Biosorption of Lead(II) by Arthrobacter sp. 25: Process Optimization and Mechanism

Abstract

In the present work, Arthrobacter sp. 25, a lead-tolerant bacterium, was assayed to remove lead(II) from aqueous solution. The biosorption process was optimized by response surface methodology (RSM) based on the Box-Behnken design. The relationships between dependent and independent variables were quantitatively determined by second-order polynomial equation and 3D response surface plots. The biosorption mechanism was explored by characterization of the biosorbent before and after biosorption using atomic force microscopy (AFM), scanning electron microscopy, energy dispersive X-ray spectroscopy, X-ray diffraction, and Fourier transform infrared spectroscopy. The results showed that the maximum adsorption capacity of 9.6 mg/g was obtained at the initial lead ion concentration of 108.79 mg/l, pH value of 5.75, and biosorbent dosage of 9.9 g/l (fresh weight), which was close to the theoretically expected value of 9.88 mg/g. Arthrobacter sp. 25 is an ellipsoidal-shaped bacterium covered with extracellular polymeric substances. The biosorption mechanism involved physical adsorption and microprecipitation as well as ion exchange, and functional groups such as phosphoryl, hydroxyl, amino, amide, carbonyl, and phosphate groups played vital roles in adsorption. The results indicate that Arthrobacter sp. 25 may be potentially used as a biosorbent for low-concentration lead(II) removal from wastewater.

Introduction

Heavy metal contamination is a serious global environmental hazard, since all metals or elements with metallic characteristics can form toxic compounds. The presence of these compounds, especially in drinking water, even at trace levels, may cause anemia, itai-itai, or carcinoma and other illnesses in humans [7,9,17,24,27]. Lead (Pb) is a malleable and heavy post-transition metal generally applied in building construction, e-manufacturing, mining, and lead-acid batteries [13,26,30,39]. China is the largest producer and consumer of lead with an annual output near 135 million tons and annual consumption of 80 million tons. Outdated technology and equipment and a lack of effective environmental protection measures by Chinese industries have increased the lead contamination in the ecosystem, which may increase the probability of lead affecting humans through inhalation, ingestion, dermal contact, or other means. Acute and chronic exposure to lead can cause anemia, liver and kidney disease, central and peripheral nervous system damage, stunted growth, and high blood lead levels [33,36], especially in young children; these effects are frequently reported in China [19].

The efficient removal of lead ions from aqueous waste is essential for environmental safety, but current techniques such as ionic exchange, membrane separation, and activated carbon adsorption are inefficient for the treatment of wastewater containing low concentrations of lead [12]. In China, the threshold limit of lead-containing liquid from sewage and industrial effluents was set to 1.0 mg/l (Environmental Quality Standard for Surface Water of China, GB8978-1996), but wastewater without efficient treatment procedures cannot achieve the discharge standards. The biosorption process utilizes various natural materials that exhibit metal sequestering properties, which helps to decrease the concentration of heavy metal ions from the ppm to the ppb level [6,14,15,28].

Bacterial biomass shows tremendous potential for use in heavy metal removal because of its small size, ubiquity, and applicability to various environments [1,20,37]. Polysaccharides, proteins, and lipids on bacterial cell walls may provide functional groups such as amino, hydroxyl, carboxyl, and phosphate groups that are able to attract and bond with metals [25,28]. Many bacterial species have been used for the sorption of lead(II). A lead-resistant bacterium, Enterobacter cloacae strain P2B, was isolated from effluent discharged by a lead battery manufacturing company and was shown to resist lead nitrate up to 1.6 mM [29]. Pseudomonas aeruginosa ASU 6a is a promising biosorbent for the removal of lead(II) and nickel(II) ions from aqueous solutions [10]. In addition, Bacillus sp. PZ-1 exhibits lead resistance and has been used as an adsorbent to remove Pb(II) from aqueous solution even at low temperature [37]. Microbes, including Chlorella vulgaris, Bacillus sp., and yeast, are used commercially to adsorb or remove heavy metals by B.V.SORBEX Inc., Visa Tech Ltd., and the US Bureau of Mines, respectively [48].

Arthrobacter is a genus of bacteria that is commonly found in soil. Several species in this genus are efficient in bioremediation. Arthrobacter ps-5 isolated from the rhizosphere of the maidenhair tree (Ginkgo biloba) produces extracellular polymeric substances (EPS) that exhibit high adsorption capacity for copper and lead ions [49]. A. crystallopoietes has been shown to reduce hexavalent chromium levels in contaminated soil [3]. Additionally, free and immobilized biomass from pond-isolated Arthrobacter sp. was successfully utilized for the removal of copper both in a batch and a continuous system [18].

In this study, a lead-tolerant bacterium, Arthrobacter sp. 25, was found to show biosorption of low concentrations of lead ions in aqueous solution. This work aims to optimize the biosorption process using response surface methodology (RSM) and determine the removal mechanism of lead(II) by Arthrobacter sp. 25.

 

Materials and Methods

Preparation of Biosorbent, Lead(II) Solution, and Batch Biosorption Experiments

Arthrobacter sp. 25 used in the present study is an aerobic, elliptical shaped, and gram-negative bacterium with a GenBank accession number of JQ086764 for its 16S rDNA sequence, which was isolated from a lead-zinc mine in northeast China [22]. The free-living cells were collected at the late exponential phase by centrifugation at 8,000 rpm for 7 min and prepared as biosorbents after being fully rinsed by phosphate buffer solution and sterile deionized water. Lead(II) solutions were prepared by dilution of 1 g/l stock solution obtained by dissolving a predetermined quantity of Pb(NO3)2 (Sinopharm, China) in distilled and deionized water.

Batch biosorption experiments were carried out to investigate the effects of parameters on lead(II) adsorption in 150 ml flasks in a thermostatic shaker. After centrifugation and filtration, the residual lead(II) concentration in the supernatant was determined on an atomic absorption spectrophotometer (AA-6800; Shimadzu-GL, Japan). All experiments were conducted in triplicates and the mean values were calculated. The Q (biosorption rate, %) and the q (adsorption capacity, mg/g) were determined as Eqs. (1) and (2), respectively:

where C0 and C are the initial and final concentrations of lead(II) (mg/l). V and M are the volume of solution (ml) and the weight of adsorbent (g), respectively.

Optimization of Biosorption Process by Response Surface Methodology

Plackett-Burman (P-B) design. P-B design was done to screen out key variables with SPSS 16.0 software. Six main variables involved in biosorption process were selected, including initial lead(II) concentration (X1), pH (X2), contact time (X3), rotation rate (X4), temperature (X5), and biosorbent dosage (X6), where the ranges of independent variables were determined based on preliminary single factor experiments (not described in this article). Each variable was tested at a low level “-1” and a high level “1”, and three virtual columns (X7, X8, X9) were set to investigate the experimental error. On the basis of results obtained by the P-B test, the fitted first-order model is

where q is the predicted value of adsorption capacity, β0 and βi are the constant coefficients, Xi is the coded independent variable or factor, and k is the number of the independent variables.

Steepest ascent experiment. To move rapidly towards the neighborhood of the optimum response, a steepest ascent experiment was used to approximate the best response area. The new units were determined according to the estimated coefficient ratio from the first-order model. To move away from the first design center along the path of steepest ascent, the variables were set to about 16% to 25% higher or lower than the “0” level [40].

Box-Behnken test. With Design Expert 8.0.5 software (State-Ease Inc., USA), the Box-Behnken test was carried out to ascertain the optimum adsorption conditions as well as the interaction between independent variables. The coded levels (-1, 0, 1) were used to describe the characteristics of the response surface in the optimum region. The total number of treatment was 2k + 2k + n0, where k and n0 were the numbers of independent variables and experiment repetitions, respectively, at the center point. The results of Box-Behnken test were fitted with a second-order polynomial equation by multiple regression analysis:

where q is the predicted value of adsorption capacity, β0 is the offset term, βi is the ith linear coefficient; βii is the ith quadratic coefficient; and βij is the ijth interaction coefficient.

Mechanism Analyses

TEM. Transmission electron microscopy (TEM) was used for morphological observation of the bacterium. A droplet of the bacterial suspension was placed onto a Formvar film that had been transferred onto a copper index TEM grid (300 mesh) for 5 min, and the remaining solution was wiped off using a piece of filter paper. The specimens were then stained by placing a drop of 2% aqueous phosphotungstic acid for 3 min and examined with a transmission electron microscope (Hitachi H-7650, Japan) [42].

SEM-EDX and AFM. The surface morphology of Arthrobacter sp. 25 before and after lead(II) biosorption was characterized using scanning electron microscopy (SEM) (QuANTA200 model; FEI, USA). The elements were analyzed by an energy dispersive X-ray analysis system (EDX) (QuANTA200 model; FEI, USA). Prior to analysis, the samples were coated with a thin layer of gold under an argon atmosphere to improve electron conductivity and image quality.

To obtain complementary information, whole cells were also examined by Nanoscope IIIα atomic force microscopy (AFM) (Digital Instruments, USA). Both lead-loaded and lead-unloaded cells were collected by centrifugation, lightly rinse in a sterile phosphate-buffered saline solution and sterile deionized water, and then the specimens were transferred to glass slides and examined by AFM.

FTIR. The surface functional groups of Arthrobacter sp. 25 were analyzed by Fourier transform infrared spectroscopy (FTIR). Samples were ground with KBr powder and pressed into pellets for FTIR measurement in the frequency range of 4,000–400 cm-1 (ALPHA-T, Bruker, German) [38].

XRD. The crystalline structure of the biosorbent was investigated by X-ray diffraction (XRD) (D/max2200 model; Hitachi, Japan) equipped with a Cu Kα radiation. The operating conditions were U = 40 kV and I = 30 mA. The scans were performed between 5° and 80° with a scan rate of 4°/min and step size of 0.02° at room temperature.

 

Results and Discussion

Optimization of Biosorption Process

P-B design. The P-B design to screen key variables is shown in Table 1. The minimum and maximum values of the investigated parameters and the regression analysis for biosorption are listed in Table 2. The order of key variables that impacted the adsorption characteristics was X2 > X6 > X1 > X3 > X4 > X5. The pH value (X2) had the largest effect on the adsorption sites on the cell surface and the chemical state and distribution of metal ions. In this study, the pH value showed a positive effect on lead(II) uptake. This might result from competition between hydrogen ions at low pH and the weakly acidic nature of the active groups on the adsorbent, which favors metal adsorption [31]. Biosorbent dosage (X6) and initial lead(II) concentration (X1) also significantly affected the adsorption process (p < 0.05), but other variables were not significant (p > 0.05). The first-order model equation was derived as follows:

Table 1.Plackett-Burman design for screening of key variables.

Table 2.Results of Plackett-Burman and regression analysis for biosorption.

Steepest ascent experiment. The steepest ascent experiment is a procedure for moving sequentially along the path of steepest ascent. Using the steepest ascent experiment, we determined the following parameters as the response surface center: initial lead(II) concentration of 100 mg/l, pH of 6.0, biosorbent dosage of 10 g/l, and q of 8.9 mg/l.

Box-Behnken design of central composite. In order to test the key variables at different levels of low (−1), high (+1), and basal (0), a Box-Behnken design of central composite was employed. According to the experimental design, a total of 17 runs were required, and the experimental and predicted q values for each run are shown in Table 3. The relationship between independent variables and response values was expressed using the following second-order polynomial model (Eq. (6)):

Table 3.Results from Box-Behnken design for lead(II) uptake (mg/g).

The regression model was well-fitted and was able to explain 95.48% of the observed trends in the experiments. The coefficients R2 and Adj-R2 were 0.9548 and 0.8967, respectively, which indicated a high correlation between the experimental and predicted values. The values of “Prob>F” less than 0.05 (Table 4) indicated the model was significant and appropriate to describe the process of lead(II) adsorption [11].

Table 4.Box-Behnken analysis of adsorption capacity.

The model was appropriate to predict the lead(II) removal efficiency as a function of the variable factors, including initial lead(II) concentration, pH, and biosorbent dosage, based on statistical analysis. However, this model alone does not physically explain the biosorption phenomena.

Optimal conditions of lead(II) biosorption. The hierarchical linear model (Eq. (6)) obtained in this study was used to represent the response surface. The biosorption process was optimized using the response surface and contour plots. The adsorption capacities exhibited at different pH, biosorbent dosage, and initial lead(II) concentrations are presented in Figs. 1A, 1B, and 1C. The contour plots show the effects on adsorption capacity of varying two factors when the other factors were maintained at a basal level.

Fig. 1.3D surface plots of sorption capacity versus two independent factors: (A) pH and biosorbent dosage, (B) initial lead(II) concentration and biosorbent dosage, and (C) pH and initial lead(II) concentration.

The pH value can affect the protonation of functional groups on the biosobent as well as the metal chemistry [34]. The effects of pH on the lead(II) adsorption capacity of Arthrobacter sp. 25 are shown in Figs. 1A and 1C. As the pH value of the solution increased from 5.0 to 5.7, the adsorption capacity increased from 9.3 to 9.9. When the pH value was higher than 5.7, the insoluble metal hydroxides were precipitated and the adsorption capacity decreased. Similar results were observed for lead(II) removal using dried cells of Lactobacillius bulgaricus and shells of Pistacia vera L. [40,51]. The cell surface of bacteria contains complex polysaccharides, proteins, and lipids that can provide amino, carboxyl, and sulfate groups [2]. At low pH, cell surface ligands can protonate, which impedes interaction with metal cations owing to the repulsive force. As higher pH values, more ligands such as amino, phosphate, and carboxyl groups become exposed, and these metal-binding functional groups are able to attract more metal ions. However, as pH further increased, lead(II) interacted with the oxygen or hydroxyl ions resulting in oxide or hydroxide precipitation, subsequently hindering the adsorption process [43].

Biosorbent dosage showed a significant influence on the biosorption process at a specified initial concentration. The effect of biomass dosage on lead(II) removal is depicted in Figs. 1A and 1B. The biosorption capacity increased as biomass dosage increased from 7.5 to 10 g/l, but at higher dosages, the adsorption capacity decreased greatly. This trend could be due to the partial aggregation of biomass at higher biomass concentrations, decreasing the effective surface area available for biosorption [23]. When the concentration of adsorbents and initial lead(II) were equivalent, the adsorption capacity reached a maximum.

Figs. 1B and 1C show the effect of initial lead(II) concentration on the biosorption capacity of Arthrobacter sp. 25. The adsorption capacity increased gradually when the lead(II) concentration increased from 80 to 109 mg/l, and then decreased as the concentration was additionally increased. At lower metal ion concentrations, the ratio of the initial moles of metal ion to the available surface area was low; subsequently, the fractional sorption was independent of the initial concentration. However, at higher concentrations, the sites available for sorption were fewer relative to the moles of metal ions; thus, the removal of metal ion was strongly dependent upon the initial concentration [45].

These results showed that the optimal conditions for lead(II) biosorption by Arthrobacter sp. 25 (free-living cells) from aqueous solution were an initial lead(II) concentration of 108.79 mg/l, pH of 5.75, and a biosorbent dose of 9.9 g/l. The predicted maximum theoretical adsorption capacity of lead(II) was 9.88 mg/g, as indicated in the response surface and contour plots (Fig. 1).

Model validation test. Under optimum conditions, the observed adsorption capacity was 9.6 mg/g, close to the theoretical maximum value of 9.88 mg/g, for a 97.2% prediction accuracy. This level of accuracy proved the feasibility of RSM in the optimization of the adsorption conditions.

Other bacteria have been used for lead biosorption and exhibited adsorption capacities in the range of 1.29-110 mg/g under different operational conditions (Table 5). This large range of efficacy can be attributed to differences in surface structure, functional groups, available surface area, and the conditions of the cells (living or lyophilized/heat-dried cells) [4,21,32,41]. In this study, free-living cells were used as biosorbent and exhibited an adsorption capacity of 9.6 mg/g (using a wet/dry ratio of 8.42:1). The capacity could theoretically reach 83.2 mg/g with dried cells, and a direct comparison between these different results is not valid.

Table 5.Uptake capacity of lead by different bioadsorbents.

Mechanism Analyses

TEM. The cells of Arthrobacter sp. 25 were ellipsoidal in shape with an average length of 1.25 ± 0.2 μm and width of 0.83 ± 0.2 μm, as shown in Fig. 2. A thick layer of EPS extended up to 0.5 μm from the cell membrane and uniformly covered the cell. Both cell walls and the EPS layer provided abundant sites for the binding of metal ions [42].

Fig. 2.TEM image of Arthrobacter sp. 25.

SEM-EDX and AFM analyses. SEM images of Arthrobacter sp. 25 with and without lead(II) loading are shown in Fig. 3 (A, B). Lead(II)-loaded cells differed in morphology from the unloaded ones. For the samples containing the unloaded cells, the majority of cells remained intact, smooth, and closely connected with one another, providing a large surface area for biosorption (Fig. 3A). In the case of the loaded samples, the matrix layers of the cell wall appeared to shrink and stick. Naik et al. [29] observed that cells of Enterobacter cloacae strain P2B shrunk significantly after exposure to lead nitrate, which may be consistent with our observations. The changes in cellular morphology and size may result from mechanical force and reciprocation between surface-active components and metallic ions [20]. Therefore, this structural change was attributed to the strong crosslinking of metal (lead) and negatively charged chemical groups on the cell wall polymers.

Fig. 3.SEM images of Arthrobacter sp. 25 before (A) and after (B) lead(II) uptake.

The EDX analysis showed that the elemental composition of strain 25 was significantly changed after lead(II) adsorption (Figs. 4A, 4B; Table 6). The atomic percentage of element Pb (At %) was 6.31, C (At %) decreased from 22.47 to 8.23, and O and P increased respectively from 34.04 to 36.61 and 6.19 to 12.34 after lead(II) uptake. These results indicated that lead(II) could covalently bond with C-, O-, and P-containing functional groups. The functional groups (carboxylate, hydroxyl, amino, and phosphate) on the bacterial cell surface were previously found to be essential for lead(II) adsorption [25,28,44,50]. The adsorption of lead(II) by strain 25 occurred primarily on the cell surface. The atomic percentage of Na and K decreased from 2.94 to 1.82 and 1.20 to 0.60, respectively, after the cells were mixed with lead(II), indicating the possible exchange of Pb with Na and K on the cell surface [8,47].

Fig. 4.Energy dispersive X-ray analyses of Arthrobacter sp. 25 before (A) and after (B) lead(II) uptake.

Table 6.Atomic ratio analyses of Arthrobacter sp. 25 before (A) and after (B) lead(II) uptake.

To evaluate the changes of surface roughness, depth, and width in response to lead(II) adsorption, the two-dimensional and its corresponding three-dimensional topographic AFM images of the strain 25 cell were compared and analyzed as shown in Fig. 5. Before adsorption, the cells were elliptical in shape with an average size of 2-3 μm length and 1 μm width (Fig. 5A). This was longer than the length measured by SEM, suggesting the layer of extracellular polymeric substances may contribute to the increased thickness of the cell surface. The cell surface was relatively smooth with a RMS (root mean square) of 91.7 nm and Ra (roughness average) of 20.4 nm (data not shown). After lead(II) adsorption, the cell appeared to be coated with a soft, compressible material. The RMS and Ra values, respectively, increased to 160.0 nm and 21.9 nm. Obviously protuberant components were unevenly distributed on the cell surface, and stretched out upon retraction of the tip to hundreds of nanometers in length (Fig. 5B), possibly resulting from the bonding of lead(II) with the EPS polysaccharide, protein, and amide [52]. The results indicated that lead(II) adsorption by strain 25 included processes of surface adsorption and micro-precipitation. The irregular shape and topography of the cell resulted from the interactions between lead(II) and the surface of the biosorbent. These observations are consistent with the TEM and SEM analyses. Moreover, the morphological changes after lead(II) adsorption were partly attributed to different interaction forces between the cell and substrate surfaces [5].

Fig. 5.AFM images of Arthrobacter sp. 25 before (A) and after (B) contacting with lead(II). Each group of images includes two-dimensional topographic images of the cell surface (left); sectional analysis determining the depth of cell (middle); and three-dimensional topographic images of the cell surface (right). The lead(II) concentration is 100 mg/l.

XRD analyses. XRD analyses before and after lead(II) adsortpion are shown in Fig. 6. There was a wide diffraction peak at the incident-beam 2θ of 14°-30°, which was formed by polysaccharide and other organic ingredients. The shape of this peak indicated that the original absorbents had an amorphous structure. The XRD pattern of the lead(II)-loaded biomass was distinct and complex with some peaks narrowing and two new peaks emerging at 26.52° and 30.26°, indicating the deposition of lead hydroxides and lead carbonates in the form of crystallized lead [53]. According to Jade 5.0 software analysis, the new peaks were generated by the integration of lead(II) with P=O and –OH. Together with the EDX analysis, it was deduced that phosphoryl and hydroxyl were the main cellular surface functional groups for lead(II) biosorption.

Fig. 6.X-ray powder diffraction analyses of Arthrobacter sp. 25 before (A) and after (B) lead(II) uptake.

FTIR analyses. FITR was used to identify the biomass functional groups involved in the adsorbing process, which is important for elucidation of the surface-bonding mechanism. The FTIR spectrum with and without the lead(II) revealed significant differences in the absorption peaks of functional groups (Figs. 7A, 7B). The FTIR spectrum showed that the peaks of hydroxyl and amine groups shifted from 3,296.61 to 3,307.13 cm-1, indicating the formation of more -OH and -NH by complexation; an anti-symmetric stretching peak of -CH shifting from 2,927.06 to 2,929.03 cm-1 by interaction with lead(II); a symmetric stretching peak of C=O shifting from 1,652.48 to 1,652.59 cm-1; an enhancement of an amide peak between 1,535 to 1,560 cm-1; a stretching vibration peak of C-O-C shifting from 1,234.47 to 1,231.30 cm-1; a stretching peak of C-OH located at around 1,000 to 1,100 cm-1 now migrated from 1,066.11 to 1,055.33 cm-1; and the reduced transmittance of the phosphate peak at 953.35 cm-1 indicated that the presence of metals caused less stretching. NH- stretching was found to be responsible for copper(II) and lead(II) bonding, and the C-O bond played an important role in lead(II) sorption [34]. Additionally, a report by Gupta and Rastogi [16] found that the presence of amino, carboxyl, hydroxyl, and carbonyl groups on the surface of the green alga Spirogyra was responsible for the bonding of lead [45]. These phenomena further supported the conclusion that available hydroxyl, amino, amide, carbonyl, and phosphate functional groups on the Arthrobacter sp. 25 cell surface were associated with lead(II) biosorption.

Fig. 7.FTIR spectrometry of Arthrobacter sp. 25 before (A) and after (B) lead(II) adsorption.

In conclusion, the optimum conditions for lead(II) adsorption by Arthrobacter sp. 25 were an initial lead ion concentration of 108.79 mg/l, pH 5.75, and biosorbent dosage of 9.9 g/l. Under these conditions, the maximum adsorption capacity was 9.6 mg/g.

The experimental results showed that the lead-resistant strain Arthrobacter sp. 25 can be effectively used for the removal of lead ions from wastewater at low concentrations. The complexity of the lead(II) removal mechanism was also confirmed.

The cells of Arthrobacter sp. 25 are ellipsoidal shaped and covered with EPS. The mechanism of biosorption of lead(II) involves surface phenomena such as physical adsorption and microprecipitation as well as ion exchange. Phosphoryl, hydroxyl, amino, amide, carbonyl, and phosphate functional groups were all engaged in lead(II) adsorption.

This lead-resistant bacterial strain may be utilized as a potential biotechnological agent for the bioremediation of lead-contaminated water. This work may provide some new insights into heavy metal removal in real-life bioprocesses and contribute to the application of functional Arthrobacter species.