pollut Res

Environ Sci Pollut Res

PHE biosorption isotherms of the natural and modified biomasses

Fig. 3 Sorption isotherms of phenanthrene on the natural and modified biomasses at 298.15 K. The treatments of S0–S5 are the same as shown in Table 1. The symbol, solid line, and broken line are measured data, Freundlich model predictions, and linear model predictions, respectively

hydroxyl groups (Yuan et al. 2011; Song et al. 2014). For the lipid removal biomasses (S3), the C 1s peak (287.3 eV) and the O 1s peak (531.2 eV) were associated with some nonpolar functional groups (Song et al. 2014), and they decreased from 28.61 % (S0) to 13.18 % (S3) and from 54.98 % (S0) to 25.50 % (S3), respectively. The carboxylate (C 1s peak at 287.0 eV) in biomasses was reduced during carboxyl esterification (S4). The O 1s peak at 531.6 eV disappeared, and a new peak (536.7 eV) was observed after base hydrolysis (S5). It suggested that the ester had transformed to micromolecular ester or carboxylate (Song et al. 2014). In short, the FTIR and XPS spectra results indicated that the natural and modified biomasses were mainly composites of hydroxyl (–OH), carboxyl (–COO−), ester group (O–C= O), primary amine groups (–NH2), amide (CO–NH), carbon– carbon double bond (C = C), and aromatic components. Moreover, the corresponding functional groups were changed after chemical modifications.

Table 3 Fitted parameters of linear and Freundlich sorption models for the phenanthrene sorption isotherms

Sorbent

Linear model

Sorption isotherms of PHE by the natural and modified biomasses are presented in Fig. 3 and Table 3. In addition, evaluation showed that our data did not fit the Langmuir model well; thus, it is not reported in the paper. As shown in Table 3, the sorption isotherm of PHE by the natural biomasses was practically linear because of N = 0.98 (approach to 1). Therefore, the partitioning effects, the macropore diffusion into the accessible site, and the instantaneous utilization of the most readily available adsorbing sites on the sorbent surface (Valili et al. 2013) mainly dominated the sorption process of PHE in the natural biomasses. Nevertheless, the partitioning of PHE to biomasses contributed only to the linear part of the sorption isotherms (Zhang et al. 2014). Compared to the linear model, the Freundlich model was better for describing PHE sorption (R2, 0.995–0.998) by the modified biomasses. The sorption isotherms of PHE by the modified biomasses were nonlinear as further indicated by the Freundlich exponent N ranging from 0.84 to 0.94. The Freundlich exponent N decreased with the increase of nonlinearity due to the existence of heterogeneous sorption sites, structure, and/or composition (Liu et al. 2010; Zeng et al. 2014). Many researchers also reported the nonlinear sorption of PAHs by wood fibers, soil humic acids, pine sawdust, and lignin (Kang and Xing 2005; Huang et al. 2006; Xi and Chen 2014b; Zhang et al. 2014). Wen et al. (2007) pointed out that the polarity and aromaticity of sorbents play a vital role in PAHs sorption nonlinearity. However, no significant correlations were observed between N and the polarity (r=−0.217, p>0.05) and aromaticity (r=−0.875, p>0.05) of the modified biomasses (S1–S5) (Fig. 4). Except for the methylation of amine groups (S1), a significant negative correlation (r= −0.961, p S 0 (1768.8 μg g−1)>S2 (1570.9 μg g−1)>S1 (1552.9 μg g−1)> S3 (1346.4 μg g−1)>S4 (991.0 μg g−1) (Fig. 5). The biomasses in this study showed higher PHE sorption capacity than some of the previously reported sorbents such as the natural fibric peat (803 μg g−1), surfactant-modified peat (854 μg g−1) (Tang et al. 2010), non-saponifiable tea leaf residue fraction (585.22 μg g−1) (Xi and Chen 2014a), and hematite nanoparticles (82.77 μg g−1) (Zeng et al. 2014). Therefore, the biomasses used in this study, especially the base-hydrolyzed ones, are expected to be promising biosorbents to remove PHE. The sorption kinetics of PHE on the natural and modified biomasses were fit to Lagergren’s pseudo-first- and pseudosecond-order kinetic models, and the corresponding parameters (qe, k1, and k2) and regression coefficients (R2) are given in Table 4. The R2 values of the pseudo-second-order kinetic model (0.999 for all biomasses) were higher than those of the pseudo-first-order kinetic model (R2, 0.961–0.994). In addition, the calculated values of qe based on the pseudo-secondorder kinetic model agreed fairly well with the experimental data. Thus, we concluded that the sorption of PHE by the natural and modified biomasses followed the pseudo-second-

Environ Sci Pollut Res Table 4 Kinetic parameters of the Lagergren’s pseudo-first- and second-order models for phenanthrene sorption on the natural and modified biomasses (S0–S5)

Sorbent

S0 S1 S2 S3 S4 S5

Pseudofirst-order model

Pseudo-second-order model

qe (μg g−1)

k1 (g μg−1 min−1)

R2

qe (μg g−1)

k2 (g μg−1 min−1)

R2

1726.6±20.9 1500.5±27.1 1516.0±28.8 1334.2±9.6 986.1±10.2 1973.5±24.1

0.13569±0.01955 0.08323±0.01165 0.09982±0.01672 0.15074±0.01478 0.09964±0.00907 0.11637±0.01437

0.983 0.966 0.961 0.994 0.988 0.983

1770.0±3.0 1574.5±25. 8 1583.3±29.9 1351.3±13.7 1022.4±3.2 2042.4±46.2

0.00034±0.00009 0.00008±0.00003 0.00011±0.00002 0.00048±0.00011 0.00016±0.00003 0.00001±0.00001

0.999 0.999 0.999 0.999 0.999 0.999

The treatments of S0–S5 are the same as shown in Table 1

order model (Fig. 6). As shown in Table 4, the largest sorption rate (k2, 0.00048 g μg−1 min−1) was observed in the treatment of S3, followed by S0 (0.00034 g μg−1 min−1), and then S4 (0.00016 g μg−1 min−1), S2 (0.00011 g μg−1 min−1), S5 (0.00010 g μg−1 min−1), and finally S1 (0.00008 g μg−1 min−1). More recent reports indicated that the hydrophobicity, polarity, and aromaticity of sorbents were the most important factors affecting PAHs uptake capacity (Wen et al. 2007; Yuan et al. 2010; Wang et al. 2011; Zhou et al. 2012; Xi and Chen 2014a; Zhang et al. 2014), while our analyses showed that the sorption capacities and sorption rates (k2) did not correlate to hydrophobicity (O/C), polarity ((N+O)/C), and aromaticity (H/C) of the test biomasses, which implies that there may exist other interactions between the biosorbents and PHE (Wen et al. 2007; Zhang et al. 2014). The analyses of FTIR and XPS in this study revealed the presence of abundant conjugated structures in the surface of natural and modified biomasses, such as aromatic components and carbon–carbon double bond (C=C). These conjugated structures are able to interact with the planar aromatic ring

Fig. 6 Lagergren’s pseudo-second-order kinetics for phenanthrene sorption on the natural and modified biomasses. The treatments of S0– S5 are the same as shown in Table 1

of PHE molecules by π-electrons overlap or π–π interaction in different geometries (face to face, offset, edge to face) when the attractive interactions between π-electrons and σframework outweigh the repulsions between π-electrons (Hunter and Sanders 1990; Wijnja et al. 2004; Huang et al. 2006; Wang et al. 2006, 2011; Chen et al. 2010; Yuan et al. 2010; Zhang et al. 2014). Additionally, the electron donor– acceptor (EDA) interactions between the aromatic structures and functional groups such as carbonyl (–COO−) in the surface of biomasses (electron acceptors) and the aromatic rings in PHE (electron donors) could enhance PHE sorption (Zhang et al. 2014). The π–π or EDA interaction could contribute to the PHE sorption in our study; thus, further studies are needed to confirm the mechanism of biosorption utilizing better and more quantitative ways (e.g., nuclear magnetic resonance). PHE biosorption thermodynamics study Results showed that all sorption isotherms were nonlinear at low temperature (288.15 K) with the Freundlich exponent N that ranged from 0.82 to 0.90 (data not shown), but highly linear at high temperature (308.15 K) with N approximate to 1. Moreover, the PHE sorption capability of each biomass decreased significantly with increasing temperature, as shown by the decreasing slope (Kd) of sorption isotherms fitted by the linear models (Fig. 7). The increase of ln Kd with the increase of 1/T (Fig. 8) further proved this point. Meanwhile, the sorption coefficients (Koc and Kf) also decreased as temperature increased. The average Koc and Kf values at 288.15, 298.15, and 308.15 K were 18.94, 16.63, 13.50 L g−1 and 16.14, 11.35, 5.37 (μg g−1) (g L−1) −N, respectively. According to previous literatures (Saltali et al. 2007; Vijayaraghavan and Yun 2007; Gupta et al. 2010; Zhang et al. 2014), a temperature increase may lead to the damage of active binding sites in the biomass and the decrease of sorption affinity between PHE and biomass, which increases the tendency to desorb adsorbents from the interface to the solution. Thermodynamic parameters such as the standard enthalpy change (△H0), standard entropy change (△S0), and standard

Environ Sci Pollut Res

Fig. 7 Sorption isotherms of phenanthrene on the natural and modified biomasses at different temperatures. The treatments of S0–S5 are the same as shown in Table 1. The symbol, solid line, and broken line are measured

data, Freundlich model predictions, and linear model predictions, respectively

Gibbs free energy change (△G0) were used to further demonstrate whether or not the sorption processes are exothermic and spontaneous. The thermodynamic parameters were determined from the slopes and the intercepts of the plots of ln Kd versus 1/T (Fig. 8), and they are presented in Table 5. The △G0 values ranged from −5.633 to −2.848 kJ mol−1; △H0 and △S0 ranged from −17.184 to −9.462 kJ mol−1 and from −0.040 to −0.009 J mol−1 K−1, respectively. The negative values of △G0 indicate the spontaneous nature of PHE adsorption by all biomasses (Aksu and Tunc 2005; Gupta et al. 2010; Zhang et al. 2014). On the other hand, △H0 values were also negative in

this study, indicating that the sorption of PHE on each biomass was exothermic (Gupta et al. 2010; Zhang et al. 2014). During the biosorption processes, the highest released heat amount (△H0, −17.184 kJ mol−1) was in the base hydrolysis treatment (S5), which consisted the strongest biosorption ability. Negative values of △S0 revealed the decreasing randomness at the solid–solution interface during the interaction of the PHE with the active sites of the biosorbent (Vijayaraghavan and Yun 2007; Gupta et al. 2010). In conclusion, an increase of temperature would lead to the decrease of sorption capability. Moreover, biosorption of PHE on both the natural and modified biomasses studied were spontaneous and exothermic. PHE sorption mechanism

Fig. 8 Plot of ln Kd versus 1/T for phenanthrene sorption on the natural and modified biomasses at different temperatures. The treatments of S0– S5 are the same as shown in Table 1

As discussed above, π-electrons overlap, π–π interaction, and EDA interaction between biomass surfaces and PHE were widely proposed to dominate PHE sorption (Huang et al. 2006; Wang et al. 2006; Chen et al. 2010; Yuan et al. 2010; Zhang et al. 2014). In this study, the largest PHE sorption capacity of the biomasses after base hydrolysis (S5) could be mainly attributed to the enhanced EDA interactions between the increased carbonyl (–COO−) and PHE after base hydrolysis. Additionally, the swollen lignocellulosic materials after base hydrolysis could also enhance PHE adsorption due to increase in internal surface area, decrease in crystallinity and polymerization degree, separation of the structural links between polymers and hydrocarbons, and structure destruction of polymers (Valili et al. 2013). Therefore, a high PHE uptake

Environ Sci Pollut Res Table 5 The thermodynamic parameters of phenanthrene sorption on the natural and modified biomasses (S0–S5)

Parameters

T (K)

△H0 (kJ mol−1) △S0 (kJ mol−1 K−1) △G0 (kJ mol−1)

288.15 298.15 308.15

Sorbent S0

S1

S2

S3

S4

S5

−16.432

−9.851

−7.662

−10.216

−9.462

−17.184

−0.037 −5.667 −5.512 −4.915

−0.019 −4.930 −4.869 −4.586

−0.009 −4.955 −4.813 −4.768

−0.019 −4.709 −4.594 −4.325

−0.021 −3.279 −3.112 −2.849

−0.040 −5.633 −5.430 −4.827

The treatments of S0-S5 are the same as shown in Table 1 △H0 standard enthalpy change, △S0 standard entropy change, △G0 standard Gibbs free energy change

capability of the biomasses after base hydrolysis was inevitable. The PHE sorption capacities of the biomasses after acetylation (S2, 1570.9 μg g−1) and methylation (S1, 1552.9 μg g−1) were lower than that of the natural biomasses (S 0 , 1768.8 μg g−1). According to Ji et al. (2013), being strong electron donors, the –OH group or –NH2 group could enhance the π electron density of the aromatic structures and induce strong π–π and EDA interactions for PHE adsorption. In the methylated and acetylated modification, the amino groups and the hydroxyl groups, respectively, were blocked. Thus, the weak π–π and EDA interactions led to low PHE sorption capacities of the biomasses after acetylation (S2) and methylation (S1) (Huang et al. 2006; Yuan et al. 2010; Zhang et al. 2014). Based on the similarity–intermiscibility theory, the lipid-removed biomasses (S3) showed the lower PHE uptake capability (1346.4 μg g−1) due to a decrease of nonpolar functional groups. The esterified biomasses (S4) had the lowest sorption capacity (991.0 μg g−1) as a result of carboxyl reduction (Fig. 1), which decreased the EDA interaction between the biomasses and PHE (Yuan et al. 2010).

Conclusions The natural and modified biomasses, especially the basehydrolyzed biomasses, have great potential to remove PHE with higher sorption capacity, shorter equilibrium time, and without potential adverse environment impacts. Sorption occurred with a predominant linear partition process for the natural biomass (S0), while the modified biomasses (S1–S5) showed relatively nonlinear isotherms. The sorption coefficients (Kf, Kd, Koc) decreased with the increase of temperature. The PHE adsorption was fitted well to the pseudo-second-order kinetic model. Thermodynamic calculations proved that PHE sorption on all biomasses was a spontaneous and exothermic process. The sorption coefficients (Kf, Kd, Koc), N values, sorption capacities, and sorption rates (k2) were uncorrelated to polarity ((N+O)/C) or aromaticity (H/C) of the biomasses. However, the existence of abundant conjugated structures (C=C and

aromatic components) and some functional groups (–OH, – COO−, O–C=O, –NH2, CO–NH) in the biomasses may act as active adsorption sites and adsorb planar structure PHE by the π–π and EDA interactions. Acknowledgments This work was supported by the National Natural Science Foundation of China (41171252, 41090284) and the Fundamental Research Funds for the Central Universities in China. Ethics statement All authors have no potential conflict of interest.

References Aksu Z (2005) Application of biosorption for the removal of organic pollutants: a review. Process Biochem 40:997–1026 Aksu Z, Tunc O (2005) Application of biosorption for penicillin G removal: comparison with activated carbon. Process Biochem 40: 831–847 Chen B, Ding J (2012) Biosorption and biodegradation of phenanthrene and pyrene in sterilized and unsterilized soil slurry systems stimulated by Phanerochaete chrysosporium. J Hazard Mater 229:159– 169 Chen B, Johnson EJ, Chefetz B, Zhu L, Xing B (2005) Sorption of polar and nonpolar aromatic organic contaminants by plant cuticular materials: role of polarity and accessibility. Environ Sci Technol 39: 6138–6146 Chen B, Zhou D, Zhu L (2008) Transitional adsorption and partition of nonpolar and polar aromatic contaminants by biochars of pine needles with different pyrolytic temperatures. Environ Sci Technol 42:5137–5143 Chen B, Wang Y, Hu D (2010) Biosorption and biodegradation of polycyclic aromatic hydrocarbons in aqueous solutions by a consortium of white-rot fungi. J Hazard Mater 179:845–851 Das SK, Bhowal J, Das AR, Guha AK (2006) Adsorption behavior of rhodamine B on Rhizopus oryzae biomass. Langmuir 22:7265–7272 Das SK, Ghosh P, Ghosh I, Guha AK (2008) Adsorption of rhodamine B on Rhizopus oryzae: role of functional groups and cell wall components. Colloid Surf B Biointerfaces 65:30–34 Fang L, Zhou C, Cai P, Chen W, Rong X, Dai K, Liang W, Gu J, Huang Q (2011) Binding characteristics of copper and cadmium by cyanobacterium Spirulina platensis. J Hazard Mater 190:810–815 Fang Q, Chen B, Lin Y, Guan Y (2014) Aromatic and hydrophobic surfaces of wood-derived biochar enhance perchlorate adsorption via hydrogen bonding to oxygen-containing organic groups. Environ Sci Technol 48:279–288

Environ Sci Pollut Res Fu Y, Viraraghavan T (2002) Dye biosorption sites in Aspergillus niger. Bioresour Technol 82:139–145 Gupta VK, Rastogi A, Nayak A (2010) Biosorption of nickel onto treated alga (Oedogonium hatei): application of isotherm and kinetic models. J Colloid Interface Sci 342:533–539 Huang L, Boving T, Xing B (2006) Sorption of PAHs by aspen wood fibers as affected by chemical alterations. Environ Sci Technol 40: 3279–3284 Hunter CA, Sanders JKM (1990) The nature of π–π interactions. J Am Chem Soc 14:5525–5534 Ji L, Chen W, Xu Z, Zheng S, Zhu D (2013) Graphene nanosheets and graphite oxide as promising adsorbents for removal of organic contaminants from aqueous solution. J Environ Qual 42:191–198 Kang S, Xing B (2005) Phenanthrene sorption to sequentially extracted soil humic acids and humins. Environ Sci Technol 39:134–140 Kaya EMO, Ozcan AS, Gok O, Ozcan A (2013) Adsorption kinetics and isotherm parameters of naphthalene onto natural- and chemically modified bentonite from aqueous solutions. Adsorption 19:879–888 Li Y, Chen B, Zhu L (2010) Enhanced sorption of polycyclic aromatic hydrocarbons from aqueous solution by modified pine bark. Bioresour Technol 101:7307–7313 Liu W, Xu S, Xing B, Pan B, Tao S (2010) Nonlinear binding of phenanthrene to the extracted fulvic acid fraction in soil in comparison with other organic matter fractions and to the whole soil sample. Environ Pollut 158:566–575 Park D, Yun Y, Park JM (2005) Studies on hexavalent chromium biosorption by chemically-treated biomass of Ecklonia sp. Chemosphere 60:1356–1364 Saltali K, Sari A, Aydin M (2007) Removal of ammonium ion from aqueous solution by natural Turkish (Yildizeli) zeolite for environmental quality. J Hazard Mater 141:258–263 Song C, Sun X, Xing S, Xia P, Shi Y, Wang S (2014) Characterization of the interactions between tetracycline antibiotics and microbial extracellular polymeric substances with spectroscopic approaches. Environ Sci Pollut Res 21:1786–1795 Tang X, Zhou Y, Xu Y, Zhao Q, Zhou X, Lu J (2010) Sorption of polycyclic aromatic hydrocarbons from aqueous solution by hexadecyltrimethylammonium bromide modified fibric peat. J Chem Technol Biotechnol 85:1084–1091

Valili S, Siavalas G, Karapanagioti HK, Manariotis ID, Christanis K (2013) Phenanthrene removal from aqueous solutions using wellcharacterized, raw, chemically treated, and charred malt spent rootlets, a food industry by-product. J Environ Manag 128:252–258 Vijayaraghavan K, Yun Y (2007) Utilization of fermentation waste (Corynebacterium glutamicum) for biosorption of Reactive Black 5 from aqueous solution. J Hazard Mater 141:45–52 Wang X, Sato T, Xing B (2006) Competitive sorption of pyrene on wood chars. Environ Sci Technol 40:3267–3272 Wang X, Guo X, Yang Y, Tao S, Xing B (2011) Sorption mechanisms of phenanthrene, lindane, and atrazine with various humic acid fractions from a single soil sample. Environ Sci Technol 45:2124–2130 Wen B, Zhang J, Zhang S, Shan X, Khan SU, Xing B (2007) Phenanthrene sorption to soil humic acid and different humin fractions. Environ Sci Technol 41:3165–3171 Wijnja H, Joseph P, Malekani K (2004) Formation of π–π complexes between phenanthrene and model π-acceptor humic subunits. J Environ Qual 33:265–275 Xi Z, Chen B (2014a) The effect of structural compositions on the biosorption of phenanthrene and pyrene by tea leaf residue fractions as model biosorbents. Environ Sci Pollut Res 21:3318–3330 Xi Z, Chen B (2014b) Removal of polycyclic aromatic hydrocarbons from aqueous solution by raw and modified plant residue materials as biosorbents. J Environ Sci 26:737–748 Yuan M, Tong S, Zhao S, Jia C (2010) Adsorption of polycyclic aromatic hydrocarbons from water using petroleum coke-derived porous carbon. J Hazard Mater 181:1115–1120 Yuan S, Sun M, Sheng G, Li Y, Li W, Yao R, Yu H (2011) Identification of key constituents and structure of the extracellular polymeric substances excreted by Bacillus megaterium TF10 for their flocculation capacity. Environ Sci Technol 45:1152–1157 Zeng F, He Y, Lian Z, Xu J (2014) The impact of solution chemistry of electrolyte on the sorption of pentachlorophenol and phenanthrene by natural hematite nanoparticles. Sci Total Environ 466:577–585 Zhang M, Ahmad M, Lee S, Xu L, Ok YS (2014) Sorption of polycyclic aromatic hydrocarbons (PAHs) to lignin: effects of hydrophobicity and temperature. Bull Environ Contam Toxicol 93:84–88 Zhou Y, Chen L, Wang X, Xu Y, Lu J (2012) Adsorption of phenanthrene by quaternary ammonium surfactant modified peat and the mechanism involved. Water Sci Technol 66:810–815

The characteristics of phenanthrene biosorption by chemically modified biomass of Phanerochaete chrysosporium.

The natural (S0) and chemically modified Phanerochaete chrysosporium including the methylation of amino groups (S1), acetylation of hydroxyl groups (S...
1MB Sizes 3 Downloads 9 Views