Effects of non-dissolved redox mediators on a hexavalent chromium bioreduction process

ABSTRACT

The effects of six non-dissolved redox mediators (RM) immobilized in cellulose acetate beads on enhancing Cr(VI) reduction by Mangrovibacter plantisponsor CR1 were investigated. In addition, the voltammetric behaviours and electron transfer capacities of the redox mediators were examined using electrochemical methods. Compared to the control beads, the Cr(VI) bioreduction rate with 1-chloroanthraquinone cellulose acetate beads (1-CAQ/CA beads) was increased up to 4.5-fold, which was mainly attributed to enhanced electron transfer by 1-CAQ. The redox mediators also improved the oxidation–reduction potential values of the Cr(VI) bioreduction processes, which might assist in Cr(VI) bioreduction. The role of the redox mediators was discussed based on the cyclic voltammetric characteristics (E₀') of the redox mediators and the electrochemical impedance spectroscopy characteristics (R_ct) of the RM/CA beads. A linear correlation was found for the reaction constant k and the 1-CAQ concentration (C_1-CAQ), which was k = 1.5674 C_1-CAQ + 4.8506 (R² = 0.9683). The Cr(VI) bioreduction was affected by temperature, and the optimum pH for the Cr(VI) bioreduction was 6.5. The results of repeated-batch operations showed that 1-CAQ/CA beads exhibited good stability and persistence. This study contributes to a better understanding of the effects of the redox mediator on Cr(VI) bioreduction process and demonstrates its promising potential for environmental bioremediation applications.

KEYWORDS:

• Hexavalent chromium
• bioreduction
• non-dissolved redox mediators
• immobilization

Introduction

Chromium (Cr), one of the most widespread elements on earth, is an important industrial material that is used extensively in electroplating, dye and pigment manufacturing, wood preservation, leather tanning, alloy production and other applications.[1,2] Large quantities of industry effluents containing chromium are often released into natural waters without proper treatment, resulting in severe environmental pollution.[3] Chromium exists in different valence states ranging from −2 to +6, and only hexavalent chromium [Cr(VI)] and trivalent chromium [Cr(III)] are stable oxidation states in the natural environment. Cr(VI) and Cr(III) differ widely in physicochemical behaviour, mobility and toxicity.[4] Cr(VI) is approximately 100 times more toxic and 1000 times more mutagenic than Cr(III) due to its solubility across nearly the entire pH range and its greater mobility.[5] The European Union recommends total chromium limits of 0.05 and 0.1 mg/L for potable and industrial wastewater, respectively,[6] and the United States Environmental Protection Agency has set a maximum contaminant level of 0.05 mg/L for Cr(VI) in domestic water.[1] Thus, the transformation of Cr(VI) to Cr(III) is an attractive and useful process for the remediation of Cr(VI) pollution.

Conventional physicochemical treatment processes for Cr(VI) detoxification generally involve the reduction of Cr(VI) to Cr(III) by a strong reducing agent and precipitation of Cr(III) by adjusting the pH to a neutral range. However, these processes require large amounts of chemicals and energy and they produce secondary wastes. Biological reduction of Cr(VI) to Cr(III) is an ideal alternative because it is economical, safe and sustainable. A Cr(VI)-reducing ability has been demonstrated by many aerobic and anaerobic bacteria, such as Bacillus megaterium,[7] Desulfovibrio vulgaris,[8] Thermus scotoductus,[9] Ochrobactrum spp.,[10] Shewanella oneidensis,[11] Escherichia coli [12] and Cellulomonas spp.[2] Nevertheless, the biological reduction of Cr(VI) occurs slowly due to toxic effects on bacteria or electron transfer limitations, and bioreduction processes invariably require several days or even weeks.

It has been shown that redox mediators, such as humus substances and their quinoid analogues, could enhance the biotransformation of azo dyes,[13] nitroaromatics,[14,15] aliphatic and aromatic poly-halogenated compounds,[16–18] nitrates,[19] metals [3,20] and metalloids.[21] Redox mediators could decrease electron transfer limitations, accelerating the biotransformation of these contaminations. Liu et al. [3] evaluated the catalysis effect of certain quinones on Cr(VI) bioreduction, and the result showed that the average reduction rate of Cr(VI) was improved by nearly 3.3-fold in the presence of 0.2 mmol/L 2-hydroxy-1,4-naphthoquinone (lawsone). However, to maintain the increased conversion rates, the dissolved redox mediators had to be provided continuously, which was uneconomical and caused recontamination problems. Therefore, applications of non-dissolved redox mediators and effective methods to immobilize redox mediators have lately been explored. Redox mediators have been immobilized by entrapment in polyvinyl alcohol-H₃BO₃ and agar,[13] by covalent binding on ceramsite [22] and polyurethane foam,[15,23] by adsorption on metal oxide nanoparticles,[24] by electrostatic attraction on the surface of anion-exchange resins [25] and by electropolymerization on activated carbon felt.[26,27] Previously, Li et al. [28] reported that non-dissolved redox mediators immobilized in cellulose acetate could enhance the nitrite denitrification efficiency, but little is known regarding how to affect the Cr(VI) bioreduction of non-dissolved redox mediators.

In this work, the characteristics of Cr(VI) bioreduction by Mangrovibacter plantisponsor CR1 with non-dissolved redox mediators immobilized in cellulose acetate beads were investigated. The electron transfer properties of the redox mediators were evaluated based on cyclic voltammetric characteristics and electrochemical impedance spectroscopy (EIS) characteristics. This work will support a better understanding of the effects of the redox mediators on Cr(VI) bioreduction process and improve practical bioremediation processes.

Materials and methods

Chemicals

Anthraquinone (AQ), 1-chloroanthraquinone (1-CAQ), 2-chloroanthraquinone (2-CAQ), 1,5-dichloroanthraquinone (1,5-DCAQ), 1,8-dichloroanthraquinone (1,8-DCAQ) and 1,4,5,8-tetrachloroanthraquinone (1,4,5,8-TCAQ) were of analytical grade and supplied by Sigma–Aldrich. The chemical structures of these redox mediators are shown in the Supplemental data (Figure S1). Cellulose acetate (CA) was purchased from Chemical Reagent Corporation (Shanghai, China); triolein (98% purity) was purchased from Jinlong Chemical Reagent Corporation (Beijing, China); and Tween 80 (A.R.), acetone (A.R.), Mg(ClO₄)₂ (A.R.) and liquid olefin were purchased from Chemical Reagent Corporation (Beijing, China). Other chemical reagents were of analytical grade and purchased from Xiandai Ltd. (Shijiazhuang, China).

Bacterial strains and culture conditions

The Cr(VI)-reducing bacterial strain CR1 was isolated from tanning sludge collected from a tannery in Xinji City (Hebei, China). It was identified as M. plantisponsor (GenBank accession no. KF530843) based on its biochemical properties and 16S rRNA gene sequence homology analysis.[29] Cr(VI) bioreduction was performed in mineral medium containing the following: 1.0 g/L NH₄Cl, 1.0 g/L NaHCO₃, 0.2 g/L K₂HPO₄, 0.2 g/L MgSO₄·7H₂O, 1.0 g/L NaCl and 2.0 g/L glucose with 2 mL of a trace element solution. The trace element solution included the following: 15 g/L ethylene diamine tetraacetic acid (EDTA), 0.43 g/L ZnSO₄·7H₂O, 0.24 g/L CoCl₂·6H₂O, 0.99 g/L MnCl₂·4H₂O, 0.25 g/L CuSO₄·5H₂O, 0.22 g/L NaMoO₄·2H₂O, 0.19 g/L NiCl₂·6H₂O, 0.21 g/L NaSeO₄·10H₂O, 0.014 g/L H₃BO₄, and 0.05 g/L NaWO₄·2H₂O. Strain CR1 was cultivated aerobically in mineral medium at 30 °C until the optical density at 600 nm (OD₆₀₀) reached 0.75. A stock solution of Cr(VI) was prepared by dissolving K₂CrO₄ in deionized water. A certain volume of Cr(VI) from the stock solution was taken, sterilized and added to the medium to obtain the desired Cr(VI) concentration. The biomass concentration of strain CR1 was 0.31 g of dry weight per litre in all of the experiments. The pH was adjusted with 0.1 mol/L NaOH or 0.1 mol/L HCl.

Immobilization of redox mediators

To prepare immobilized redox mediators (RM), 7.0 g of cellulose acetate was dissolved in 55 mL of acetone, and ultrasonication (Kunshan Co., KQ218, China) for 20 min with a fixed power of 200 W was used to promote dissolution. Then, 5 mL of deionized water was added to the mixture, followed by 1.5 g of Mg(ClO₄)₂ and 0.5 mL of triolein. Subsequently, different redox mediators were added in different amounts. The mixtures were kept at 35 °C for 48 h and stirred to achieve uniform admixture. Thus, viscous syrups were carefully prepared and homogeneous phases were obtained. Each viscous syrup was added drop by drop to a glass tube containing the liquid olefin (approximately 2/3 of the volume). The cellulose acetate solution and liquid olefin were maintained at room temperature. During the process of dripping the viscous syrup through the air to the column of liquid olefin, a spherical immobilization bead was formed by the action of surface tension and was collected at the bottom of the glass tube. Finally, the cellulose acetate beads with redox mediators (RM/CA beads) were washed with deionized water to remove all soluble impurities. The redox mediator concentration was calculated by the mass ratio of the added redox mediator and the number of beads. In addition, cellulose acetate beads without redox mediators (i.e. control beads) were prepared by the same procedures.

Cr(VI) bioreduction with redox mediators

Bioreduction assays were performed in 250 mL serum glass bottles with the mineral medium described above. The atmosphere in the headspace of the bottles was changed with a mixture of N₂/CO₂ (80%/20%) in order to create anaerobic conditions. Cr(VI) bioreduction was performed with different non-dissolved redox mediators to obtain the optimal redox mediator. Cr(VI) bioreduction with the optimum redox mediator was conducted under different redox mediator concentrations (0–0.072 mol/L), temperatures (20 °C–50 °C) and initial pH values (4.0–9.0) to optimize the parameters simultaneously. Six repeated batch operations were also studied to investigate the stability and persistence of enhancement of the Cr(VI) bioreduction. At pre-determined time intervals, 2 mL aliquots were withdrawn for reduction experiments at set time intervals. All of the experiments were performed in triplicate.

Analytical methods

Bacterial growth was monitored by measuring OD₆₀₀ using an ultraviolet/visible (UV/vis) spectrophotometer (Tianmei Co., UV2600, China). Cr(VI) concentration was analysed according to the standard colorimetric method of the American Public Health Association involving a reaction with 1,5-diphenylcarbazide and measured via the same UV/vis spectrophotometer at 540 nm.[30] The pH was measured using a digital pH meter (Mettler Toldedo Co., Delta-320, China). Electrochemical measurements were performed using the reported three-electrode configuration [13] with a Voltalab Powersuite (Lanlike Co., LK98B, China).

EIS was used to explore the roles of redox mediators in Cr(VI) bioreduction. A platinum wire was used to string four beads along their radical direction. Due to good elasticity and toughness, the beads tightly contacted the platinum wire to produce a sufficient contact area for good conductivity. The bead-platinum composite electrode was used as the working electrode for the EIS measurement. EIS measurements were performed using an electrochemical workstation (CHI 660E, Chenhua Co., China) with a saturated calomel electrode as a reference electrode and a Pt counter electrode in a KCl electrolyte with 5 mmol/L Fe(CN)₆^3-/4−. EIS was measured at the open circuit voltage over a frequency range of 100 kHz–0.01 Hz with a potential amplitude of 5 mV. The electric resistance of the beads was then evaluated according to the Nyquist plots of the impedance spectra.

Results and discussion

Effects of different non-dissolved redox mediators on Cr(VI) bioreduction

Without bacteria, the same volume of control beads and RM/CA beads were examined for their adsorption of Cr(VI). Control beads and RM/CA beads caused less than 3% adsorption of Cr(VI) within 48 h (data not shown) under a fixed initial Cr(VI) concentration (52 mg/L), which suggested that physicochemical processes, i.e. absorption or chemical reduction, were not responsible for the Cr(VI) bioreduction. Cr(VI) with the control beads was reduced by only 16.2% by strain CR1 within 10 h, and Cr(VI) was reduced by 90.4%, 76.1%, 69.8%, 54.7% and 49.6% with 0.036 mol/L 1-CAQ, 2-CAQ, 1,5-DCAQ, AQ and 1,8-DCAQ within 10 h, respectively (Figure 1). The highest Cr(VI) reduction rate was obtained in the 1-CAQ/CA beads system, which showed an increase of nearly 4.5-fold in comparison with the control bead system. The accelerating effect of 1,4,5,8-TCAQ was not remarkable, and only 20.3% Cr(VI) was reduced. The acceleration order during the bioreduction process was 1-CAQ > 2-CAQ > 1,5-DCAQ > AQ > 1,8-DCAQ > 1,4,5,8-TCAQ > control. As the number of –Cl substituents increased, the acceleration effect decreased. Meanwhile, the position of the –Cl substituent influenced the acceleration effect, except for AQ. Previous studies have reported that the acceleration orders of redox mediators during the process of nitrite denitrification [28] and azo dye decolourization [19] were 1,4,5,8-TCAQ > 1,5-DCAQ > 1,8-DCAQ > AQ and 1,5-DCAQ > 1,8-DCAQ > AQ > 1,4,5,8-TCAQ, respectively. Compared to the reported results, the difference in acceleration order was due to the different biological reduction processes. The accelerating effect of the six quinone compounds was due to the presence and the position of the electrophilic functional groups in their structures, and the steric effects of the different targeted pollutants may also affect the acceleration efficiency.

Figure 1. Effects and changes in the oxidation–reduction potential (ORP) of different non-dissolved redox mediators on Cr(VI) reduction by strain CR1 under anaerobic conditions.

Different redox couples co-exist in different systems, and all of the redox couples contribute to the oxidation–reduction potential values (ORP) of the system. The ORP features of the Cr(VI) bioreduction processes may indirectly reflect specific chemical, biochemical and biological changes over time. Therefore, the ORP changes in the Cr(VI) bioreduction processes with different non-dissolved redox mediators were monitored. Throughout the bioreduction processes, all of the ORP values were negative. At the beginning of the bioreduction processes, a rapid drop in ORP occurred in both the control bead and the RM/CA bead systems (Figure 1). In the control, 1-CAQ/CA, 2-CAQ/CA, 1,5-DCAQ/CA, AQ/CA, 1,8-DCAQ/CA and 1,4,5,8-TCAQ/CA systems, the ORP values stabilized at approximately −550, −400, −470, −460, −540, −470, and −495 mV, respectively (Figure 1). The results showed that the redox mediators could improve the ORP values of the Cr(VI) bioreduction processes, and the optimal ORP values were approximately 150 mV higher with 1-CAQ than the controls. Previous research also demonstrated that low ORP values were required for Cr(VI) bioreduction.[31] The results of this experiment indicated that higher ORP values (−400 mV with 1-CAQ/CA beads vs. −550 mV with control beads) were associated with faster reduction rates. The results vary due to the differences in the bioreduction systems, which are related to different bacterial species and their characteristics.

Effects of different 1-CAQ concentrations on Cr(VI) bioreduction

Mediated Cr(VI) bioreduction was studied with different initial 1-CAQ concentrations (0–0.072 mol/L). The bioreduction process with 1-CAQ followed a modified pseudo-zero order model, which was described by k = dC_Cr(VI)/dt (Figure 2A). The reaction constant k (mg Cr(VI) g⁻¹ dry cell weight h⁻¹) was 3.87, 6.60, 9.45, 11.45, 12.54, 14.11 and 15.37 at 1-CAQ concentrations of 0, 0.012, 0.024, 0.036, 0.048, 0.060 and 0.072 mol/L, respectively. The linear correlation for the k value and the 1-CAQ concentration (0–0.072 mol/L) was k = 1.5674 C_1-CAQ + 4.8506 (R² = 0.9683) (Figure 2B). The average reduction rates k increased with increasing concentrations of the redox mediator. Furthermore, the mediator gradients showed that the effect is especially strong at low mediator concentrations. The acceleration gradient levels were weak at higher concentration, possibly due to the lack of a sufficient electron source.[32] Similar linear relationships were obtained in previous studies on Cr(VI) bioreduction with dissolved redox mediators.[12]

Figure 2. Effects of different 1-CAQ concentrations on Cr(VI) reduction by strain CR1 under anaerobic conditions. Cr(VI) concentration (A); Cr(VI) reduction rate (B).

Effects of temperature and pH on Cr(VI) bioreduction

As shown in Figure 3(A), the reduction efficiency increased with rising the temperature from 20 °C to 50 °C, which was quite similar to the result observed in the case of E. coli BL21.[12] The reduction rate constant k' (mg Cr(VI) g⁻¹ dry cell weight h⁻¹) was 7.44, 10.24, 12.01, 15.04 and 17.27 at 20 °C, 30 °C, 35 °C, 40 °C and 50 °C, respectively. The pH notably affected the chromium form in solution as well as the chromium binding sites on the cell surface. Therefore, experiments were performed to monitor the optimum pH. As shown in Figure 3(B), the optimum pH for Cr(VI) reduction by CR1 was 6.5. The Cr(VI) bioreduction activity was nearly inhibited at pH 4.0, and it decreased gradually with increasing pH from 6.5 to 9.0. Previous research demonstrated that the standard redox potential for the Cr(VI)/Cr(III) couple decreased with increasing pH, which would affect the Cr(VI) bioreduction.[33]

Figure 3. Effects of temperature (A) and pH (B) on Cr(VI) reduction by CR1 under anaerobic conditions.

Repeat batch tests

The anaerobic bioreduction of Cr(VI) with 1-CAQ during repeated experiments was investigated. The results presented in Figure 4 showed that the bioreduction efficiency decreased by less than 5%, and the redox mediator immobilized in cellulose acetate beads exhibited good stability and persistence under the applied experimental conditions after six cycles. Compared to the calcium alginate beads with immobilized redox mediators, the RM/CA beads exhibited good reusability (data not shown), which might significantly reduce the operation costs in industrial applications.

Figure 4. Bioreduction of Cr(VI) by strain CR1 with repeated batch cycles under anaerobic conditions.

Electrochemical performances of non-dissolved redox mediators

EIS analysis was conducted to evaluate the charge-transfer resistance of the redox mediators used in this study. A typical Nyquist plot for the three electrodes is shown in Figure 5. The EIS Nyquist plots consist of a semicircular portion (high frequency region) and a subsequent straight line (low frequency region). The diameter of the former corresponds to the charge-transfer resistance (R_ct) at an electrode/electrolyte interface, which usually represents the resistance of electrochemical reactions on the electrode, and a smaller R_ct value indicates a faster rate of the electron transfer.[34] The order of the R_ct value was control > 1,8-DCAQ > AQ > 1,5-DCAQ > 1-CAQ > 1,4,5,8-TCAQ > 2-CAQ. The results showed that the R_ct of the control beads was higher than the R_ct of the RM/CA beads, and redox mediators could enhance the electron transfer process.

Figure 5. EIS of control beads and RM/CA beads.

The effect of redox mediators on the biological reduction process may be directly related to the standard redox potential (E₀'), which should ideally be between the values of the two eventual half reactions, the reduction of Cr(VI) and the oxidation of a primary electron donor.[35] Therefore, the cyclic voltammetry characteristics of the six redox mediators were determined (Table 1) at a 50 mV/s scan rate and a redox mediator concentration of 3.6 mmol/L. The six redox mediators displayed a reversible oxidation peak in the potential range of −0.120 to −0.226 V vs. Hg–Hg₂Cl₂(s) and a reversible reduction peak in the range of −0.224 to −0.271 V vs. Hg–Hg₂Cl₂(s). The E₀' values for the six redox mediators are in the range of −0.175 to −0.249 V vs. Hg–Hg₂Cl₂(s), which are lower than the values for Cr(VI)/Cr(III).[21] The redox mediator 1,4,5,8-TCAQ had the lowest reduction peaks (E_a) due to its four chloro substituents. The E₀' order of the redox mediators was 2-CAQ ≥ 1,5-DCAQ ≥ 1-CAQ > 1,8-DCAQ > AQ > 1,4,5,8-TCAQ. Li et al. [28] reported that the acceleration effect during the nitrite denitrification process was related to E₀' of the redox mediators. However, the acceleration effect of the redox mediators obtained in this study (see ‘Effects of different non-dissolved redox mediators on Cr(VI) bioreduction’) did not agree with E₀'. This result indicated that the acceleration effect of the non-dissolved redox mediators was determined by many factors, including the EIS characteristics (i.e. R_ct) and cyclic voltammetry characteristics (i.e. E₀') of the redox mediators. Further work is needed to elucidate the relationship between the electrochemical performances of the redox mediators and the acceleration effect of the redox mediators during the Cr(VI) bioreduction process.

Table 1. Cyclic voltammetric characteristic values of six redox mediators.

Redox mediators	Reduction peaks (Ea)/V	Oxidation peaks (Er)/V	Standard potentials (E0')/V
AQ	−0.253	−0.182	−0.218
1-CAQ	−0.240	−0.120	−0.180
2−CAQ	−0.230	-0.120	−0.175
1,8-DCAQ	−0.248	−0.157	−0.203
1,5-DCAQ	−0.224	−0.132	−0.178
1,4,5,8-TCAQ	−0.271	−0.226	−0.249

Conclusions

The characteristics of Cr(VI) bioreduction by M. plantisponsor CR1 with non-dissolved redox mediators immobilized in cellulose acetate beads were investigated. The non-dissolved redox mediators were found to enhance the reduction of Cr(VI) to Cr(III) and to improve the ORP values of the Cr(VI) bioreduction processes, which might contribute to accelerating the Cr(VI) bioreduction. There was a linear correlation between the reaction constant k value and C_1-CAQ. The Cr(VI) bioreduction was affected by temperature and initial pH, and the RM/CA beads exhibited better stability and persistence performance. The electrochemical performances of the redox mediators showed that the role of the redox mediators is to accelerate the electron transfer of Cr(VI) bioreduction. This work will extend the practical application of redox mediators in accelerating the biotransformation of recalcitrant pollutants.

Supplemental data for this article can be accessed at http://dx.doi.org/10.1080/13102818.2015.1134277.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This study was supported by the National Natural Science Foundation of China [grant number 51208170]; the Hebei Natural Science Foundation for Distinguished Young Scholars [grant number E2012208012]; the Supporting Program for Hundred Outstanding Innovation Talents in Hebei Universities (II) [grant number BR2-211]; the Program for New Century Excellent Talents in University [grant number NCET-10-0127].