Evaluating Enhanced Sulfate Reduction and Optimized Volatile Fatty

Jan 21, 2015 - *(B.-J.N.) Phone: +61 7 33463230; fax +61 7 33654726; e-mail: [email protected]., *(Y.Z.) Phone: +86 411 84706460; fax: +86 411 84706263; ...
0 downloads 7 Views 2MB Size
Download PDF
Article pubs.acs.org/est

Evaluating Enhanced Sulfate Reduction and Optimized Volatile Fatty Acids (VFA) Composition in Anaerobic Reactor by Fe (III) Addition Yiwen Liu,† Yaobin Zhang,*,‡ and Bing-Jie Ni*,† †

Advanced Water Management Centre, The University of Queensland, St. Lucia, Queensland 4072, Australia Key Laboratory of Industrial Ecology and Environmental Engineering, Ministry of Education, School of Environmental Science and Technology, Dalian University of Technology, Dalian 116024, China

‡

S Supporting Information *

ABSTRACT: Anaerobic reactors with ferric iron addition have been experimentally demonstrated to be able to simultaneously improve sulfate reduction and organic matter degradation during sulfate-containing wastewater treatment. In this work, a mathematical model is developed to evaluate the impact of ferric iron addition on sulfate reduction and organic carbon removal as well as the volatile fatty acids (VFA) composition in anaerobic reactor. The model is successfully calibrated and validated using independent long-term experimental data sets from the anaerobic reactor with Fe (III) addition under different operational conditions. The model satisfactorily describes the sulfate reduction, organic carbon removal and VFA production. Results show Fe (III) addition induces the microbial reduction of Fe (III) by iron reducing bacteria (IRB), which significantly enhances sulfate reduction by sulfate reducing bacteria (SRB) and subsequently changes the VFA composition to acetate-dominating effluent. Simultaneously, the produced Fe (II) from IRB can alleviate the inhibition of undissociated H2S on microorganisms through iron sulfide precipitation, resulting in further improvement of the performance. In addition, the enhancement on reactor performance by Fe (III) is found to be more significantly favored at relatively low organic carbon/SO42− ratio (e.g., 1.0) than at high organic carbon/SO42− ratio (e.g., 4.5). The Fe (III)-based process of this work can be easily integrated with a commonly used strategy for phosphorus recovery, with the produced sulfide being recovered and then deposited into conventional chemical phosphorus removal sludge (FePO4) to achieve FeS precipitation for phosphorus recovery while the required Fe (III) being acquired from the waste ferric sludge of drinking water treatment process, to enable maximum resource recovery/reuse while achieving high-rate sulfate removal.

■

INTRODUCTION

Dissimilatory ferric iron reduction is an important terminal electron accepting process in many anaerobic environments.11−14 Iron reducing bacteria (IRB) can use ferric iron as an electron acceptor to oxidize organic matters.15 For the organic wastewater rich with sulfate, both IRB and SRB could utilize organic acids (VFAs) to mineralize organic carbon.16−18 Recently, Zhang et al.19 demonstrated that the addition of Fe2O3 in the acidogenic sulfate reducing reactor could substantially enhance microbial sulfate reduction by ca. 30%. Theoretically, 0.67 g organic carbon is removed associated with the reduction of 1 g SO42− by SRB.20 Therefore, the enhancement of sulfate reducing process could also be an alternative way to improve organic carbon removal for sulfaterich wastewater treatment instead of the sensitive methanogenesis process. Furthermore, it has been reported that sulfate reduction can occur at a short hydraulic retention time (HRT)

Anaerobic treatment process has been widely applied to the treatment of organic industrial wastewater due to low operational cost and high removal efficiency.1−6 Under anaerobic conditions, organic matter is initially hydrolyzed and then fermented into acetate acid and hydrogen, which can be further utilized by methanogenic archaea (MA) for methane production.7 Industries such as pharmaceutical, chemical units, and paper production often discharge wastewater containing high sulfate concentrations. The sulfate-rich wastewater can induce some significant problems in terms of the anaerobic treatment performance, such as the deterioration of methanogenesis.8 Dissimilatory sulfate reduction associated with the growth of sulfate reducing bacteria (SRB) is inevitably involved in anaerobic reactors. It is well-known that SRB can compete with MA for substrate (i.e., acetate and hydrogen) under sulfate-rich conditions.9,10 In addition, the activity of SRB would lead to a high concentration of undissociated hydrogen sulfide, which in turn would result in inhibition on MA and eventually the failure of the anaerobic reactor. © 2015 American Chemical Society

Received: Revised: Accepted: Published: 2123

August 27, 2014 January 19, 2015 January 21, 2015 January 21, 2015 DOI: 10.1021/es504200j Environ. Sci. Technol. 2015, 49, 2123−2131

Article

Environmental Science & Technology in the acidogenic phase,21 which is more competitive than methanogenesis that requires a longer HRT and is more susceptible to environmental conditions.8 Mathematical modeling of anaerobic treatment processes is of great importance toward a full understanding of the system and optimization of its practical application. The well-known Anaerobic Digestion Model No. 1 (ADM1) has been established and widely applied to simulate anaerobic processes previously.22 The ADM1 comprises multiple steps to describe biochemical processes in anaerobic degradation, namely disintegration, hydrolysis, acidogenesis, acetogenesis, and methanogenesis. Several extensions have been developed and incorporated into the original model, such as the processes accounting for homoacetogenesis,23 inhibitory effects of long chain fatty acids,24 and acetic oxidation pathway.25 However, there is still lack of a structure model to describe the simultaneous sulfate and iron reduction processes. In particular, little effort has been dedicated to modeling the anaerobic reactor treating sulfate-rich wastewater under ferric iron reducing condition, which involves complex microbial interactions between acidogens, acetogens, SRB and IRB as well as inhibitory effects of undissociated hydrogen sulfide. Therefore, the aim of this study was to develop a mathematical model based on ADM1 in order to evaluate the impact of ferric iron addition on sulfate reduction and organic carbon removal as well as the VFA composition in anaerobic reactor. The developed model is calibrated and validated to describe independent long-term experimental data sets from the anaerobic reactor with Fe (III) addition under different operational conditions and provide insights into the dynamics of sulfate reduction, organic carbon removal and VFA production in the system.

even at a neutral pH of ca. 7 with extremely low rate of Fe(III) dissolution. In particular, under the unlimited organic substrate (VFAs) condition in our experimental system,19 the competition between IRB and SBR for electron donors (VFA) is actually not a limiting factors for their respective metabolism. Instead, the functions of these two species on consuming extra organic acids could create a better pH condition for their coexistence, which has been clearly confirmed by the microbial analysis using 16S rRNA gene sequences.19 Therefore, in this work, microbial iron reduction processes is integrated with the sulfate reduction processes to form the new model in order to describe and evaluate the impact of Fe (III) addition on sulfate reduction and organic carbon removal as well as the VFA composition in anaerobic reactor. In the anaerobic reactor treating sulfate-containing wastewater, the addition of Fe2O3 would induce the release of Fe (II) from Fe (III) reduction by IRB (through utilizing Fe (III) from Fe2O3 dissolution or directly on the Fe2O3 oxide surface), and the produced Fe (II) by IRB can precipitate with sulfide.19 Our aim is to develop a practically applicable model. As such, the model does not replicate all the possible involved processes to avoid overparameterization. In the new model, these complex conversion processes were simplified into following kinetic processes: Fe2O3 corrosion process (Fe2O3 + 6H+ ⇒ 2Fe3+ + 3H2O), microbial iron reduction (Fe3+ + e− ⇒ Fe2+) and iron sulfide precipitation processes (Fe2+ + S2− ⇒ FeS). The possible chemical reduction of Fe (III) by sulfide was not considered specifically in this model as it likely contributed negligible part of iron reduction due to the slow reaction rate.34 SRB and IRB would compete for organic carbon (i.e., VFAs). The H2S produced by SRB have inhibition effects on both SRB and IRB. In addition, sulfate reduction could be promoted by Fe2O3 addition in this system, as demonstrated by Zhang et al.19 Therefore, Fe2O3 corrosion, sulfate and ferrous iron reduction, inhibition of undissociated H2S, ferrous sulfide precipitation, and promotion of Fe2O3 addition on SRB processes were incorporated into ADM1 to form the new model. The developed model was summarized in Table S1−S5 in Supporting Information (SI). Table S6 in the SI lists the definitions, values, units, and sources of all parameters used in the developed model. The new model describes the relationships among nine soluble species, that is, sugar, butyrate, propionate, acetate, hydrogen, sulfate, sulfide, ferric and ferrous iron; as well as 15 particulate species, that is, sugar degraders, butyrate degraders, propionate degraders, IRBs (grown on butyrate, propionate, acetate and hydrogen), SRBs (grown on butyrate, propionate, acetate and hydrogen), slowly biodegradable organic carbon, inert organic carbon, Fe2O3 and ironsulfide precipitation (SI Table S1). Five types of biological processes were considered, namely hydrolysis, acidogenesis, acetogenesis, iron, and sulfate reduction (SI Table S2 and S3). Acidogenesis converts the biodegradable organic carbon to four fermentation products, namely H2, acetate acid, propionate acid and butyrate acid. Acetogenesis includes uptake of butyrate acid and propionate acid. Both IRBs and SRBs consume H2, acetate acid, propionate acid and butyrate acid to reduce ferric iron and sulfate for ferrous iron and sulfide production, respectively.35 Kinetic control of all the enzymatic reaction rates is described by the Michaelis−Menten equation. The hydrolysis of slowly degradable substrate, which is either from the wastewater or generated through biomass decay, can produce the biodegrad-

■

MATERIALS AND METHODS Model Development. It is well-known that the rate of microbial iron reduction is closely related with the thermodynamic stability of iron (hydr)oxides. Solubility of iron (hydr)oxides generally decreases from ferrihydrite (Kso =10−39) to goethite (Kso =10−41) and to hematite (Kso =10−43) at circumneutral pH. The dissimilatory iron-reducing bacteria (GS-15) was reported to reduce a natural amorphic Fe(III) oxide but did not significantly reduce highly crystalline Fe(III) forms like hematite.26 Thus, IRB is generally believed to outcompete SBR for organic substrates when microbiologically reducible Fe (hydr)oxides (i.e., poorly crystalline phases) are available,27,28 while DIRB may not outcompete DSRB in the presence of crystalline Fe(III) phases (e.g., Fe2O3). However, microbial iron reduction does not necessarily require high-level direct dissolution of Fe2O3. In fact, the thermodynamic properties (ΔGr, Ks) of iron (hydr)oxides was recently found to have only a secondary control on Fe(III) reduction rates by iron-reducing bacteria.29 Instead, Fe(III) reduction rates of various Fe (hydr)oxides appear to be correlated with surface area.29,30 The competitiveness of bacterial Fe(III) oxide reduction as a pathway for organic matter oxidation in anoxic environments is not based on assumed thermodynamic properties of the Fe(III) oxide phase(s), leading to potential direct microbial hematite reduction with the presence of sulfate reduction other than dissolved Fe(III) reduction. For example, it has been shown that crystalline Fe(III) oxides are preferentially reduced and responsible for the oxidation of carbon within some natural sediments.31 In many other studies,32,33 microbial reduction of Fe2O3 to Fe(II) can occur 2124

DOI: 10.1021/es504200j Environ. Sci. Technol. 2015, 49, 2123−2131

Article

Environmental Science & Technology

Figure 1. Model calibration results using the experimental data from R1 (with Fe2O3) and R2 (without Fe2O3) during Phase I (0−60 days): (a) Effluent sulfate and sulfide in R1; (b) Effluent sulfate and sulfide in R2; (c) Effluent organic carbon and VFA in R1; (d) Effluent organic carbon and VFA in R2; (e) Sulfate and organic carbon removal in R1; (f) Sulfate and organic carbon removal in R2.

Experimental Data for Model Evaluation. Experimental data from an upflow anaerobic sludge bed (UASB) reactor with Fe2O3 addition (R1) and another identical UASB reactor without Fe2O3 addition (R2) treating same sulfate-containing wastewater previously reported in Zhang et al.19 were used for the model calibration and validation. R1 had a working volume of 2 L (Φ100 mm × 280 mm), in which about 60 g Fe2O3 powder (Tianjing Chemical Reagent Factory, China) was dosed into the reactor. R2 had the same setting as R1 except without the Fe2O3 addition, which was run as a control reactor (R2) for comparison. The seed sludge for both reactors was collected from a laboratory-scale UASB reactor treating sucrose wastewater. The ratio of volatile suspended substances to total suspended substances (VSS/TSS) of the sludge was 0.74. About 1-L sludge was inoculated in each reactor, resulting in an initial TSS of 14.2 g/L in both reactors. Both R1 and R2 were continuously operated at a HRT of 5 h and a temperature of 25 ± 1 °C, fed with an effluent from an acidogenic reactor treating a sucrose wastewater. The detailed composition of this wastewater could be found in Zhang et al.19

able organic carbon. Four physicochemical processes, namely acid−base reactions (including butyrate, propionate, acetate, inorganic carbon, inorganic nitrogen and sulfide, SI Table S4), liquid−gas transfer (including CO2, H2S and H2, SI Table S5), Fe2O3 corrosion (r13 = kFe(XFe3/(KFe3 + XFe3)), SI Table S3) and sulfide precipitation with ferrous ions (r14 = kpreSH2SSFe2, SI Table S3) are included in the model. In addition, the inhibition effects of pH, hydrogen, sulfide and ammonia are also included. A promotion factor (Pfe3) was added into the kinetics of sulfate reduction processes in order to describe the promotion effect of Fe2O3.19 The experimental data revealed that Fe (III) addition induced the microbial reduction of Fe (III) by IRB, forming a synthetic interaction between SRB and IRB, which stimulated sulfate reduction rate by SRB. To describe this enhancement after Fe (III) addition, we employed this promotion factor to capture the enhancement of Fe (III) addition on sulfate reduction rate. This kinetic approach has been previously applied successfully for describing improved activity by iron addition.36 Furthermore, the undissociated H2S could also inhibit the acetogenesis, iron reduction and sulfidogenesis. 2125

DOI: 10.1021/es504200j Environ. Sci. Technol. 2015, 49, 2123−2131

Article

Environmental Science & Technology

Figure 2. Model validation results using the experimental data from R1 (with Fe2O3) and R2 (without Fe2O3) during Phase I (61−100 days): (a) Effluent sulfate and sulfide in R1; (b) Effluent sulfate and sulfide in R2; (c) Effluent organic carbon and VFA in R1; (d) Effluent organic carbon and VFA in R2; (e) Sulfate and organic carbon removal in R1; (f) Sulfate and organic carbon removal in R2.

studies (e.g., ADM1). Thus, literature values reported with similar model structure were directly adopted for these parameters, as presented in SI Table S6. However, limited information is available in literature for the parameters related to the Fe2O3 corrosion and the promotion effects of Fe2O3 addition. Parameter estimation based on experimental measurements was then carried out for these new parameters, that is, maximum uptake rate of Fe2O3 (k Fe), uptake affinity constant for Fe2O3 (K Fe3), promotion constant of Fe2O3 (KP,Fe), uptake affinity constant for Fe3+ by IRB (K Fe3+), and inhibition constant by H2S on IRB (K h2s,irb) (SI Table S6). The model was calibrated using experimental data of sulfate reduction, sulfide production, organic carbon removal and VFA variation during Phase I (0−60 days) from both R1 (with Fe2O3 addition) and R2 (without Fe2O3 addition), which include three different operational sulfate-feeding conditions, that is, 300 mg/L, 700 mg/L, and 1000 mg/L. The parameter values were estimated by minimizing the sum of squares of the

A designed amount of sodium sulfate was added to this wastewater as sulfate source before feeding to R1 and R2. The influent organic carbon for R1 and R2 was around 1400 mg chemical oxygen demand (COD)/L during the 100-days operation. The influent sulfate concentrations for R1 and R2 were stepwise increased from 300 mg/L to 700 mg/L, 1000 mg/L and finally to 1400 mg/L during the 100-days operation, forming different operational phases with different organic carbon/SO42− ratio conditions. More details about the reactor setup and operation were described in Zhang et al.19 The effluent organic carbon, sulfate, sulfide and VFAs were monitored to evaluate and compare the performance of the two reactors, which were used for model calibration and validation. Model Calibration and Validation. The developed model includes 25 biochemical processes and 60 stoichiometric and kinetic parameters as summarized in SI Table S2 and S3. Most of model parameter values are well established in previous 2126

DOI: 10.1021/es504200j Environ. Sci. Technol. 2015, 49, 2123−2131

Article

Environmental Science & Technology

Figure 3. Model validation results using the experimental data of the percentage VFA compositions in reactor effluents of R1 and R2 during Phase II (61−100 days): (a) VFA composition in R1 (with Fe2O3); and (b) VFA composition in R2 (without Fe2O3).

(Figure 2e) due to the fact that more sulfate was removed by utilizing organic carbon. The effluent sulfide concentration and pH averaged at ca. 84 mg-S/L (Figure 2a) and 5.6, respectively, similar to those of R2, due to the iron sulfide precipitation. These results clearly indicated the enhanced sulfate reduction and organic carbon removal by Fe (III) addition in anaerobic reactors, particularly under high sulfate loading conditions (low organic carbon/SO42− ratio). Model Calibration. A two-step procedure was applied to calibrate the new model. In the first step, the regular kinetics of hydrolysis, acidogenesis, acetogenesis, and sulfate reduction were tested using the organic carbon, VFA, sulfate and sulfide data from the control reactor without Fe2O3 addition (R2) during Phase I (0−60 days, right panel in Figure 1). Then, the Fe (III)-related parameters were further calibrated using the experimental data from the anaerobic reactor with Fe2O3 addition (R1) during Phase I (0−60 days, left panel in Figure 1) in the second step. In this work, the literature reported parameter values could well describe the organic carbon, VFA, sulfate and sulfide profiles without Fe2O3 addition in R2, as shown in Figure 1. We then calibrated the developed model with the Fe (III)related parameters (SI Table S6). The calibration of the new model involved optimizing these parameter values by fitting simulation results to the experimental data from R1 with Fe2O3 addition during Phase I (0−60 days, left panel in Figure 1). Figure 1 compares the model predicted results with the experimental data from R1. The model predictions regarding the enhanced sulfate reduction and organic carbon removal as well as the sulfide production and VFA profiles matched very well with the experimental measurements (Figure 1). The calibrated parameter values giving the optimum model fittings with the experimental data are listed in SI Table S6. There is no statistical difference of model prediction between R1 and R2 (p > 0.05) and experimental data between R1 and R2 (p > 0.05). The good agreement between these simulated and measured data supported that the developed model properly captures the impact of Fe (III) addition on sulfate reduction and organic carbon conversion in anaerobic reactor. Parameter values of Fe2O3 corrosion (kFe and K Fe3) and the promotion effects of Fe2O3 addition (KP,Fe) giving the optimum model fit with the experimental data are listed in SI Table S6.

deviations between the measured data and the model predictions using the secant method embedded in AQUASIM 2.1d.37 Model validation was then carried out with the calibrated model parameters (SI Table S6) by the other sets of the monitoring data from both long-term operations of both R1 and R2, with different dynamic inflow sulfate conditions which has not been used to estimate the parameters during model calibration. The validation data sets of sulfate reduction, sulfide production, organic carbon removal and VFA variation were from Phase II (61−100 days) of both R1 (with Fe2O3 addition) and R2 (without Fe2O3 addition), which include two different operational sulfate-feeding conditions (1000 mg/L and 1400 mg/L).

■

RESULTS Enhanced Sulfate Reduction and Organic Carbon Removal by Fe (III). At a low sulfate feeding concentration of 300 mg/L during the first 14 days, both R1 (with Fe2O3) and R2 (without Fe2O3) showed similar performances (Figure 1) in terms of sulfate removal with removal rates of ca. 52% and 50%, respectively. R1 exhibited an organic carbon removal of ca. 32%, slightly higher than that of ca. 26% in R2. Both sulfate and organic carbon removal efficiency in R1 and R2 showed a gradual increase tendency, likely due to the adaption of microorganisms to the environmental conditions. Therefore, the enhancement of Fe (III) on sulfate and organic carbon removal at a low SO42− loading concentration (high organic carbon/SO42− ratio) was not significant. With the stepwise increase of sulfate feeding concentrations from 300 to finally 1400 mg/L, R2 (without Fe2O3) showed substantial deteriorations in terms of sulfate removal (gradually dropped from 50% to 31%, Figure 2b). The organic carbon removal in R2 only increased slightly from ca. 26% to 30% (Figure 2d). The effluent sulfide and pH were ca. 82 mg-S/L (Figure 2b) and 5.2, respectively. In contrast, sulfate removal in R1 (with Fe2O3) increased from 52% to 56% and kept relatively stable despite of the significantly increasing sulfate loading (Figure 2a). Following each shock (sudden increase of influent sulfate concentrations), the sulfate removal of R1 presented much quicker recovery than that of R2. The organic carbon removal also substantially increased from 32% to ca. 51% 2127

DOI: 10.1021/es504200j Environ. Sci. Technol. 2015, 49, 2123−2131

Article

Environmental Science & Technology

developed model was confirmed by independent data sets from anaerobic systems with and without Fe (III) addition under different operational conditions. The successful application of the model in this work indicates it is applicable to describe the enhanced sulfate reduction and optimized VFA composition in anaerobic reactor by Fe (III) addition. Modeling results indicated that both R1 and R2 exhibited similar performances with respect to sulfate and organic carbon removal at the beginning of the study (0−14 days) due to the relative low IRB biomass and activity (Figure 1). The slightly improved sulfate removal at this stage was attributing to the FeS precipitation, which alleviated the H2S inhibition on SRB and other acetogenic bacteria and thus indirectly increased sulfate reduction and organic carbon removal. With the increase of sulfate feeding concentrations from 1000 to 1400 mg/L, the model simulation showed an increased sulfate removal in R1 (Figure 2a). On the contrary, the sulfate removal in R2 decreased significantly (Figure 2b). The decreased performance of R2 was due to the strong inhibitions on the acetogenesis and sulfate reduction processes by undissociated H2S.42 In contrast, H2S inhibition was partially reduced by FeS precipitate in R1. It has been demonstrated that SRB could grow in acidic conditions without significant pH inhibitory effect at pH of 5−6.43 Thus, the SRB metabolism and turnover rates in the two reactors would not varied significantly at pH range of 5−6. In addition, we also performed additional simulation for R2 using the same pH condition as R1 and the results showed the sulfate and organic carbon removal in R2 did not change much, confirming that pH should not determine the different performances between R1 and R2. Previous experimental studies also showed minor interference of pH (e.g., 5−6) on the activity of SRB and/or IRB.43,44 The contribution of IRB to total organic removal (∼2%) was much less compared to SRB (∼98%), indicating IRB played a minor role in terms of direct organic utilization. However, the organic removal rate in R1 (42.2%, with Fe2O3 addition) was significantly higher than in R2 (23.6%, without Fe2O3 addition), suggesting a synthetic interaction between SRB and IRB for the stimulated sulfate reduction by SRB with Fe2O3 addition. These results confirmed that the Fe (III) addition induced the microbial reduction of Fe (III) by IRB in anaerobic reactor, which could significantly enhance sulfate and organic carbon reduction by SRB and subsequently changes the VFA composition to acetatedominating effluent (much more propionate would be consumed by SRB). Simultaneously, the produced Fe (II) from IRB could alleviate the inhibition of undissociated H2S on SRB through iron sulfide precipitation, resulting in further improvement of the performance. This novel high-rate Fe (III)-based anaerobic sulfate and organic carbon removal technology could be easily operated under high sulfate loading concentrations and low HRTs condition with better performance as demonstrated in this work, thus having the potential to replace sensitive methanogenesis for organic carbon removal during treating high-sulfate containing wastewater.19 The generated acetate-dominating effluent with less propionate (Figure 3) could be beneficial for subsequent anaerobic processes for methane production or biological nitrogen removal processes as carbon source for denitrification.45,46 From an integrated environmental and economic perspective, nutrients source in water and wastewater treatment systems should be managed such that both good nutrients removal performance and high resource recovery or reuse can

These values clearly suggested a coexistence of IRB and SRB in the system with the SRB activity being significantly promoted, consistent with the experimental observations. The calibrated H2S inhibition constant K h2s,irb (340 g S m−3) is slightly higher than that of SRB (213−256 g S m−3), indicating a lower effect of H2S on IRB compared to SBR. The obtained value of uptake affinity constant by IRB K Fe3+ (10 g Fe m−3) is similar to the literature reported value (7.42 g Fe m−3).35 Model Validation. Model and parameter validation was performed based on the comparison between the model predictions (using the same parameters shown in SI Table S6) and the experimental data from both R1 and R2 during Phase II (61−100 days) with highly different operational sulfate loading concentration of 1000−1400 mg/L compared to that during Phase I (300−1000 mg/L), together with the different dynamics of influent organic carbon concentrations, resulting in reliable and solid validation of our model. The model and its parameters were first evaluated with the organic carbon, VFA, sulfate and sulfide data from R2 (without Fe2O3) during Phase II (61−100 days). The model predictions and the experimental results are shown in Figure 2 (right panel). The validation results show that the model predictions match the measured data in the validation experiment, which supports the validity of the developed model. The developed model and the parameters were then evaluated with the organic carbon, VFA, sulfate and sulfide data from R1 (with Fe2O3) during Phase II (61−100 days). The performance of R1 was clearly improved with increasing sulfate concentrations. In comparison, the treatment efficiency dropped significantly in R2. The model predictions well matched these experimental results in R1 as shown in Figure 2 (left panel), again supporting the validity of the developed model for anaerobic reactor with Fe (III) addition. The experimental results of the percentage VFA compositions in reactor effluents of R1 and R2 during Phase II (61−100 days) were also used to evaluate the developed model. The experimental and simulated VFA compositions are shown in Figure 3. The percentage of butyrate in R1 and R2 was similar. Comparatively, the percentage of acetate in R1 was about 53%, which is much higher than that of 38% in R2. The percentage of propionate in R1 was about 31%, which was significantly lower than that of 42% in R2, likely due to the enhanced sulfate reduction in R1 would consume more propionate than in R2. As can be seen in Figure 3, the model predictions are consistent with these experimental observations, further suggesting that the model is appropriate to describe the impact of Fe (III) in anaerobic reactors.

■

DISCUSSION Recently, anaerobic biological treatment of sulfate-containing wastewater has attracted more attentions due to the lack of effective point source treatment technology.8,38−41 Among them, biological sulfate reduction with ferric iron addition is a novel and promising process for simultaneously enhancing sulfate reduction and organic carbon removal.19 In this work, a mathematical model considering the interactions between acidogens, acetogens, SRB, IRB, Fe (III), and Fe (II) in anaerobic reactor with Fe2O3 addition was constructed for the first time based on the known metabolisms. The set of best-fit parameter values are shown in SI Table S6. The parameter values obtained were robust in their ability to predict sulfate, sulfide, organic carbon and VFA dynamics during the long-term operation under different conditions. The validity of this 2128

DOI: 10.1021/es504200j Environ. Sci. Technol. 2015, 49, 2123−2131

Article

Environmental Science & Technology

Figure 4. A proposed strategy to integrate Fe (III)-based enhancing anaerobic sulfate removal with phosphorus recovery, with the produced sulfide being recovered and then deposited into conventional chemical phosphorus removal sludge (FePO4) to achieve FeS precipitation for phosphorus recovery while the required Fe (III) being acquired from the ferric sludge of drinking water treatment process, to enable maximum resource recovery/reuse while achieving high-rate sulfate removal.

be achieved. Based on the findings of this work, a new strategy could be proposed to treat high sulfate containing wastewater as well as resource recovery/reuse through integrating the Fe (III)-based enhancing anaerobic sulfate removal system with a commonly used strategy for phosphorus recovery, as presented in Figure 4. Instead of externally dosing Fe2O3, ferrous iron could be economically acquired from the waste ferric sludge in coagulation process at drinking water treatment process.47,48 The produced sulfide in the effluent of the Fe (III)-based anaerobic sulfate removal reactor can be recovered through sparging and absorbing by sodium hydroxide.49 The recovered sulfide could then be deposited into the FePO4 sludge generated from conventional chemical phosphorus removal process (the ferric sludge from drinking water treatment plant could also be used in this process, Figure 4) to achieve FeS precipitation for phosphorus recovery.50,51 The “cleaned” effluent without sulfide but with plentiful biodegradable substrate (mainly VFAs) dominating with acetate can be delivered to other components in wastewater treatment plant, such as anaerobic methane production or biological nitrogen removal (Figure 4). The possibility of ferric sludge recycling within the wastewater and drinking water process loop (Figure 4) would not only represent a significant process cost reduction, but also improve the sulfate-containing wastewater treatment, enabling maximum resource recovery/reuse while achieving high-rate sulfate removal. In summary, a mathematical model is developed based on ADM1 to evaluate the impact of Fe (III) addition on sulfate reduction and organic carbon removal as well as the VFA composition in anaerobic reactor. The developed model has been calibrated and validated to reproduce the experimental data from long-term operation of two reactors with different operational conditions successfully. The modeling results

confirm that Fe (III) addition induced the microbial reduction of Fe (III) by IRB, which could significantly enhance sulfate reduction by SRB and subsequently changes the VFA composition to acetate-dominating effluent. The Fe (III)based anaerobic sulfate removal technology can be integrated with a commonly used strategy for phosphorus recovery.

■

ASSOCIATED CONTENT

S Supporting Information *

Additional information as noted in the text. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Authors

*(B.-J.N.) Phone: +61 7 33463230; fax +61 7 33654726; email: [email protected]. *(Y.Z.) Phone: +86 411 84706460; fax: +86 411 84706263; email: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This study was supported by the Australian Research Council (ARC) through Project DP130103147. Yiwen Liu gratefully received the Endeavour International Postgraduate Research Scholarship (IPRS) and The University of Queensland Centennial Scholarship (UQCent). Bing-Jie Ni acknowledges the supports of ARC Discovery Early Career Researcher Award (DE130100451) and ARC Linkage Project (LP110201095). Yaobin Zhang acknowledges the supports of National Natural Scientific Foundation of China (51378087 and 21177015). 2129

DOI: 10.1021/es504200j Environ. Sci. Technol. 2015, 49, 2123−2131

Article

Environmental Science & Technology

■

(21) Mizuno, O.; Li, Y. Y.; Noike, T. The behavior of sulfate-reducing bacteria in acidogenic phase of anaerobic digestion. Water Res. 1998, 32, 1626−1634. (22) Batstone, D. J.; Keller, J.; Angelidaki, I.; Kalyuzhnyi, S. V.; Pavlostathis, S. G.; Rozzi, A.; Sanders, W. T. M.; Siegrist, H.; Vavilin, V. A. The IWA anaerobic digestion model no 1(ADM 1). Water Sci. Technol. 2002, 45, 65−73. (23) Ni, B. J.; Liu, H.; Nie, Y. Q.; Zeng, R. J.; Du, G. C.; Chen, J.; Yu, H. Q. Coupling glucose fermentation and homoacetogenesis for elevated acetate production: Experimental and mathematical approaches. Biotechnol. Bioeng. 2011, 108, 345−353. (24) Zonta, Ž .; Alves, M. M.; Flotats, X.; Palatsi, J. Modelling inhibitory effects of long chain fatty acids in the anaerobic digestion process. Water Res. 2013, 47, 1369−1380. (25) Shimada, T.; Morgenroth, E.; Tandukar, M.; Pavlostathis, S. G.; Smith, A.; Raskin, L.; Kilian, R. E. Syntrophic acetate oxidation in twophase (acid-methane) anaerobic digesters. Water Sci. Technol. 2011, 64, 1812−1820. (26) Lovley, D. R.; Phillips, E. J. P. Novel mode of microbial energy metabolism: Organic carbon oxidation coupled to dissimilatory reduction of iron or manganese. Appl. Environ. Microbiol. 1988, 54, 1472−1480. (27) Lovley, D. R.; Phillips, E. J. P. Competitive mechanisms for inhibition of sulfate reduction and methane production in the zone of ferric iron reduction in sediments. Appl. Environ. Microbiol. 1987, 53, 2636−2641. (28) Chapelle, F. H.; Lovley, D. R. Competitive exclusion of sulfate reduction by Fe (lll)-reducing bacteria: A mechanism for producing discrete zones of high-iron ground water. Groundwater 1992, 30, 29− 36. (29) Roden, E. E. Fe (III) oxide reactivity toward biological versus chemical reduction. Environ. Sci. Technol. 2003, 37, 1319−1324. (30) Roden, E. E.; Zachara, J. M. Microbial reduction of crystalline iron (III) oxides: Influence of oxide surface area and potential for cell growth. Environ. Sci. Technol. 1996, 30, 1618−1628. (31) Stucki, J. W.; Lee, K.; Goodman, B. A.; Kostka, J. E. Effects of in situ biostimulation on iron mineral speciation in a sub-surface soil. Geochim. Cosmochim. Acta 2007, 71, 835−843. (32) Yan, B.; Wrenn, B. A.; Basak, S.; Biswas, P.; Giammar, D. E. Microbial reduction of Fe (III) in hematite nanoparticles by Geobacter sulf urreducens. Environ. Sci. Technol. 2008, 42, 6526−6531. (33) Lovley, D. R.; Phillips, E. J. P. Organic matter mineralization with reduction of ferric iron in anaerobic sediments. Appl. Environ. Microbiol. 1986, 51, 683−689. (34) Poulton, S. W.; Krom, M. D.; Raiswell, R. A revised scheme for the reactivity of iron (oxyhydr) oxide minerals towards dissolved sulfide. Geochim. Cosmochim. Acta 2004, 68, 3703−3715. (35) Somasundaram, V.; Philip, L.; Bhallamudi, S. M. Experimental and mathematical modeling studies on Cr (VI) reduction by CRB, SRB and IRB, individually and in combination. J. Hazard. Mater. 2009, 172, 606−617. (36) Xiao, X.; Sheng, G.-P.; Mu, Y.; Yu, H.-Q. A modeling approach to describe ZVI-based anaerobic system. Water Res. 2013, 47, 6007− 6013. (37) Reichert, P. Aquasim 2.0-User Manual, Computer Program for the Identification and Simulation of Aquatic Systems; Swiss Federal Institute for Environmental Science and Technology (EAWAG), 1998. (38) Sierra-Alvarez, R.; Hollingsworth, J.; Zhou, M. S. Removal of copper in an integrated sulfate reducing bioreactor-crystallization reactor system. Environ. Sci. Technol. 2007, 41, 1426−1431. (39) Amaral, F. M.; Kato, M. T.; Florêncio, L.; Gavazza, S. Color, organic matter and sulfate removal from textile effluents by anaerobic and aerobic processes. Bioresour. Technol. 2014, 163, 364−369. (40) Ozdemir, S.; Cirik, K.; Akman, D.; Sahinkaya, E.; Cinar, O. Treatment of azo dye-containing synthetic textile dye effluent using sulfidogenic anaerobic baffled reactor. Bioresour. Technol. 2013, 146, 135−143.

REFERENCES

(1) Chan, Y. J.; Chong, M. F.; Law, C. L.; Hassell, D. G. A review on anaerobic−aerobic treatment of industrial and municipal wastewater. Chem. Eng. J. 2009, 155, 1−18. (2) Jeganathan, J.; Nakhla, G.; Bassi, A. Long-term performance of high-rate anaerobic reactors for the treatment of oily wastewater. Environ. Sci. Technol. 2006, 40, 6466−6472. (3) Van Eekert, M. H. A.; Van Ras, N. J. P.; Mentink, G. H.; Rijnaarts, H. H. M.; Stams, A. J. M.; Field, J. A.; Schraa, G. Anaerobic transformation of β-hexachlorocyclohexane by methanogenic granular sludge and soil microflora. Environ. Sci. Technol. 1998, 32, 3299−3304. (4) Choi, W.-H.; Shin, C.-H.; Son, S.-M.; Ghorpade, P. A.; Kim, J.-J.; Park, J.-Y. Anaerobic treatment of palm oil mill effluent using combined high-rate anaerobic reactors. Bioresour. Technol. 2013, 141, 138−144. (5) Liu, Y.; Xu, H.-L.; Yang, S.-F.; Tay, J.-H. Mechanisms and models for anaerobic granulation in upflow anaerobic sludge blanket reactor. Water Res. 2003, 37, 661−673. (6) Hulshoff Pol, L. W.; de Castro Lopes, S. I.; Lettinga, G.; Lens, P. N. L. Anaerobic sludge granulation. Water Res. 2004, 38, 1376−1389. (7) Massé, D. I.; Droste, R. L. Comprehensive model of anaerobic digestion of swine manure slurry in a sequencing batch reactor. Water Res. 2000, 34, 3087−3106. (8) Zhang, J.; Zhang, Y.; Quan, X.; Liu, Y.; An, X.; Chen, S.; Zhao, H. Bioaugmentation and functional partitioning in a zero valent ironanaerobic reactor for sulfate-containing wastewater treatment. Chem. Eng. J. 2011, 174, 159−165. (9) Chou, H. H.; Huang, J. S.; Chen, W. G.; Ohara, R. Competitive reaction kinetics of sulfate-reducing bacteria and methanogenic bacteria in anaerobic filters. Bioresour. Technol. 2008, 99, 8061−8067. (10) Yoda, M.; Kitagawa, M.; Miyaji, Y. Long term competition between sulfate-reducing and methane-producing bacteria for acetate in anaerobic biofilm. Water Res. 1987, 21, 1547−1556. (11) Yu, R.-Q.; Flanders, J. R.; Mack, E. E.; Turner, R.; Mirza, M. B.; Barkay, T. Contribution of coexisting sulfate and iron reducing bacteria to methylmercury production in freshwater river sediments. Environ. Sci. Technol. 2012, 46, 2684−2691. (12) Lovley, D. R. Organic matter mineralization with the reduction of ferric iron: A review. Geomicrobiol. J. 1987, 5, 375−399. (13) Achtnich, C.; Schuhmann, A.; Wind, T.; Conrad, R. Role of interspecies H2 transfer to sulfate and ferric iron-reducing bacteria in acetate consumption in anoxic paddy soil. FEMS Microbiol. Ecol. 1995, 16, 61−70. (14) Canfield, D. E.; Thamdrup, B.; Hansen, J. W. The anaerobic degradation of organic matter in Danish coastal sediments: Iron reduction, manganese reduction, and sulfate reduction. Geochim. Cosmochim. Acta 1993, 57, 3867−3883. (15) Somasundaram, V.; Philip, L.; Bhallamudi, S. M. Laboratory scale column studies on transport and biotransformation of Cr (VI) through porous media in presence of CRB, SRB and IRB. Chem. Eng. J. 2011, 171, 572−581. (16) Pruden, A.; Messner, N.; Pereyra, L.; Hanson, R. E.; Hiibel, S. R.; Reardon, K. F. The effect of inoculum on the performance of sulfate-reducing columns treating heavy metal contaminated water. Water Res. 2007, 41, 904−914. (17) Yang, F.; Shi, B.; Bai, Y.; Sun, H.; Lytle, D. A.; Wang, D. Effect of sulfate on the transformation of corrosion scale composition and bacterial community in cast iron water distribution pipes. Water Res. 2014, 59, 46−57. (18) Wang, H.; Hu, C.; Hu, X.; Yang, M.; Qu, J. Effects of disinfectant and biofilm on the corrosion of cast iron pipes in a reclaimed water distribution system. Water Res. 2012, 46, 1070−1078. (19) Zhang, J.; Zhang, Y.; Chang, J.; Quan, X.; Li, Q. Biological sulfate reduction in the acidogenic phase of anaerobic digestion under dissimilatory Fe (III)Reducing conditions. Water Res. 2013, 47, 2033−2040. (20) Isa, Z.; Grusenmeyer, S.; Verstraete, W. Sulfate reduction relative to methane production in high-rate anaerobic digestion: Technical aspects. Appl. Environ. Microbiol. 1986, 51, 572−579. 2130

DOI: 10.1021/es504200j Environ. Sci. Technol. 2015, 49, 2123−2131

Article

Environmental Science & Technology (41) Bratkova, S.; Koumanova, B.; Beschkov, V. Biological treatment of mining wastewaters by fixed-bed bioreactors at high organic loading. Bioresour. Technol. 2013, 137, 409−413. (42) O’Flaherty, V.; Colohan, S.; Mulkerrins, D.; Colleran, E. Effect of sulphate addition on volatile fatty acid and ethanol degradation in an anaerobic hybrid reactor. II: Microbial interactions and toxic effects. Bioresour. Technol. 1999, 68, 109−120. (43) Elliott, P.; Ragusa, S.; Catcheside, D. Growth of sulfate-reducing bacteria under acidic conditions in an upflow anaerobic bioreactor as a treatment system for acid mine drainage. Water Res. 1998, 32, 3724− 3730. (44) Fredrickson, J. K.; Gorby, Y. A. Environmental processes mediated by iron-reducing bacteria. Curr. Opin. Biotechnol. 1996, 7, 287−294. (45) Uberoi, V.; Bhattacharya, S. K. Effects of chlorophenols and nitrophenols on the kinetics of propionate degradation in sulfatereducing anaerobic systems. Environ. Sci. Technol. 1997, 31, 1607− 1614. (46) Liu, Y.; Zhang, Y.; Zhao, Z.; Li, Y.; Quan, X.; Chen, S. Enhanced azo dye wastewater treatment in a two-stage anaerobic system with Fe0 dosing. Bioresour. Technol. 2012, 121, 148−153. (47) Mejia Likosova, E.; Keller, J.; Poussade, Y.; Freguia, S. A novel electrochemical process for the recovery and recycling of ferric chloride from precipitation sludge. Water Res. 2014, 51, 96−103. (48) Razali, M.; Zhao, Y. Q.; Bruen, M. Effectiveness of a drinkingwater treatment sludge in removing different phosphorus species from aqueous solution. Sep. Purif. Technol. 2007, 55, 300−306. (49) Sayers, R. H.; Langlais, R. J. Removal and Recovery of Sulfide Tannery Wastewater; Environmental Protection Agency, Office of Research and Development, Industrial Environmental Research Laboratory, 1977; Vol. 1. (50) Mejia Likosova, E.; Collins, R. N.; Keller, J.; Freguia, S. Anodic reactivity of ferrous sulfide precipitates changing over time due to particulate speciation. Environ. Sci. Technol. 2013, 47, 12366−12373. (51) Kato, F.; Kitakoji, H.; Oshita, K.; Takaoka, M.; Takeda, N.; Matsumoto, T. Extraction efficiency of phosphate from pre-coagulated sludge with NaHS. Water Sci. Technol. 2006, 54, 119−129.

2131

DOI: 10.1021/es504200j Environ. Sci. Technol. 2015, 49, 2123−2131