The removal of fluoranthene by Agaricus bisporus immobilized in Ca-alginate modified by Lentinus edodes nanoparticles

Abstract

Fruiting bodies of Agaricus bisporus (A. bisporus) were entrapped in Ca-alginate modified by Lentinus edodes nanoparticles (CA-LENP) to adsorb and biodegrade fluoranthene (FLU) efficiently from an aqueous solution in a fluidized bed bioreactor. A modification of alginate with 3% LENP could reduce the leakage of the fruiting bodies and enzymes from matrix beads effectively during the immobilization, and then increase the biodegradation efficiency. The removal efficiency of FLU reached 95.20 ± 0.66% with an immobilized A. bisporus volume of 2% and a hydraulic retention time of 12 h. The biosorption, biodegradation and removal of FLU were well fitted to the pseudo-second-order, first-order reaction and third-order reaction model, respectively. Langmuir and Freundlich models could describe the biosorption equilibrium data very well. The calculated thermodynamic parameters demonstrated that the biosorption of FLU onto CA-LENP beads was spontaneous and endothermic in nature. The CA-LENP beads with A. bisporus were reused 10 times and ultimately removed more than 88.86% of FLU. The enzymatic activity of CA-LENP beads with the fruiting bodies of A. bisporus could retain almost 66.84% of the initial activity after 60 days of storage at 25 °C. These results indicate that the CA-LENP beads with A. bisporus in FBB can be applied as a wastewater treatment system for the removal of PAHs.

---

1. Introduction

Polycyclic aromatic hydrocarbons (PAHs) are widespread environmental pollutants in aquatic environments and can be released into wastewater from natural and anthropogenic sources.^1–3 The presence of PAHs in the environment poses a serious threat to both the environment and human health owing to their toxic, mutagenic and carcinogenic properties.⁴ Fluoranthene (FLU) is one of the most abundant PAHs and is often detected in the aquatic environment.^5,6 Many rivers and lakes around the world have been found to have high concentrations of FLU which have reached levels of several micrograms per liter.^7–10 Therefore, there is an urgent demand for the effective treatment of FLU contaminated water.

Ozone oxidation,¹¹ photolysis,¹² biodegradation^2,3 and biosorption¹³ have been proposed for the removal of PAHs from the aquatic environment. Unlike physical–chemical treatments, biodegradation is an efficient and economic technique, and generally produces little secondary pollution.^2,14 It has been reported that immobilized extracellular ligninolytic enzymes, such as laccase¹⁵ (E.C. 1.10.3.2) and manganese peroxidase¹⁶ (MnP, E.C. 1.11.1.13), could degrade the FLU efficiently in the environment. However, such pure ligninolytic enzymes are too expensive to produce on a large scale. Extracellular ligninolytic enzymes have been observed in the edible mushroom Agaricus bisporus (A. bisporus) that is widely cultivated in China and is readily available all the year round.¹⁷ The immobilization of A. bisporus offers the possibility for the biodegradation of PAHs with the reduction in cost. Since alginate is inexpensive, biocompatible and easily gelled under mild conditions, alginate has been widely used for cell immobilization to treat different kinds of wastewater,^18–20 thus it can be used to immobilize the fruiting body of A. bisporus. Nevertheless, leaching of enzyme from alginate matrix bead during its fabrication and incubation should be considered seriously.

Adsorption is considered as an effective approach for the removal of PAHs owing to its efficiency, simplicity, and applicability.^21–24 It has been reported that various types of sorbents, such as activated carbon,²⁵ bentonite²⁶ and functional graphene oxide²⁷ could remove PAHs efficiently from the aquatic environment, but the high preparation cost and energy consumption may restrict their practical application in water treatment.^28,29 Due to the wide source of raw materials and technological innovation, Lentinus edodes (L. edodes) nanoparticles (LENP) combine the advantages of nanoparticles adsorbents with low-cost adsorbents, presenting high surface-area-to-volume ratio, high density of reactive sites and low capital cost. In the previous study, Pleurotus ostreatus nanoparticles have been applied as a biosorbent for the removal of Mn(II) from aqueous solution with high biosorption capacity.³⁰ To avoid the release of enzyme from immobilized carrier and improve its biosorption capacity, LENP are applied for the modification of Ca-alginate (CA) matrix beads.

Biotreatment of wastewater in a fluidized bed bioreactor (FBB) is receiving considerable attention. Compared with other bioreactor configurations used in wastewater treatment, FBB shows many advantages including good mixing and efficient mass transfer.³¹ In the FBB, the intimate contact between matrix beads and liquid phase could provide a larger surface area for the substrate diffusion and utilization.³² FBB with immobilized cells has been applied for the treatment of wastewaters like petroleum³¹ and phenol¹⁹ contaminated wastewater successfully. However, the application of immobilized fruiting bodies of A. bisporus in the FBB for the treatment of wastewater is never reported.

Humic substances generated by microbial degradation of plants and animals, are usually observed in natural water bodies. Besides, humic acid (HA) can interact with organic contaminants and inhibit the activities of many enzymes.^14,33 Hence, it is indispensable to investigate the effect of HA on the removal of FLU from aqueous solution.

In this study, CA-LENP beads with fruiting bodies of A. bisporus were prepared to compare their enzymatic activities with the entrapped pure enzyme in the same material and to treat FLU contaminated water in FBB. Fourier transform infrared spectrometry (FTIR), Scanning Electron Microscope (SEM) and Brunauer–Emmett–Teller (BET) method were employed to characterize the LENP, CA and CA-LENP beads. The work investigated the effect of different parameters including dosage of CA-LENP beads with A. bisporus, hydraulic retention time (HRT) and HA on the removal efficiency of FLU by immobilized A. bisporus in FBB. The mechanism for biosorption and biodegradation of FLU by CA-LENP beads was studied on the basis of biosorption isotherm models, reaction kinetic models and thermodynamic analysis. Meanwhile, the storage stability and reusability of immobilized A. bisporus were also studied.

2. Experimental section

2.1. Reagents and materials

Fresh A. bisporus and L. edodes were supplied by a mushroom production base in the suburbs of Chengdu, Sichuan Province, China. The laccase from A. bisporus with an activity of 9.86 U mg⁻¹ protein and its substrate 2,2-azinobis-3-ethylbenzthiazoline-6-sulfonate (ABTS, ≥99% purity), FLU (≥97% purity), sodium alginate and HA were purchased from Sigma-Aldrich. Other reagents were of analytical grade and acquired from Kelong Chemical Reagent Factory (Chengdu, Sichuan Province). All solutions were prepared using the ultrapure water. The stock solution of FLU was prepared in analytical grade acetone at a concentration of 1.0 g L⁻¹, and then diluted to the appropriate concentration.

2.2. Immobilization

2.2.1. Preparation of mushroom powder and LENP particles. Fresh A. bisporus was washed with ultrapure water and cut into small pieces. Then mushroom pieces were lyophilized and grinded into powder by a pulverizing mill (Joyoung, JYL-350B). The L. edodes was fabricated into nanoparticles by the method described in the previous study.³⁰ The maximum size of LENP was under 300 nm and the arithmetic average of nanoparticles examined by small angle X-ray scattering (MAX-2500, Japan) was 232 nm. The fresh A. bisporus was boiled with distilled water three times for 3–5 min each, filtered with the ultrapure water, and then dried at 50 °C for two days in an oven to obtain the dead A. bisporus.

2.2.2. Immobilization of laccase and fruiting bodies of A. bisporus. Sodium alginate was dissolved in ultrapure water to obtain 10 mL alginate solution containing 1.13 U mL⁻¹ of laccase. Varying concentrations of sodium alginate (2.0%, 2.5%, 3.0%, 3.5%, 4.0% w/v) were prepared to obtain high enzymatic activity during the immobilization of laccase. To modify the alginate matrix bead with LENP, the same amount of laccase was introduced into the alginate solution containing different concentrations (2.0%, 3.0%, 4.0%, 9.0% w/v) of LENP and mixed thoroughly for homogenization. In the case of entrapment of A. bisporus, a certain amount of mushroom powder of A. bisporus (0.1–0.4 g) was added into 10 mL alginate solution with the optimum concentration of sodium alginate and LENP, and the mixture was well stirred for 2 h to ensure the homogeneity of whole solution. Each alginate solution was dropped into stirred solution containing 2.5% (w/v) CaCl₂ and cross-linked for 1 h before forming beads with a mean particle size of 5 mm diameter. Then the matrix beads were washed with ultrapure water and immersed in water at 4 °C to avoid drying and shrinkage from exposure to air. Approximately 53.4 g CA and 49.4 g CA-LENP beads (wet weight) could be got from 100 mL alginate solution in the immobilization of laccase. For the entrapment of fruiting bodies of A. bisporus, the weight of CA and CA-LENP beads prepared from 100 mL alginate solution in this study were about 56.2 g and 52.3 g (wet weight), respectively.

2.3. Characterization of matrix beads

The morphology of CA and CA-LENP beads were characterized using SEM (JSM-7500F, Japan). The specific surface area of matrix bead was measured utilizing BET method (Micromeritics ASAP-2020, America) with nitrogen as the adsorbate. A FTIR Spectrometer (NEXUS-650, America) was used to characterize the surface of LENP before and after biosorption of FLU.

2.4. Activity and stability assays

The activities of laccase and MnP were detected by analyzing the change of absorbance during the catalytic oxidation of ABTS with a UV-vis spectrophotometer (UV 2100, Shimadzu, Japan). The activity of immobilized laccase was measured according to the method previously reported by Daâssi et al.²⁰ MnP activity was determined by incubating the immobilized A. bisporus for 5 min at 25 °C with stirring (120 rpm) in the 3.0 mL of mixture containing 2.2 mL (0.1 M) of sodium malonate, 0.3 mL (0.5 mM) of ABTS, and 0.3 mL (0.2 mM) of MnSO₄, and monitoring the oxidation of ABTS spectrophotometrically. The reaction was initiated by introducing 0.2 mL (0.1 mM) of H₂O₂. The absorbance of oxidized product was determined at 420 nm and corrected for laccase activity. One MnP activity unit (U) was defined as the amount of enzyme oxidizing 1 μmol ABTS per minute. All experiments were carried out in triplicate.

To examine the storage stability of immobilized fruiting bodies of A. bisporus, the residual activities of extracellular ligninolytic enzymes of the immobilized A. bisporus in CA and CA-LENP beads were determined at certain intervals. All the samples were stored in phosphate buffer (pH 4.5) at 25 °C.

2.5. FBB studies

A schematic diagram of the FBB is shown in Fig. 1. The column had the dimensions of 50 mm internal diameter, 500 mm height with a conical bottom. This bioreactor was thermostated and its working volume was 0.8 L.

Fig. 1 A schematic diagram of the FBB.

Initially the bioreactor was operated in batch mode, and the dependence of FLU removal efficiency was studied on the dosage of immobilized A. bisporus (6.05–30.24 g) added into the column. For the batch experiments, the bioreactor was filled up with 0.8 L of synthetic wastewater contaminated by FLU (90 μg L⁻¹) and the immobilized A. bisporus were incubated in FBB for 24 h. The pH of solution was adjusted with HNO₃ (0.1 M) or NaOH (0.1 M) to conduct the biodegradation process. In the batch experiments, the immobilized A. bisporus showed best performance in the FLU removal from solutions at pH 5.0. Moreover, temperature had only a slight effect on the immobilized A. bisporus, which is similar to the results observed by Niu et al.³ Then the FLU removal experiments in this study were performed in the FBB at pH 5.0 and 25 °C with an air velocity of 0.8 cm s⁻¹. After the batch experiments, the bioreactor was switched to continuous flow condition with a 20 mL min⁻¹ flow rate, at which the reaction system was no longer limited by external mass transfer influences. A series of experiments were carried out to investigate the effect of HRT (8, 12, 16, 20, 24 h) and HA concentrations (0–40 mg L⁻¹) on the FLU removal efficiency by the immobilized A. bisporus.

Effluent samples (1 mL) in the bioreactor were decanted at certain intervals and extracted by liquid–liquid extraction (1 : 1 water : benzene) for overnight, and then the aqueous extracts were evaporated under a gentle nitrogen hood. The residues were dissolved in acetone and the concentration of FLU was analyzed using gas chromatograph (Agilent GC model 7890A) with a flame ionization detector and a BP-5 capillary column (30 m × 0.32 mm × 0.25 μm film thickness). The oven program was 80 °C for 2 min followed by ramping at 10 °C min⁻¹ up to 300 °C maintained for 2 min. Both injection and detector temperatures were held at 280 °C. Retention time for FLU was observed at 27 min. The concentration of FLU was measured by comparison against the FLU standard (Sigma-Aldrich).³⁴

To examine the reusability of immobilized A. bisporus, CA and CA-LENP beads were used repeatedly in the FBB to treat the wastewater under the optimum conditions. The matrix beads with fruiting bodies of A. bisporus were removed from the FBB after one assay and washed with ultrapure water three times before being reused for another cycle. All experiments were conducted in triplicate.

2.6. Biosorption experiments

Biosorption experiments were conducted in the FBB under the optimum conditions. In the experiment, biosorption kinetic was determined from solution with 90 μg L⁻¹ FLU. The biosorption isotherm experiments were conducted with an initial FLU concentration ranging from 45 to 1000 μg L⁻¹. Biosorption experiments were also performed to investigate the thermodynamic parameters at different temperatures (15, 25, 35, 45 °C). The amount of adsorbed FLU (q, μg g⁻¹) and the removal efficiency can be calculated by the following equations:where C₀ and C_e are the initial and equilibrium concentration of FLU (μg L⁻¹), respectively; V is the volume of the solution (L) and m is the mass of matrix beads with dead A. bisporus (g). The blank experiments confirmed that the loss of FLU resulted from abiotic processes such as, natural dissipation or volatilization during the incubation was negligible.

The Langmuir model and the Freundlich model were employed to describe the isotherm experiment data. Besides, pseudo-first-order model (PFOM), pseudo-second-order model (PSOM) and intra-particle diffusion model were adopted to analyze the biosorption kinetic data and explore the biosorption mechanism. The equations of all the above-mentioned biosorption isotherm and kinetics models are summarized in Table 1.

Table 1 Models applied for fitting of biosorption kinetics and isotherm as well as removal kinetics in this study^a

Category	Model	Equation	Symbols representation
a qt, biosorption amount at time t (μg g^−1); qe, equilibrium biosorption amount (μg g^−1); C, intercept (μg g^−1); Ce, equilibrium solution phase concentration (μg L^−1); Q, biosorption capacity (μg g^−1); C0, initial solution phase concentration (μg L^−1); Ct, solution phase concentration at time t (μg L^−1).
Sorption kinetics models	Pseudo-first order model	ln(qe − qt) = ln qe − k1t	k1, rate constant of pseudo-first order (min^−1); k2, rate constant of pseudo-second order rate (g μg^−1 min^−1); kid, rate constant of intra-particle diffusion (g μg^−1 min^−1/2) for the adsorption process
	Pseudo-second order model
	Intra particle diffusion model of Weber and Morris	qt = kidt^1/2 + C
Sorption isotherm models	Langmuir isotherm model	,	b, Langmuir constant (L μg^−1); Kf, Freundlich affinity coefficient (μg g^−1); 1/n, Freundlich exponential coefficient (L g^−1); RL, the dimensionless separation factor of equilibrium parameter
	Freundlich isotherm model
Removal and degradation kinetics models	First-order reaction	ln(C0/Ct) = k′1t + C	k′1, rate constant of first-order reaction (min^−1); k′2, rate constant of second-order reaction (L μg^−1 min^−1); k′3, rate constant of third-order reaction (L^2 μg^−2 min^−1)
	Second-order reaction	(1/Ct − 1/C0) = k′2t + C
	Third-order reaction	(1/Ct^2 − 1/C0^2) = 2k′3t + C

---

2.7. Biodegradation kinetic experiments

The biodegradation kinetic experiments with initial FLU concentration of 90 μg L⁻¹ were performed in the FBB under the optimum conditions. Variations in the biodegradation efficiency for FLU with the incubation time by immobilized A. bisporus were studied in kinetic experiments. The biodegradation efficiency could be expressed as follows:

Biodegradation efficiency (%) = 1 − A₁ − A₂ (3)

where A₁ is the mass ratio of FLU retained in the aqueous solution; A₂ is the mass ratio of FLU retained in the CA or CA-LENP beads. The first-order reaction, second-order reaction and three-order reaction models were applied to investigate the removal and biodegradation of FLU by immobilized A. bisporus.

2.8. Statistical analysis

All the experiments were repeated three times, and the data reported in this study referred to the mean values. Data were analyzed by analysis of variance (ANOVA) and Duncan's test was operated to study the statistical significance of differences (P < 0.05) between means. All statistical analysis was operated by the Social Sciences (SPSS) 19.0 software and error bars representing the standard deviation (SD) were supplied in figures wherever possible. The linear regression method was adopted to analyze the models' parameters.

3. Results and discussion

3.1. Immobilization

3.1.1. Immobilization of laccase. As shown in Table 2, the activity of entrapped laccase in the CA beads increased from 0.77 ± 0.12 to 0.88 ± 0.09 U g⁻¹ with the increasing alginate concentration from 2.0% to 3.0% (w/v). However, there was a gradual decline in the activity of entrapped laccase with 3.5% and 4.0% sodium alginate, which could be attributed to the fact that the high sodium alginate concentration might interfere with the entry of enzymes into matrix beads leading to lower immobilization efficiency ultimately.³⁵ Furthermore, the activity of laccase in the gelling solution (the CaCl₂ solution after the formation of beads) decreased with the increasing sodium alginate concentration (2.0–3.5%), indicating the reduction in the leaching of enzyme from CA bead. This can be ascribed to the different pore size of the beads. Riaz et al.³⁶ has reported that the lower concentration of sodium alginate solution resulted in the greater pore size of the beads facilitating the leakage of enzyme from the beads. The highest activity of entrapped laccase occurred at 2.5% and 3.0% sodium alginate concentration and the activity observed in their corresponding gelling solution was relatively lower (Table 2). Hence, the most suitable concentration of alginate solution was 2.5% for the immobilization of laccase. This is in accordance with the results reported by Daâssi et al.²⁰

Table 2 The activities of immobilized laccase in CA and CA-LENP beads (d = 5.0 mm) with different contents of CA (w/v) and LENP (w/v) and the activities of laccase in their corresponding gelling solution^a

Carrier	Activity (U g^−1 beads)	Activity in gelling solution (U mL^−1)
a Results are expressed as mean ± SD of three independent experiments. Means in the same column with different letters are significantly different from each other (P < 0.05) according to the Duncan's test.
2.0% CA	0.77 ± 0.12 a	0.24 ± 0.06 d
2.5% CA	0.86 ± 0.04 a	0.18 ± 0.03 cd
3.0% CA	0.88 ± 0.09 a	0.17 ± 0.02 cd
3.5% CA	0.82 ± 0.04 a	0.15 ± 0.03 bcd
4.0% CA	0.79 ± 0.06 a	0.21 ± 0.02 cd
CA-1% LENP	1.68 ± 0.12 b	0.13 ± 0.13 bc
CA-2% LENP	2.02 ± 0.11 c	0.06 ± 0.01 ab
CA-3% LENP	2.19 ± 0.07 d	0.01 ± 0.01 a
CA-4% LENP	2.21 ± 0.06 d	0.03 ± 0.01 a
CA-9% LENP	1.87 ± 0.11 c	0.06 ± 0.02 ab

---

For better manipulation of matrix bead permeability and retention of immobilized laccase, LENP were used to modify the CA matrix beads. The activity of entrapped laccase improved considerably with the increasing LENP concentration from 1.0% to 3.0% (w/v), but the improvement slowed down when LENP concentration was higher than 3.0%. In addition, the activity in gelling solution decreased with the increase of LENP concentration. When the LENP concentration was 9.0%, the enzymatic activity of immobilized laccase in CA beads was relatively lower, which could be attributed to the decrease in the homogeneous distribution of laccase in the bead.³⁷ These results suggested that 3.0% of LENP was sufficient for the modification of CA beads (Table 2). In comparison to the entrapped laccase in CA beads, the enzymatic activity in the gelling solution for the entrapped laccase in CA-LENP beads was much lower. Besides, the enzymatic activity of the entrapped laccase in CA-LENP beads was much higher than that in CA beads, suggesting that CA-LENP beads showed a better performance in the retention of laccase. Therefore, CA-LENP beads with 3.0% of LENP and 2.5% of sodium alginate were applied for further investigations.

3.1.2. Immobilization of fruiting bodies of A. bisporus. When 100 mg mushroom powder of A. bisporus was used to be immobilized in CA and CA-LENP beads, the immobilized A. bisporus in CA and CA-LENP beads showed similar activity to the entrapped laccase in the same material, suggesting that the fruiting bodies of A. bisporus were successfully entrapped in the matrix beads (Tables 2 and 3). Moreover, the enzymatic activity in the gelling solution for immobilized A. bisporus in CA bead was lower than the entrapped laccase in CA bead, indicating that the loss of enzymatic activity in the gelling solution during the immobilization of fruiting body was relatively lower. For these reasons, the immobilization of fruiting body might be more efficient than the entrapment of purified laccase.

Table 3 The enzymatic activities of immobilized fruiting bodies of A. bisporus (laccase, 0.12 U mg⁻¹ dry weight of fruiting body; MnP, 0.05 U mg⁻¹ dry weight of fruiting body) in CA beads and CA-LENP beads (d = 5.0 mm) and the enzymatic activities in their corresponding gelling solution^a

Carrier	Concentration of A. bisporus (g mL^−1 polymer solution)	Activity (activity (U g^−1 beads))		Activity in gelling solution (U mL^−1)
		MnP	laccase	MnP	laccase
a Results are expressed as mean ± SD of three independent experiments. Means in the same column with different letters are significantly different from each other (P < 0.05) according to the Duncan's test.
CA	10	0.39 ± 0.12 a	0.93 ± 0.09 a	0.05 ± 0.02 ab	0.08 ± 0.03 b
CA-LENP	10	0.68 ± 0.07 b	2.12 ± 0.13 b	0.03 ± 0.01 a	0.02 ± 0.01 a
CA-LENP	40	0.93 ± 0.07 c	2.71 ± 0.09 c	0.06 ± 0.01 b	0.09 ± 0.02 b

---

When the concentration of A. bisporus increased from 10 to 40 g mL⁻¹, the enzymatic activities of laccase and MnP of matrix beads with A. bisporus were enhanced to 2.71 ± 0.09 and 0.93 ± 0.07 U mg⁻¹ beads, respectively, but they did not increase proportionally to the amount of mushroom of A. bisporus (Table 3). This is because of the diffusion resistance of the matrix beads. Furthermore, the enzymatic activity in the gelling solution for the CA-LENP beads with A. bisporus increased with the increase in the concentration of A. bisporus. Therefore, the immobilized A. bisporus in CA-LENP beads with 10 g mL⁻¹ of mushroom which could achieve similar activity as the entrapped laccase, was more suitable for the biodegradation of FLU from aqueous solution.

3.2. Characterization of LENP particles and matrix beads

According to the BET analysis, the CA beads had a specific surface area of 28.4 m² g⁻¹ with a mean pore diameter of 39.7 nm. Compared with CA beads, CA-LENP beads showed a larger BET specific surface area (66.3 m² g⁻¹) and a lower mean pore diameter (10.5 nm).

Fig. 2 represents the FTIR spectra of LENP before and after biosorption of FLU. The broad biosorption peak around 3384 cm⁻¹ was indicative of –OH stretching of carboxylic groups and stretching of –NH groups. The peaks at 2925 and 2867 cm⁻¹ were related to the –CH₂ antisymmetric and asymmetric stretching vibration, respectively.³⁸ The strong biosorption band near 1653 cm⁻¹ was assigned to the stretching vibration of –COO. In addition, the band of 1040 cm⁻¹ arose from C–OH stretching vibration.³⁰ The peaks at 1557 and 1454 cm⁻¹ were correlated with the stretching bond of the aromatic rings.³⁹ These results showed the existence of hydroxide (–OH), amide (–NH₂) and carboxyl (–COOH) on the surface of LENP. Compared with LENP before biosorption, the significant peaks at 1541 and 1456 cm⁻¹ on the FTIR spectra of LENP after biosorption indicated that FLU had been successfully adsorbed onto the surface of LENP. Besides, the increase in the biosorption peak intensity of 3432, 1653 and 1038 cm⁻¹ implied that the hydroxide (–OH) and carboxyl (–COOH) might be involved in the biosorption process.

Fig. 2 FTIR spectra of LENP before biosorption of FLU (a) and after biosorption of FLU (b).

According to the SEM image of LENP, LENP were spherical or approximately spherical and had a rough surface (Fig. 3a). It was observed that the surface of CA bead was rough, rugged and opaque, and also had many pores which could result in the leakage of enzyme leading to a lower enzymatic activity (Fig. 3b). CA-LENP bead also showed a rough surface with many gibbose parts, but the pore size and the number of pores on its surface were much smaller than that on CA bead (Fig. 3c). The surface modification of CA bead by LENP could form a physical barrier to avoid the leaching of enzyme.⁴⁰ The existence of LENP contributed to the reduced pores on the surface of bead and the large specific surface area of CA-LENP bead as well as the high biosorption capacity of FLU.

Fig. 3 SEM of LENP (a), CA beads (b) and CA-LENP beads (c).

3.3. Factors affecting the removal of FLU from aqueous solution

3.3.1. Effect of dosage of matrix beads. Fig. 4 depicts the FLU removal efficiency for different dosages of matrix beads after reacting for 24 h in the FBB. The removal efficiency of FLU was obviously enhanced with the increase in the quantity of CA-LENP beads. However, the increase in the dosage of CA-LENP beads might increase the cost and hinder the movement of beads leading to the reduction in the removal efficiency, so a larger number of CA-LENP beads were not examined in this study. It was shown that the increase in FLU removal efficiency slowed down as the quantity of CA-LENP beads higher than 12.10 g. From the economic point of view, 12.10 g of CA-LENP beads, which is equivalent to a volume of 18 mL of beads (2% of the total bioreactor volume), was suitable for the wastewater treatment in FBB.

Fig. 4 Effect of dosage of immobilized fruiting bodies of A. bisporus in CA-LENP beads on the removal efficiency of FLU (initial FLU concentration: 90 μg L⁻¹, pH: 5.0, reaction time: 24 h, temperature: 25 °C).

3.3.2. Effect of HRT. HRT is a very important parameter to be considered when the FBB was employed for the wastewater treatment. As illustrated in Fig. 5, the biodegradation and removal efficiencies of FLU by immobilized A. bisporus increased with the increase in HRT (P < 0.05). Moreover, the biodegradation efficiency of CA-LENP beads was higher than that of CA beads, which may be due to the higher enzymatic activity and lager specific surface area of CA-LENP beads. The maximum removal efficiency of 98.83 ± 0.78% for CA-LENP beads with A. bisporus was achieved at 24 h, whereas a removal efficiency of 95.20 ± 0.66% occurred when HRT was 12 h, after which the difference in FLU concentrations was negligible. For the immobilized A. bisporus in CA beads, the removal efficiency reached a plateau as HRT increased to 12 h. Therefore, the optimum HRT for the immobilized A. bisporus in the FBB should be 12 h. When the bioreactor was switched from batch mode to continuous mode, the removal rate of FLU for the immobilized A. bisporus improved significantly (Fig. 4 and 5), presenting the merit of FBB in the application of wastewater treatment.

Fig. 5 Effect of HRT on the biodegradation and removal efficiencies of FLU by CA and CA-LENP beads with fruiting bodies of A. bisporus (dosage of immobilized A. bisporus: 12.10 g, initial FLU concentration: 90 μg L⁻¹, pH: 5.0, temperature: 25 °C). Different lowercase letters denote the significant difference between HRT at P < 0.05.

3.3.3. Effect of HA. Fig. 6 represents the biodegradation and removal efficiencies of FLU from aqueous solution with different concentrations of HA. HA powders were dissolved in ultrapure water to prepare the HA solution, and the concentration was determined by TOC analyzer. The removal efficiency of FLU by immobilized A. bisporus decreased with the increase in HA concentration (P < 0.05). Since HA molecules can also be adsorbed by matrix beads, the HA will compete with FLU for the biosorption.⁴¹ As a result, there are fewer active sites for the biosorption of FLU. Nevertheless, the removal efficiency of FLU increased when the HA concentration exceeded 30 mg L⁻¹, at which FLU would be adsorbed by the HA molecules. Although HA shows an inhibitory effect on the activities of many enzymes, it can interact with FLU forming larger aggregates which are too large to pass through the pores on the surface of matrix beads. Consequently, the effect of high concentrations of HA on the biodegradation of FLU by immobilized A. bisporus was insignificant. The biodegradation efficiency of CA-LENP bead with A. bisporus was higher than that of CA bead with A. bisporus owing to the small pores and high enzyme activity of CA-LENP bead with A. bisporus.

Fig. 6 Effect of HA on the biodegradation and removal efficiencies of FLU by CA and CA-LENP beads with fruiting bodies of A. bisporus (dosage of immobilized A. bisporus: 12.10 g, HRT: 12 h, initial FLU concentration: 90 μg L⁻¹, pH: 5.0, temperature: 25 °C). Different lowercase letters denote the significant difference between humic acid concentrations at P < 0.05.

3.4. Biosorption experiments

3.4.1. Biosorption kinetics. As shown in Fig. 7a, the biosorption of FLU by CA and CA-LENP beads with dead A. bisporus reached the equilibrium after incubation for 8 h, at which 71.81 ± 0.64% and 81.06 ± 0.62% of FLU were adsorbed from solution, respectively. The biosorption rate of CA-LENP beads was faster than that of CA beads owing to the larger surface area of CA-LENP beads and the biosorption of FLU by LENP.

Fig. 7 The biosorption efficiency of FLU for CA and CA-LENP beads with dead A. bisporus (a), and the biodegradation and removal efficiencies of FLU for immobilized A. bisporus in CA and CA-LENP beads (b) (dosage of immobilized A. bisporus: 12.10 g, HRT: 12 h, initial FLU concentration: 90 μg L⁻¹, pH: 5.0, temperature: 25 °C).

In this study, PFOM⁴² and PSOM⁴³ were applied to describe biosorption kinetics of FLU from aqueous solution on beads. All the parameters of these two biosorption kinetic models were given in Table 4. Compared with the PFOM, PSOM can describe the biosorption behavior of FLU onto beads better, which could be confirmed in Fig. 8a. PFOM predicted a considerable lower value of the equilibrium uptake capacity than the experimental data and the values of linear correlation coefficients (R²) of PFOM were below 0.97, while the R² values for the PSOM reached 0.99 and the calculated q_e values at equilibrium from the PSOM were very close to the experimental data. For these reasons, the biosorption behavior of FLU onto beads adheres to the PSOM, suggesting that the chemical interactions may be involved in the biosorption process.¹⁴ Both the FLU molecules and the C=O groups on the surface of beads contain the π electrons, which can result in the π–π bonding interactions between them. Furthermore, the aromatic ring in FLU can act as π electron donors and form π–π electron-donor–acceptor interaction with the aromatic structures in the polymer surface.⁴⁴ In addition, the –OH groups on the polymer surface can form hydrogen bonds with FLU.²¹ Thus, the chemical interactions involved in biosorption processes might include the hydrogen bonding interactions and π–π bonding interactions.

Table 4 Kinetic parameters for biosorption of FLU on CA and CA-LENP beads with dead A. bisporus

	qe (exp.) (μg g^−1)	Pseudo-first-order			Pseudo-second-order			Intra particle diffusion
		qe (cal.) (μg g^−1)	k1 × 10^2 (min^−1)	R^2	qe (cal.) (μg g^−1)	k2 × 10^2 (g μg^−1 min^−1)	R^2	kid (g μg^−1 min^−1/2)	C	R^2
CA-LENP	4.86 ± 0.09	2.64	0.64	0.94	5.06	0.57	0.99	0.21	1.12	0.95
CA	4.09 ± 0.16	2.86	0.73	0.96	4.29	0.26	0.99	0.19	0.57	0.97

---

Fig. 8 Pseudo-second-order model plots (a) and intra-particle diffusion model plots (b) for the biosorption of FLU onto the CA and CA-LENP beads with dead A. bisporus.

To investigate the rate-limiting step of the biosorption process, intra-particle diffusion model was also employed in the present study.⁴⁵ The model predicts that if the plots of q versus t^1/2 pass through the origin, intra-particle diffusion will be the only rate-limiting step. As illustrated in Fig. 8b, there were three separate linear regions about the biosorption of FLU on matrix beads, demonstrating that more than one process affected the biosorption. The plots indicated that there were three different stages during the biosorption process, which might begin from external mass transfer, followed by intra-particle diffusion and finally came into a plateau for the lower concentration of FLU.⁴⁶ Since the straight lines of second and third stages did not pass through the origin, the biosorption process was controlled by more than one mechanism. The kinetic parameters of the second stage were summarized in Table 4. There were a higher value of correlation coefficient (0.97) and a large pore size for CA bead, which implied that intra-particle diffusion or pore-filling played an important role in controlling the kinetics of the biosorption process. For the CA-LENP bead, the hydrogen bonding interaction and π–π bonding interaction might be the main biosorption mechanism due to its high specific area and small pore volume.

3.4.2. Biosorption isotherm experiments. In this study, Langmuir and Freundlich models are applied to describe isotherm data.^47,48 The parameters of Langmuir and Freundlich models for the biosorption isotherms are shown in Table 5. The biosorption isotherms were well fitted with Langmuir model (R² = 0.96) and Freundlich model (R² = 0.99), which demonstrated that both monomer and multilayer biosorption were involved in the biosorption of FLU onto matrix beads. Since the value of b was positive at 298.15 K, all the values of R_L were in the range of 0–1 for initial concentration of FLU ranging from 45 to 1000 μg L⁻¹, indicating that the biosorption of FLU was favorable at the conditions being studied. The biosorption capacity of CA-LENP beads and CA beads for FLU increased with the increase in the equilibrium concentration (Fig. 9), but it did not reach a platform in the FLU concentration range analyzed. As the initial FLU concentration was the driving force to overcome limitation of substrate diffusion between the biosorbent and fluid phases, a high initial concentration of FLU could increase the driving force and improve the biosorption capacity.⁴⁹ Due to the low solubility of FLU, the environmental concentration of FLU was usually lower than the concentrations evaluated in this study. Consequently, higher concentrations were not evaluated. According to the results of Freundlich model (0 < 1/n < 1), there was chemical interaction existing in the biosorption process.¹⁴ It further confirmed that the hydrogen bonding interactions and π–π bonding interactions played key roles in the biosorption process.

Table 5 Biosorption isotherm constants for the biosorption of FLU onto CA and CA-LENP beads with dead A. bisporus

	Langmuir model			Freundlich model
	qmax (μg g^−1)	b × 10^2 (L μg^−1)	R^2	Kf (μg g^−1)	1/n (L g^−1)	R^2
CA-LENP	71.94	0.40	0.96	0.58	0.75	0.99
CA	53.48	0.29	0.96	0.36	0.74	0.99

---

Fig. 9 Biosorption isotherms of FLU onto the CA and CA-LENP beads with dead A. bisporus (dosage of immobilized A. bisporus: 12.10 g, HRT: 12 h, pH: 5.0, temperature: 25 °C).

3.4.3. Biosorption thermodynamics. Thermodynamic parameters related to the biosorption process, such as Gibb's free energy (ΔG°), change in enthalpy (ΔH°), and entropy (ΔS°) can be given by the following equations:⁵⁰

ΔG° = ΔH° − TΔS° (5)

where K represents the equilibrium constant (acquired from Langmuir model, L mol⁻¹); R represents the ideal gas constant (8.314 J K⁻¹ mol⁻¹) and T is the absolute temperature (K). The corresponding thermodynamic parameters are listed in Table 6. The biosorption process was spontaneous and feasible, which was supported by the negative value of ΔG° for all the studied temperatures. Generally, the values of ΔG° in the range of −20 to 0 kJ mol⁻¹ implied that the physical adsorption took place.⁵⁰ The negative values of ΔG° in this study suggested the existence of physical adsorption in the biosorption of FLU onto CA and CA-LENP beads, and thus both chemical adsorption and physical adsorption were involved in the biosorption process. The enthalpy changes were positive indicating that the biosorption process is endothermic in nature. The positive ΔS° for biosorption of FLU onto matrix beads implied the increase in randomness at solid–liquid interface during the biosorption process. Similar findings had been observed for the biosorption of FLU onto activated carbon.²⁵

Table 6 Thermodynamic parameters for the biosorption of FLU by CA and CA-LENP beads with dead A. bisporus

T (K)	CA beads			CA-LENP beads
	ΔG° (kJ mol^−1)	ΔH° (kJ mol^−1)	ΔS° (J K^−1 mol^−1)	ΔG° (kJ mol^−1)	ΔH° (kJ mol^−1)	ΔS° (J K^−1 mol^−1)
288.15	−14.03	1.53	53.79	−13.18	2.12	53.09
298.15	−14.49			−13.69
308.15	−14.88			−14.25
318.15	−15.69			−14.77

---

3.5. Biodegradation kinetic experiments

The removal and biodegradation efficiencies of FLU by the immobilized A. bisporus are illustrated in Fig. 7b. The removal efficiencies of CA and CA-LENP beads with A. bisporus were much higher than their biodegradation efficiencies, which could be explained by the theory that the FLU removal by immobilized A. bisporus is the result of both matrix bead biosorption and enzymatic biodegradation.

First-order reaction, second-order reaction and third-order reaction models were applied to study the kinetics of biodegradation and removal of FLU by immobilized A. bisporus. The corresponding kinetic parameters are presented in Table 7. Among three reaction models, first-order reaction model was the most suitable equation to describe the kinetics of FLU biodegradation by immobilized A. bisporus. The biodegradation rate constant of CA-LENP beads (0.18 × 10⁻² μg L⁻¹ min⁻¹) is slightly larger than CA beads (0.15 × 10⁻² μg L⁻¹ min⁻¹). Correspondingly, the biodegradation half-live of the FLU by CA-LENP beads was shorter than CA beads. The removal kinetics of FLU by immobilized A. bisporus followed the third-order reaction model, indicating that the FLU removal process might be influenced by the synergistic effect of beads biosorption, substrate diffusion and enzyme biodegradation.^15,51 Due to the large specific surface area, high enzyme activity and better mass transfer of CA-LENP bead with A. bisporus, the removal rate of CA-LENP bead is much faster than CA bead. Compared with bacteria isolated from petroleum sludge,³⁴ the immobilized A. bisporus in CA-LENP bead is more efficient in the removal of FLU from aqueous solution.

Table 7 The kinetic parameters of FLU removal and biodegradation by immobilized A. bisporus in CA and CA-LENP beads

		Kinetics parameter	FLU
CA-LENP	Biodegradation	k × 10^2 (μg L^−1 min^−1)	0.18
		t1/2 (min)	323.42
		R^2	0.97
	Removal	k × 10^5 (μg^2 L^−2 min^−1)	3.70
		t1/2 (min)	19.05
		R^2	0.99
CA	Biodegradation	k × 10^2 (μg L^−1 min^−1)	0.15
		t1/2 (min)	420.50
		R^2	0.98
	Removal	k × 10^5 (μg^2 L^−2 min^−1)	1.35
		t1/2 (min)	24.57
		R^2	0.99

---

The removal efficiencies of immobilized A. bisporus were similar to the biosorption efficiencies of beads with dead A. bisporus at the first incubation of 15 min, after which the reaction rate of FLU removal by immobilized A. bisporus became much faster (Fig. 7a and b). This result must be attributed to the catalysis biodegradation of FLU by enzymes inside beads with fruiting bodies of A. bisporus. In the case of FLU removal from solution by immobilized A. bisporus, FLU could be adsorbed onto the beads in terms of chemical and physical interaction, and then aggregated with each other forming small drops which could diffuse rapidly into the interior area of beads through pores on the surface of beads. When the FLU concentration decreased resulting from the catalysis biodegradation by enzymes inside matrix beads, the active sites for FLU biosorption at the outer areas of matrix beads would increase simultaneously, accelerating the phase transfer of FLU from aqueous solution to the bead surface.

3.6. Storage stability and reusability of the immobilized A. bisporus

Storage stability is a crucial factor to be considered for the industrial application of immobilized A. bisporus on a large scale. The activities of MnP and laccase in CA-LENP beads with A. bisporus could retain about 66.84 ± 1.91% and 67.42 ± 2.10% of the initial activities after storage at 25 °C for 60 days, respectively (Fig. 10a and b). However, the residual activities of MnP and laccase for immobilized A. bisporus in CA beads were only 50.77 ± 2.44% and 61.41 ± 2.87% of their initial activities after 60 days, respectively. This is likely due to the fact that CA-LENP bead with smaller size of pore could effectively reduce the leaching of fruiting bodies and enzyme from matrix bead.

Fig. 10 The storage stability of MnP (a) and laccase (b) secreted by immobilized fruiting bodies of A. bisporus in CA and CA-LENP beads at 25 °C, and the repeated removal of FLU by reusing immobilized A. bisporus in CA and CA-LENP beads (c).

To investigate the reusability of immobilized A. bisporus, the FLU contaminated water with an initial concentration of 90 μg L⁻¹ was treated by the same beads with A. bisporus in the 10 consecutive cycles. The experiments were conducted in the FBB under the optimum conditions. As illustrated in Fig. 10c, the removal efficiencies of FLU for CA-LENP and CA beads with A. bisporus increased to 96.31 ± 1.78% and 92.03 ± 1.28% in the first four reaction cycles, respectively. In the further reaction cycles, the FLU removal efficiencies decreased gradually to 88.86 ± 2.19% and 80.77 ± 2.14% in the 10th reaction cycle, respectively. The immobilized A. bisporus attained a better adaption of contaminated water in the first four reaction cycles, which facilitated the biosorption and biodegradation of FLU.^19,52 At completion of the eighth treatment, beads began to break and the leakage of fruiting bodies of A. bisporus occurred, which led to the reduction of removal efficiency. The similar phenomenon has also been described in other reports.^52,53 The high removal efficiencies of FLU for CA-LENP beads with A. bisporus presented in the ten consecutive cycles demonstrated the good reusability of CA-LENP beads with A. bisporus in the application of water treatment.

4. Conclusions

Immobilized A. bisporus in CA-LENP beads achieved similar activity as the entrapped laccase and could remove FLU from solution efficiently in FBB. The removal efficiency of FLU was affected profoundly by HRT and the dosage of immobilized fruiting bodies of A. bisporus. The removal of FLU from solution by immobilized A. bisporus is the result of both CA-LENP bead biosorption and enzymatic biodegradation. The existence of LENP is conducive to the large specific surface area of CA-LENP bead and the better retention of enzyme and fruiting body released from matrix bead as well as the high biosorption capacity of FLU. Hydrogen bonding interactions and π–π bonding interactions played key roles in the spontaneous biosorption process. The biodegradation and removal of FLU were well fitted to the first-order reaction and third-order reaction model, respectively. The immobilized A. bisporus in CA-LENP beads could retain almost 66.84% of initial activity after storage at 25 °C for 60 days and showed good reusability in the FBB during the ten consecutive cycles.

Acknowledgements

This study was financially supported by the Science and Technology Supportive Project of Sichuan Province, China (no. 2013SZ0062), Science and Technology Supportive Project of Chengdu (no. 12DXYB087JH-005) and NSFC (no. J1103518). The authors wish to thank Professor Guanglei Cheng and Dong Yu from Sichuan University for their technical assistance.

Notes and references

1. H. B. Othman, C. Leboulanger, E. Le Floc'h, H. Hadj Mabrouk and A. Sakka Hlaili, J. Hazard. Mater., 2012, 243, 204–211.
2. A. K. Haritash and C. P. Kaushik, J. Hazard. Mater., 2009, 169, 1–15.
3. J. Niu, Y. Dai, H. Guo, J. Xu and Z. Shen, J. Hazard. Mater., 2013, 248–249, 254–260.
4. Y. Zhong, T. Luan, L. Lin, H. Liu and N. F. Y. Tam, Bioresour. Technol., 2011, 102, 2965–2972.
5. A. P. Rodrigues, K. K. Lehtonen, L. Guilhermino and L. Guimarães, Sci. Total Environ., 2013, 443, 454–463.
6. E. Šepič, M. Bricelj and H. Leskovšek, Chemosphere, 2003, 52, 1125–1133.
7. J.-Y. Cho, J. Son, B.-J. Park and B.-Y. Chung, Environ. Monit. Assess., 2009, 149, 385–393.
8. A. Malik, P. Verma, A. Singh and K. Singh, Environ. Monit. Assess., 2011, 172, 529–545.
9. R. J. Law, V. J. Dawes, R. J. Woodhead and P. Matthiessen, Mar. Pollut. Bull., 1997, 34, 306–322.
10. K. Maskaoui, J. L. Zhou, H. S. Hong and Z. L. Zhang, Environ. Pollut., 2002, 118, 109–122.
11. S. N. Chu, S. Sands, M. R. Tomasik, P. S. Lee and V. F. McNeill, J. Am. Chem. Soc., 2010, 132, 15968–15975.
12. J. Niu, P. Sun and K.-W. Schramm, J. Photochem. Photobiol., A, 2007, 186, 93–98.
13. J. Ding, B. Chen and L. Zhu, Chin. Sci. Bull., 2013, 58, 613–621.
14. J. Niu, J. Xu, Y. Dai, J. Xu, H. Guo, K. Sun and R. Liu, J. Hazard. Mater., 2013, 246–247, 119–125.
15. Y. Dai, L. Yin and J. Niu, Environ. Sci. Technol., 2011, 45, 10611–10618.
16. P. Baborová, M. Möder, P. Baldrian, K. Cajthamlová and T. Cajthaml, Res. Microbiol., 2006, 157, 248–253.
17. M. R. Trejo-Hernandez, A. Lopez-Munguia and R. Quintero Ramirez, Process Biochem., 2001, 36, 635–639.
18. C. Lin, L. Gan, Z. Chen, M. Megharaj and R. Naidu, Biochem. Eng. J., 2014, 90, 1–7.
19. G. González, G. Herrera, M. T. García and M. Peña, Bioresour. Technol., 2001, 80, 137–142.
20. D. Daâssi, S. Rodríguez-Couto, M. Nasri and T. Mechichi, Int. Biodeterior. Biodegrad., 2014, 90, 71–78.
21. Y. Dai, J. Niu, L. Yin, J. Xu and Y. Xi, J. Hazard. Mater., 2011, 192, 1409–1417.
22. V. K. Gupta, S. Agarwal and T. A. Saleh, J. Hazard. Mater., 2011, 185, 17–23.
23. A. Mittal, J. Mittal, A. Malviya and V. K. Gupta, J. Colloid Interface Sci., 2009, 340, 16–26.
24. T. A. Saleh and V. K. Gupta, Environ. Sci. Pollut. Res., 2012, 19, 1224–1228.
25. J. Liu, J. Chen, L. Jiang and X. Wang, Environ. Sci. Pollut. Res., 2014, 21, 1809–1818.
26. S. Changchaivong and S. Khaodhiar, Appl. Clay Sci., 2009, 43, 317–321.
27. C. Zhang, L. Wu, D. Cai, C. Zhang, N. Wang, J. Zhang and Z. Wu, ACS Appl. Mater. Interface, 2013, 5, 4783–4790.
28. A. Mittal, J. Mittal, A. Malviya and V. K. Gupta, J. Colloid Interface Sci., 2010, 344, 497–507.
29. A. K. Jain, V. K. Gupta, A. Bhatnagar and D. Suhas, Sep. Sci. Technol., 2003, 38, 463–481.
30. L. Ma, Y. Peng, B. Wu, D. Lei and H. Xu, Chem. Eng. J., 2013, 225, 59–67.
31. M. S. Kuyukina, I. B. Ivshina, M. K. Serebrennikova, A. B. Krivorutchko, E. A. Podorozhko, R. V. Ivanov and V. I. Lozinsky, Int. Biodeterior. Biodegrad., 2009, 63, 427–432.
32. I. Karapinar and F. Kargi, Enzyme Microb. Technol., 1996, 19, 140–144.
33. Y.-S. Keum and Q. Li, Appl. Microbiol. Biotechnol., 2004, 64, 588–592.
34. S. Kumar, S. K. Upadhayay, B. Kumari, S. Tiwari, S. N. Singh and P. K. Singh, J. Hazard. Mater., 2011, 102, 3709–3715.
35. G. Dey, S. Bhupinder and R. Banerjee, Braz. Arch. Biol. Technol., 2003, 46, 167–176.
36. A. Riaz, S. A. U. Qader, A. Anwar and S. Iqbal, Aust. J. Basic Appl. Sci., 2009, 3, 2883–2887.
37. D.-Z. Chen, J.-Y. Fang, Q. Shao, J.-X. Ye, D.-J. Ouyang and J.-M. Chen, Bioresour. Technol., 2013, 139, 87–93.
38. L. Xu, M. Zhang and L. Zhu, Appl. Clay Sci., 2014, 100, 29–34.
39. W. Han, L. Luo and S. Zhang, Int. J. Environ. Sci. Technol., 2012, 9, 543–548.
40. S. Gülay and G. Şanlı-Mohamed, Int. J. Biol. Macromol., 2012, 50, 545–551.
41. N. Kawasaki, F. Ogata, I. Yamaguchi and A. Fujii, J. Oleo Sci., 2008, 57, 391–396.
42. H. C. Trivedi, V. M. Patel and R. D. Patel, Eur. Polym. J., 1973, 9, 525–531.
43. Y. S. Ho and G. McKay, Process Biochem., 1999, 34, 451–465.
44. M. Zhang, M. Ahmad, S. Lee, L. Xu and Y. Ok, Bull. Environ. Contam. Toxicol., 2014, 93, 84–88.
45. S. Ayoob, A. K. Gupta, P. B. Bhakat and V. T. Bhat, Chem. Eng. J., 2008, 140, 6–14.
46. S. Yang, M. Gao and Z. Luo, Chem. Eng. J., 2014, 256, 39–50.
47. I. Langmuir, J. Am. Chem. Soc., 1918, 40, 1361–1403.
48. H. Freundlich, Z. Phys. Chem., 1906, 40, 1361–1368.
49. J. Wu and H.-Q. Yu, Bioresour. Technol., 2007, 98, 253–259.
50. U. F. Alkaram, A. A. Mukhlis and A. H. Al-Dujaili, J. Hazard. Mater., 2009, 169, 324–332.
51. H. Xie, H. Liu, Y. Xie, M. Yang, S. Guo, Z. Zhou and H. Xu, Biochem. Eng. J., 2015, 97, 8–16.
52. H. Lin, Z. Chen, M. Megharaj and R. Naidu, Appl. Clay Sci., 2013, 83–84, 336–342.
53. L. Wu, G. Ge and J. Wan, J. Environ. Sci., 2009, 21, 237–242.

---

Footnote

† The first two authors contributed equally to this paper.

---