oxic wastewater treatment systems meeting increasingly stringent treatment standards from a life cycle perspective

Bioresource Technology 126 (2012) 31–40

Contents lists available at SciVerse ScienceDirect

Bioresource Technology journal homepage: www.elsevier.com/locate/biortech

Environmental profile of typical anaerobic/anoxic/oxic wastewater treatment systems meeting increasingly stringent treatment standards from a life cycle perspective Xu Wang a, Junxin Liu a,⇑, Nan-Qi Ren b, Zuoshan Duan c a b c

Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, 18 Shuangqing Road, Haidian District, Beijing 100085, PR China State Key Laboratory of Urban Water Resource and Environment, School of Municipal and Environmental Engineering, Harbin Institute of Technology, Harbin 150090, PR China China Northwest Municipal Engineering Design Institute Co. Ltd., CSCEC AECOM Consultants Co. Ltd., 459 Dingxi Road, Lanzhou 73000, PR China

h i g h l i g h t s " We employ a life cycle perspective to estimate the wastewater treatment systems. " Six typical A/A/O alternatives are evaluated from an environmental profile standpoint. " Sophisticated treatments are at the cost of higher resource consumption and GHG emissions. " The only positive trade-off with improved treatments is observed when using bioenergy recovery. " Optimal alternatives are identified from different positive perspectives.

a r t i c l e

i n f o

Article history: Received 17 May 2012 Received in revised form 7 August 2012 Accepted 4 September 2012 Available online 13 September 2012 Keywords: Wastewater treatment plants (WWTPs) Anaerobic/anoxic/oxic process Environmental burden Resource recovery Life cycle inventory (LCI)

a b s t r a c t Stringent new legislation for wastewater treatment plants (WWTPs) is currently motivating innovation and optimization of wastewater treatment technologies. Evaluating the environmental performance of a wastewater treatment system is a necessary precursor before proposing implementation of WWTPs designed to address the global requirements for reduced resource use, energy consumption and environmental emissions. However, developing overly-sophisticated treatment methods may lead to negative environmental effects. This study was conducted to employ a process modeling approach from a life cycle perspective to construct and evaluate six anaerobic/anoxic/oxic wastewater treatment systems that include a water line, sludge line and bioenergy recovery system and was designed to meet different treatment standards in China. The results revealed that improved treatments optimized for local receiving watercourses can be realized at the cost of higher resource consumption and greenhouse gas emissions. Optimal Scenarios were also identified from different positive perspectives. Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction Wastewater treatment plants (WWTPs) have been designed and operated to minimize the environmental effects of discharging untreated water into natural aquatic systems, with a focus on preventing eutrophication and health hazards in surface water. Global demographic trends as well as new stringent legislation for WWTPs are current motivators for the development and innovation of new treatment technologies, as well as optimization of existing ones (Guerrero et al., 2011; Liu et al., 2008; Machado et al., 2009). However, increasingly sophisticated improvements in treatment have led to increased resource degradation, higher electrical energy consumption, and elevated environmental ⇑ Corresponding author. Tel./fax: +86 0 10 62849133. E-mail addresses: wangxu_offi[email protected] (X. Wang), [email protected] (J. Liu). 0960-8524/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biortech.2012.09.009

emissions (Foley et al., 2010a). Taken together, these factors lead to increased environmental burdens. To date, these additional environmental loads have largely been neglected in the regulatory push for cleaner local water environments. Furthermore, numerous WWTP options have varied performance at different treatment levels and consequently varying direct effects on the environment. To this end, there is a need for comprehensive environmental assessments of a range of wastewater treatment options to meet different treatment standards from a life cycle perspective that primarily focus on broader environmental consequences. Life cycle assessment (LCA) is a tool for evaluation of the environmental consequence of goods, as well as processes or services (Dennison et al., 1998). To date, LCA has been used to investigate the environmental consequences of wastewater treatment systems in several existing cases, with a focus on interest in the construction and demolition phase of WWTPs (Hospido et al., 2004; Pasqualino

32

X. Wang et al. / Bioresource Technology 126 (2012) 31–40

et al., 2009). However, from the perspective of long-term installations, these phases are not as important as the operational phase, especially for large-scale treatment systems (Emmerson et al., 1995; Lundin et al., 2000; Tillman and Lundstrom, 1998; Zhang and Wilson, 2000). Additionally, some assessments have paid close attention to competing technology configurations and employed LCA to examine the obvious effects of energy consumption and/or materials depletion on the overall environmental influences of operation (Vidal et al., 2002; Wu et al., 2010). Although operation-phase analysis and common treatment alternatives have been explored, several shortcomings must still be addressed. For example, field data describing most treatment alternatives used for assessment have been collected from different WWTPs with varying flow rates and influent characteristics. These variations involved high levels of uncertainty in the conclusions of such studies. Even if increasingly stringent treatment standards have been proposed for the sake of environmental protection, the relevant environmental consequences of these efforts have not been thoroughly evaluated. Existing studies have highlighted the ability of WWTPs to prevent eutrophication in the receiving environment; therefore, enhanced levels of nutrient removal are considered to be highly beneficial (Gaterell et al., 2005; Lassaux et al., 2007). However, most studies that have been conducted to date are associated with North American and European treatment standards, and the outcomes of such analyses may not be applicable to the situations of developing countries such as China. The Chinese EPA passed the Discharge Standard of Pollutants for Municipal Wastewater Treatment Plants (MEP, 2002) in 1996, which includes four Classes of treatment standards (Class 3, Class 2, Class 1B and Class 1A), with Class 1A being the most stringent. Class 1A is required when effluent is reused or discharged to landscape water (e.g., water used for recreation, scenic viewing and wetland recharge) that has low dilution ability. Few studies have assessed the environmental impacts associated with different nitrogen and

phosphorus removal requirements (Rodriguez-Garcia et al., 2011). Moreover, few studies have considered the local situation in China. Further, although different combinations of anaerobic/anoxic/oxic systems have been employed in environmental technology applications for many years (Kang et al., 2011; Mulkerrins et al., 2000; Wang et al., 2011; Zeng et al., 2010; Zhou et al., 2011), little research has been conducted to compare environmental impacts among combinations, let alone environmental performance throughout their life cycle under various treatment standards (Rodriguez-Garcia et al., 2011). Based on the literature described above, the internationally standardized LCA framework (Guinee, 2002) was employed as a guideline to quantitatively analyze anaerobic/anoxic/oxic wastewater treatment systems to determine if they meet different treatment standards. The primary objective of this study was to provide a comprehensive indicator to include the environmental vector in the decision-making process when optimization or implementation of a selected system for a WWTP is planned. To accomplish this, analysis of the treatment Scenarios begins with the definition of six different biological anaerobic/anoxic/oxic systems, including the water line (treatment of the influent), sludge line (thickening, dewatering and anaerobic digestion of the waste activated sludge) and resource recovery from the excess sludge. Extensive analysis of all treatment Scenarios under major treatment standards (Class 2, 1B and 1A) in China is presented to highlight the variability in the life cycle environmental performance of a combination of alternatives. 2. Methodology 2.1. Goal and scope definitions This study was conducted to simulate and investigate a range of wastewater treatment Scenarios that achieve varying treatment standards from a life cycle perspective. The six Scenarios investi-

Fig. 1. Flow charts of six different full-scale anaerobic/anoxic/aerobic wastewater treatment systems investigated in this study.

X. Wang et al. / Bioresource Technology 126 (2012) 31–40 Table 1 Environmental parameters and main characteristics of influent for process design and modeling. Items

Value

Environmental parameters Average flow Peak wet weather flow ratio Ambient temperature Plant altitude

200 ML/d 1.2 20 °C 40 m

Influent main characteristics pH and alkalinity Chemical oxygen demand (COD) Biological oxygen demand in 5-days (BOD5) Total nitrogen (TN) Total Kjeldahl nitrogen (TKN) Total phosphorus (TP) Total suspended solids (TSS) Total inorganic suspended solids

7.2, 5 mmole/L 500 mg COD/L 256 mg BOD5/L 50 mg N/L 50 mg N/L 12 mg P/L 190 mg TSS/L 30 mg TSS/L

Note: All other environmental parameters and influent characteristics in the simulator were set at default values, except for those mentioned above.

Table 2 Major evaluation parameters under different treatment standards in this study. Parameters

COD BOD5 TSS TN NH3–N TP a

Levels of treatment standard (MEP, GB 18918–2002) (unit: mg/L) Class 2

Class 1B

Class 1A

100 30 30 – 25(30a) 3

60 20 20 20 8(15a) 1

50 10 10 15 5(8a) 0.5

Marks in brackets are the levels when water temperature is lower than 12 °C.

gated in this paper are described in Fig. 1. The steady-state raw wastewater characteristics and global plant parameters are described in Table 1. The water line of the six Scenarios covered the most common combinations of anaerobic/anoxic/oxic biological treatment alternatives. In contrast to traditional plants, all of the constructed WWTPs modeled and simulated in this study were designed without primary sedimentation to maintain the necessary carbon to nitrogen ratio of denitrification following the most general trends. In the sludge lines of the studied plants, the biological treatment sludge from the secondary clarifier is thickened and dewatered in a belt filter, and this process is coupled with hemophilic anaerobic digestion for solids stabilization and energy recovery via biogas combustion. Scenario 1 presents the conventional A2/O process, which is still employed in systems worldwide. Scenario 2 portrays one optimization of the conventional A2/O process that cancels the mixed liquor recirculation (MLR). One obvious feature of Scenario 2 is that its return activated sludge (RAS) rate is higher than that of the conventional A2/O process, which may be beneficial for denitrification. Scenario 3 presents another modification of the conventional A2/O process, with a focus on RAS distribution. In the conventional A2/O process, RAS is completely returned to the anaerobic zone in a process termed external recirculation. However, in Scenario 3, RAS is not only returned to the anaerobic zone (20% of the total RAS), but also to the anoxic zone (80% of the total RAS). One of the advantages of this technique may be that larger parts of RAS returning to the anoxic zone would prevent failure of anaerobic conditions in the anaerobic zone and mitigate the impact of nitrate on the phosphorus-releasing reaction. Scenario 4 describes another optimization of the conventional

33

A2/O process, the University of Cape Town (UCT) process, which provides an adjustment for MLR and RAS. In Scenario 4, MLR includes two different internal cycles, the traditional one between the oxic zone and the anoxic zone, and a newly added one between the anoxic zone and the anaerobic zone. In this Scenario, the nitrate concentration is greatly reduced owing to denitrification in the anoxic zone; thus, the MLR from the anoxic zone to the anaerobic zone would avoid the adverse effects on the anaerobic environment induced by the nitrate in the returning sludge. Scenario 5 and 6 are two different types of reversed A2/O processes. Scenario 5 presents the most common conventional reversed A2/ O process, while Scenario 6 employs a step-feed mode based on Scenario 5. In Scenario 6, 80% of the influent flows to the anoxic zone, while the rest is distributed in the anaerobic zone. Detailed design parameters of all Scenarios can be found in our previous study (Wang et al., 2012), and a portion of the treatment standards are presented in Table 2. In this study, treatment standards of Class 2, 1B and 1A were included for assessment. Class 3 was excluded owing to its relatively limited application. For this investigation, treatment of 2  105 m3/d of raw municipal wastewater (500 mg COD/L, 50 mg N/L, 12 mg P/d) over 20 years was selected as the functional unit (including sewage treatment and the subsequent sludge handling and bioenergy recovery system) based on large treatment goals owing to wide employment of the anaerobic/anoxic/oxic process in large-scale WWTPs. The system boundary was defined as the raw sewage arriving at the WWTP and included all discharges (liquid phase, solid phase and gas phase) to the local environments. Processes linked to the construction and demolition phase were ignored since they are generally negligible when compared with the operating phase (Lundin et al., 2000). Furthermore, collection and transportation of wastewater and/or sludge schemes were not considered in the present study because they were assumed to be shared by all the systems; therefore, only the operational process was taken into account during the analysis. Moreover, no consideration was given to the sewer system because it was assumed to be a separate system and no storm water enters the WWTPs. Finally, it was assumed that all of the treated effluents were released to environmentally sensitive watercourses. 2.2. Evaluation data acquisition In this study, one of the most popularly applied process simulation codes, BioWinÒ Simulator (V.3.0., BW3–1952), was employed for construction of all Scenarios and data acquisition. BioWin is a Microsoft Windows based simulator used worldwide for the analysis and design of WWTPs. This system employs the integrated Activated Sludge/Anaerobic Digestion model, which is also known as the BioWin General Model (Envirosim, 2007). The BioWin involves an integrated kinetic model and mass balance approach, as well as a pH/alkalinity calculation model, and comprises over 50 state variables through more than 60 process expressions. These expressions portray the biological processes that occur in activated sludge and anaerobic digestion systems, several chemical precipitation reactions, and gas–liquid mass transfer for gasses. Steadystate simulations under average conditions were modeled and simulated in this study. Three sets of error criteria were used in the unique solution analysis to ensure that the data obtained for the final assessment were unique, and the iterations were stopped if the mass balance was closer than the error criterion designated for the simulation. To quantify the environmental burden, the evaluation indicators primarily embrace chemical use, greenhouse gas (GHG) emissions and energy consumption. During operation of a WWTP, the above three items are viewed as important stimuli associated

34

X. Wang et al. / Bioresource Technology 126 (2012) 31–40

Table 3 Summary of unique solution analysis results. Level

Error criterion

Class 2

10 10 10

3

10 10 10

3

10 10 10

3

Class 1B

Class 1A

4 5

4 5

4 5

Chemical use

Greenhouse gas emission (kg CO2-e/m3 of treated water)

Energy flow (kWh/m3 of treated water)

Nutrient recovery (t/ML of excess sludge)

Ferric salts (t/d)

Methanol (ML/d)

CH4

N2O

CO2

Consumption

Recovery

Nitrogen

Phosphorus

0.000 0.000 0.000

0.000 0.000 0.000

0.263 0.264 0.264

0.290 0.290 0.290

0.294 0.294 0.294

0.358 0.358 0.358

0.110 0.109 0.109

8.400 8.401 8.401

6.778 6.779 6.779

0.090 0.089 0.090

0.000 0.000 0.000

0.503 0.502 0.503

0.295 0.295 0.295

0.355 0.355 0.355

0.428 0.428 0.428

0.180 0.181 0.180

9.895 9.896 9.895

5.914 5.915 5.914

0.100 0.100 0.101

0.250 0.250 0.250

0.500 0.501 0.501

0.285 0.285 0.285

0.379 0.379 0.379

0.457 0.457 0.457

0.183 0.184 0.184

9.882 9.881 9.881

6.007 6.006 6.006

with abiotic depletion, global warming, ozone depletion, acidification and ecotoxicity, which are the main impact categories used in environmental impact assessment (Horne et al., 2009; Shaw et al.,

2011). To this end, chemical use, energy consumption and GHG emissions were employed herein to represent the overall environmental burden. The chemical use in this study consisted of the

Fig. 2. Inventories of chemical consumption at different treatment levels: (A) Class 2; (B) Class 1B; (C) Class 1A.

X. Wang et al. / Bioresource Technology 126 (2012) 31–40

35

Fig. 3. Energy consumption and bioenergy recovery at different treatment levels: (A) Class 2; (B) Class 1B; (C) Class 1A.

external carbon source and ferric trichloride for enhanced phosphorus removal. Additionally, GHG calculation performed herein considered the release of CH4, N2O and CO2 gasses. The current global warming potential (GWP) values of CO2, CH4 and N2O are 1, 25 and 298, respectively (IPCC, 2001). Upon evaluation of the GHG emissions, relative GWP contributions of CH4 and N2O were generally transferred to the carbon dioxide equivalent. For investigation of the energy consumption, the energy requirement for aeration, pumping for activated-sludge return and MLR, liquid mixing in the anoxic and anaerobic units, and heating for the sludge anaerobic digestion unit were considered. For determination of the resource recovery indices, it was assumed that methane is

recovered from the sludge anaerobic digester and combusted to generate energy, and that the dewatered digested sludge is further recycled on agricultural soils via composting. The N and P sludge content allows the equivalent N and P mineral fertilizer substitution. Detailed descriptions of the methods used to calculate each evaluation indicator are available elsewhere (Wang et al., 2012).

3. Unique solution analysis Unique solution analysis was also performed to ensure that the values obtained for each indicator used in this study were unique.

36

X. Wang et al. / Bioresource Technology 126 (2012) 31–40

Fig. 4. Relationship between sludge production and electricity consumption.

This analysis was conducted using three sets of error criteria, the selected error criterion of 10 3, as well as the more stringent criteria of 10 4 and 10 5. After simulations and comparisons, the relative differences among the results of the three sets of calculations were within 1%, indicating that the obtained solution was unique. Owing to length limitations, the results of the unique solution analysis were not all presented herein; nevertheless, the quantitative results of the uniqueness analysis of Scenario 1 for the various error criteria at the three treatment levels were relatively obvious (Table 3). Overall, the unique solution analysis showed a good level of uniqueness for all indicators. To reduce the iteration time, 10 3 was selected as the error criterion for the present study. 4. Results and discussion Figs. 2–6 show comparisons of selected inventory data for the six treatment alternatives under the three levels of the treatment standard. Assessment of the Scenarios employing the variously available mid-point and end-point life cycle impact assessment (LCIA) methodologies would better establish the relative environmental burdens caused by different process configurations and levels of treatment. However, this paper only presents the life cycle inventory results. Further full LCIA results and in-depth demonstrations concerning comprehensive configurations of WWTPs including all well known process systems will be given in a future study. In this paper, the essential life cycle inventory outcomes are presented to provide simple, but important information for decision-making when the implementation of an anaerobic/anoxic/ oxic system for a WWTP is proposed. 4.1. Chemical consumption Enhanced levels of wastewater treatment and nutrient elimination resulted in increased environmental loads in accordance with the consumption of synthetic chemicals, which primarily consisted of additional carbon sources (e.g., methanol in this work), and precipitating chemicals (e.g., ferric salts in this study) (see Fig. 2). The average chemical consumption of both methanol and ferric salts increased substantially from Class 2 (methanol: 0.09 ± 0.05 ML/d, no ferric salts dosing) to Class 1B (methanol: 0.12 ± 0.06 ML/d, ferric salts: 0.23 ± 0.12 t/d) and then again from Class 1B to 1A (methanol: 0.14 ± 0.05 ML/d, ferric salts: 0.56 ± 0.26 t/d). These findings

represented a negative environmental outcome; accordingly, the addition of chemicals during wastewater treatment should be minimized owing to the negative effects of this practice. Additionally, the counter ion of the salts occasionally remained in the treated effluent, resulting in increased salinity of the receiving waters. Moreover, the chemical precipitate accumulated in the sludge, leading to extra costs for treatment of the excess sludge. Furthermore, since the sludge content in a treatment system is limited to a maximum amount, large treatment systems are required to maintain the same amount of biological sludge. This results in the need for chemicals such as ferric salts and methanol, which require additional resources and energy to manufacture, as well as further resources and energy for transportation to the WWTP, which all lead to increased environmental burdens. There was a distinct increase in the chemical demand of the reversed A2/O type processes (Scenario 5 and 6) when compared with the conventional A2/O type processes (Scenario 1–4). This was primarily due to a certain portion of carbon sources being consumed in the anoxic process prior to entering the anaerobic zone. Specifically, the phosphorus-releasing process in the anaerobic zone would be disturbed owing to a lack of sufficient carbon sources in both Scenario 5 and 6. Thus, to supplement carbon sources to meet the enhanced phosphorus removal requirements, a large amount of chemicals are required for the reversed A2/O type processes. The transition from Scenario 5 to 6 was employed through a step-feed mode, which resulted in a portion of the carbon sources (30% of the influent) being introduced directly into the anaerobic zone in Scenario 6; however, this did not attenuate the additional chemical consumption. Fig. 2 shows that the increased chemical consumption for Scenario 4 was minimal among the conventional A2/O type processes (Scenario 1–4). As shown in Scenario 4, the nitrate-containing return sludge was first introduced into the anoxic zone for denitrification, after which the nitrate-free sludge and/or water mixture was partially recycled to the anaerobic zone; hence, the phosphorus-releasing process in the anaerobic zone could not be reduced owing to the direct competition for carbon sources between the heterotrophic denitrifiers and polyphosphate-accumulating bacteria, which resulted in a positive environmental effect. Additionally, the results suggest that there is likely one optimum balance between sophisticated treatments and chemical consumption that minimizes the combined environmental loads of eutrophication from effluent discharge and chemical consumption. Thus, when the implementation or optimization of an anaerobic/anoxic/oxic process is proposed, Scenario 4 would be an environmentally favorable option. 4.2. Energy consumption and recovery 4.2.1. Energy requirement During estimation of the energy requirement, aeration in the oxic zone, liquid mixing in the anoxic and anaerobic zone, pumping for RAS and MLR, and mixing and heating for the anaerobic sludge digestion unit were considered. Other sources of energy consumption were calculated to comprise 20% of the total energy consumption. The evaluation outcomes (see Fig. 3) clearly demonstrated that the transition from Class 2 to 1A required substantially increased importation of electrical energy (Class 2: 0.377 ± 0.023 kWh/m3 wastewater treated, Class 1B: 0.490 ± 0.089 kWh/m3 wastewater treated, Class 1A: 0.510 ± 0.065 kWh/m3 wastewater treated). These findings represent a negative environmental outcome from a fossil fuels consumption point of view. Herein, the largest proportion (46.9–58.8% of the total) of the power demand was for the aeration unit in the water line. It was necessary to determine how to increase the oxygen-supply efficiency, as well as to

X. Wang et al. / Bioresource Technology 126 (2012) 31–40

37

Fig. 5. Greenhouse gas (GHG) emissions by gas type (CH4, N2O, and CO2) and total at different treatment levels: (A) Class 2; (B) Class 1B; (C) Class 1A.

optimize the oxygen-transfer as a precursor for energy conservation in WWTP operations. During traditional wastewater treatment, the need to achieve stringent nutrient discharge and protect the receiving environments from eutrophication leads to increased excess sludge generation that coincides with chemical phosphorus removal. As shown in Fig. 4, there was a positive relationship between energy consumption and excess sludge production. The power consumption for sludge heating in the anaerobic process increased substantially as the waste generation increased, indicating that minimal excess sludge production would reduce the electricity demands. However, this contradictory relationship cannot be avoided in conventional WWTPs, which are required to achieve higher nutrient removal from sewage. There was a distinct improvement in the energy requirements of both Scenario 5 and 6 when all alternatives were compared. This was primarily due to their poor phosphorus removal performance and subsequent increase in excess-sludge generation to achieve phosphorus elimination. This indicated that the additional phosphorus

removal was achieved via not only increased chemical dosing, but also improved operational power input. Furthermore, these results clearly indicated that there was an increase in the power requirement of Scenario 3 when compared with the conventional (Scenario 1) and other two modified processes (Scenario 2 and 4). Herein, a multipoint recycle mode for return sludge was employed in which only a small percentage of sludge was returned to the anaerobic zone and the rest was introduced directly to the anoxic zone. These results demonstrated an environmental tradeoff between energy consumption and the decrease in nutrient levels, with additional nutrient removal being realized via increased power importation rather than chemical consumption in Scenario 3. 4.2.2. Energy recovery and net energy consumption As shown in Fig. 3, there was a clear positive outcome between improved treatment levels and bioenergy recovery (Class 2: 0.130 ± 0.027 kWh/m3 wastewater treated; Class 1B: 0.215 ±

38

X. Wang et al. / Bioresource Technology 126 (2012) 31–40

the energy balance will be inclined to the positive side and it may be possible to export a certain amount of energy. Accordingly, an attempt to develop such a system will be made in a future study. From a net energy consumption perspective, it was difficult to determine which combination of anaerobic/anoxic/oxic systems is always clearly optimal under varying treatment standards. Nevertheless, the optimal bioenergy recovery is likely to be obtained when the combined environmental burdens of excess sludge production and gross energy importation are minimized. Furthermore, if only energy conservation is considered when selecting the optimal alternative, Scenario 2 would provide the optimal results. 4.3. Greenhouse gas emissions In Fig. 5, GHG emissions are presented by gas types (CH4, N2O, and CO2) and in total. These emissions do not involve the embodied GHG emissions associated with the WWTP construction phase and chemical consumption. It is well known that methane emissions are primarily linked with effluent discharge, process generation, and direct production from anaerobic sludge digestion in WWTPs. However, most of the organic load is aerobically degraded to CO2, which is considered GHG-neutral according to the Intergovernmental Panel on Climate Change (IPCC) accounting rule. In the present study, the majority of methane generated anaerobically in the digester was collected for beneficial bioenergy production. As shown in Fig. 5, wastewater treatment with anaerobic sludge digestion and methane recovery significantly reduced the contribution of methane to the GHG emissions to 1.12–1.41% of the total. These findings provided a positive outcome, and a win–win situation for energy recovery as well as the mitigation of climate change. The process model predicted small emissions of methane and hydrogen from the secondary treatment process, primarily as a result of being stripped from solution in the highly turbulent aeration systems. However, when compared with other parts of the methane contribution, this small proportion could be neglected. There is still a great deal of uncertainty associated with the quantification of nitrous oxide emissions during biological nutrient removal processes. However, the results of the present study revealed that processes with enhanced treatments (especially greater levels of nitrogen removal) have lower nitrous oxide emission factors (Class 1A: 0.264 ± 0.033 kg CO2-e/m3 wastewater treated; Class 1B: 0.269 ± 0.028 kg CO2-e/m3 wastewater treated) than processes that achieve intermediate levels of nitrogen removal (Class 2: 0.279 ± 0.055 kg CO2-e/m3 wastewater treated). These findings are supported by other recent evidence (Foley et al., 2010b); however, based on the results shown in Fig. 5, further estimation of these emissions is required. Direct carbon dioxide emissions from power depletion showed a similar pattern as the net energy consumption. The improved treatment levels were associated with a clear increase in direct carbon dioxide emissions (Class 2: 0.310 ± 0.020 kg CO2/m3 wastewater treated; Class 1B: 0.409 ± 0.078 kg CO2/m3 wastewater treated; Class 1A: 0.426 ± 0.057 kg CO2/m3 wastewater treated). These findings indicate that there are negative environmental effects associated with the implementation of sophisticated treatment standards.

Fig. 6. Nutrients in the excess sludge for recovery potential analysis at different treatment levels: (A) Class 2; (B) Class 1B; (C) Class 1A.

0.040 kWh/m3 wastewater treated; Class 1A: 0.217 ± 0.059 kWh/ m3 wastewater treated). Moreover, although the high yield of excess sludge in Scenario 5 and 6 led to an increased energy requirement, this negative situation appeared to have been offset by the higher energy recovery from biogas generated during anaerobic sludge digestion, which represented a neutral environmental outcome from an energy balance perspective. These findings indicate that if excess sludge production can be stimulated while reducing the power requirement via optimization of the related impact factors,

Table 4 Brief results of data quality assessmenta. Category

Score a

Chemical use

1.7

Energy consumption

GHG emissions

Resource recovery

Aeration

Pumping

Mixing

Heating

Others

CH4

N2O

CO2

Energy

Compost

1.8

1.8

1.8

1.8

2.8

1.8

2.7

1.8

1.8

1.7

Best quality = 1, worst quality = 5; a detailed explanation of the data quality assessment method is available elsewhere Wang et al. (2012).

X. Wang et al. / Bioresource Technology 126 (2012) 31–40

The total GHG emissions generally increased as the treatment standards increased (Class 2: 0.970 ± 0.219 kg CO2-e/m3 wastewater treated; Class 1B: 1.298 ± 0.171 kg CO2-e/m3 wastewater treated; Class 1A: 1.307 ± 0.269 kg CO2-e/m3 wastewater treated). From a GHG emissions perspective, Scenario 3 appeared to be the most favorable option. The results also indicated that minimal GHG emissions are produced by anaerobic sludge digestion and energy recovery from biogas combustion. 4.4. Nutrient recovery potential in excess sludge The primary objective of this study was to extend the system to include the environmental impacts of nutrient recovery to agricultural application in excess sludge. Herein, the nutrient contents, including nitrogen and phosphorus, in excess sludge were examined. As shown in Fig. 6, there was no distinct increase in the nutrient recovery potentials with increased levels of treatment. This is likely because the traditional WWTPs were not proposed or designed for nutrient recovery from excess sludge. However, this initial estimation highlighted the potential benefit of WWTPs for nutrient recovery, recycling and reuse, rather than elimination simply for the safety of the receiving aquatic surroundings. The idea of nutrient recovery from sewage via excess sludge was generated from the search for a replacement for synthetic fertilizers for agricultural practice. However, further evaluation and environmental impact assessment is necessary to determine if the comprehensive impacts of excess sludge applied for agricultural purposes are beneficial or more harmful than those of the existing displaced synthetic fertilizers. Nevertheless, these preliminary outcomes clearly demonstrate that the potential for nutrient recovery from excess sludge and the generation pathways and quantification of fertilizers (e.g., via struvite) require further study. 4.5. Environmental trade-off of wastewater treatment alternatives Figs. 2–5 show enhanced levels of resource consumption (chemical resources and energy resources) and environmental emissions (GHG and sludge generation). The negative outcomes depicted in these figures are a result of integrated global and local stress toward stringent effluent treatment standards for WWTPs. The trade-off between sophisticated treatments for optimized local receiving water and the environmental cost of higher chemical and fossil fuels consumption for operation was considered. The only positive trade-off with improved levels of treatment was observed when bioenergy recovery was employed. To estimate the nutrient recovery potential, the possible nutrient sources recovery from wastewater via excess sludge was estimated (Fig. 6); however, this relationship requires further assessment and quantification. To date, there has been insufficient evidence to consider the environmental trade-offs when planning anaerobic/anoxic/oxic process systems. The results presented herein provide preliminary, but essential information for identification of the basic trade-offs of various combinations of anaerobic/anoxic/oxic processes. Specifically, from a minimal chemical consumption perspective, Scenario 4 will be the best choice; however, from an energy conservation standpoint, Scenario 2 provides the optimum results, while Scenario 3 is best for mitigation of GHG emissions. In the next study, further estimation of these negative and positive environmental trade-offs will be demonstrated using the comprehensive life cycle assessment method. 5. Implication of the outcomes This work was highly associated with the data involved and the outcomes are therefore largely dependent on the data quality. The

39

quality of each data point used in this study was analyzed in our previous study (Wang et al., 2012) and is summarized briefly in Table 4. The main uncertainty introduced in the evaluation is associated with the use of assumed accounting factors for N2O emissions. In this case, N2O was assumed to be released by the biological wastewater treatment process, effluent discharge to the watercourse, and direct volatilization from excess sludge, which are viewed as the main emission pathways of N2O in WWTPs. Because of the lack of a well-proven N2O prediction model, some representative factors were summarized from the scarce literature available pertaining to N2O quantification; accordingly, less confidence can be had in those numbers. To this end, further efforts should be concentrated on incorporation of the most accurate method for N2O quantification once the relevant prediction model becomes available. Additionally, there have been few detailed studies conducted to quantify other sources of energy consumption in WWTPs, and previously obtained statistical data were used in this study, for which there is less confidence. As mentioned in our previous work, even if chemical use, energy consumption and GHG emissions are successfully used to rapidly analyze the overall adverse environmental impact with benefits of time and effort savings, this work is not a ‘‘one size fits all’’ endeavor. For example, if there were further considerations of the quantification of a certain category, such as eutrophication, acidification, or ecotoxicity, incorporation of LCA may be more desirable. 6. Conclusions This study employed a process modeling approach to demonstrate a comprehensive environmental performance assessment of six various combinations of anaerobic/anoxic/oxic processes under different treatment standards in China from a life cycle perspective. Furthermore, optimal combinations were identified under different perspectives. Based on the analysis outcomes, sophisticated treatments designed to prevent eutrophication of the local receiving water environment result in higher chemical and fossil energy consumption during operation, and higher GHG emissions. However, bioenergy recovery from biogas combustion resulted in reduced energy consumption and GHG emissions. Acknowledgements This study was supported by the National Science Foundation of China (NSFC) under Project No. 51138009 (Key Project) and Project No. 50921064 (National Creative Research Group), and Shanghai Tongji Gao Tingyao Environmental Science & Technology Development Foundation (STGEF). References Dennison, F.J., Azapagic, A., Clift, R., Colbourne, J.S., 1998. Assessing management options for wastewater treatment works in the context of life cycle assessment. Water Sci. Technol. 38, 23–30. Emmerson, R.H.C., Morse, G.K., Lester, J.N., Edge, D.R., 1995. The life-cycle analysis of small-scale sewage treatment processes. J. Chartered Inst. Water Environ. Manage. 9, 317–325. Envirosim, 2007. BioWin Process Simulator. Envirosim Associates Ltd. Foley, J., de Haas, D., Hartley, K., Lant, P., 2010a. Comprehensive life cycle inventories of alternative wastewater treatment systems. Water Res. 44, 1654–1666. Foley, J., de Haas, D., Yuan, Z.G., Lant, P., 2010b. Nitrous oxide generation in fullscale biological nutrient removal wastewater treatment plants. Water Res. 44, 831–844. Gaterell, M.R., Griffin, P., Lester, J.N., 2005. Evaluation of environmental burdens associated with sewage treatment processes using life cycle assessment techniques. Environ. Technol. 26, 231–249. Guerrero, J., Guisasola, A., Vilanova, R., Baeza, J.A., 2011. Improving the performance of a WWTP control system by model-based setpoint optimization. Environ. Model. Softw. 26, 492–497.

40

X. Wang et al. / Bioresource Technology 126 (2012) 31–40

Guinee, J.B., 2002. Handbook on Life Cycle Assessment: Operational Guide to the ISO Standards. Springer. Horne, R., Grant, T., Verghese, K., 2009. Life Cycle Assessment: Principles, Practice and Prospects. CSIRO Publishing, Victoria, Australia. Hospido, A., Moreira, M.T., Fernandez-Couto, M., Feijoo, G., 2004. Environmental performance of a municipal wastewater treatment plant. Int. J. Life Cycle Assess. 9, 261–271. IPCC, 2001. Climate Change 2001: The Scientific Basis. Cambridge University Press, Cambridge, UK. Kang, X.S., Liu, C.Q., Zhang, B., Bi, X.J., Zhang, F., Cheng, L.H., 2011. Application of reversed A(2)/O process on removing nitrogen and phosphorus from municipal wastewater in China. Water Sci. Technol. 63, 2138–2142. Lassaux, S., Renzoni, R., Germain, A., 2007. Life cycle assessment of water from the pumping station to the wastewater treatment plant. Int. J. Life Cycle Assess. 12, 118–126. Liu, J., Yang, M., Qi, R., An, W., Zhou, J., 2008. Comparative study of protozoan communities in full-scale MWTPs in Beijing related to treatment processes. Water Res. 42, 1907–1918. Lundin, M., Bengtsson, M., Molander, S., 2000. Life cycle assessment of wastewater systems: influence of system boundaries and scale on calculated environmental loads. Environ. Sci. Technol. 34, 180–186. Machado, V.C., Gabriel, D., Lafuente, J., Baeza, J.A., 2009. Cost and effluent quality controllers design based on the relative gain array for a nutrient removal WWTP. Water Res. 43, 5129–5141. MEP, 2002. Discharge Standard of Pollutants for Municiipal Wastewater Treatment Plant (GB18918-2002). China Environment Press, Beijing. Mulkerrins, D., Jordan, C., McMahon, S., Colleran, E., 2000. Evaluation of the parameters affecting nitrogen and phosphorus removal in anaerobic/anoxic/ oxic (A/A/O) biological nutrient removal systems. J. Chem. Technol. Biotechnol. 75, 261–268. Pasqualino, J.C., Meneses, M., Abella, M., Castells, F., 2009. LCA as a decision support tool for the environmental improvement of the operation of a municipal wastewater treatment plant. Environ. Sci. Technol. 43, 3300–3307.

Rodriguez-Garcia, G., Molinos-Senante, M., Hospido, A., Hernandez-Sancho, F., Moreira, M.T., Feijoo, G., 2011. Environmental and economic profile of six typologies of wastewater treatment plants. Water Res. 45, 5997– 6010. Shaw, A., Kadava, A., Tarallo, S. 2011. Refinement of Life Cycle Assessment (LCA) Methods for Water and Wastewater Treatment Plant Design. In: 8th IWA Leading-Edge Conference on Water and Wastewater Technologies, IWA publisher, Amsterdam. Tillman, A.M., Lundstrom, H., 1998. Life cycle assessment of municipal waste water systems. Int. J. Life Cycle Assess. 3, 145–157. Vidal, N., Poch, M., Marti, E., Rodriguez-Roda, I., 2002. Evaluation of the environmental implications to include structural changes in a wastewater treatment plant. J. Chem. Technol. Biotechnol. 77, 1206–1211. Wang, J.H., Zhang, J., Xie, H.J., Qi, P.Y., Ren, Y.G., Hu, Z., 2011. Methane emissions from a full-scale A/A/O wastewater treatment plant. Bioresour. Technol. 102, 5479–5485. Wang, X., Liu, J., Ren, N.-Q., Yu, H.-Q., Lee, D.-J., Guo, X., 2012. Assessment of multiple sustainability demands for wastewater treatment alternatives: a refined evaluation scheme and case study. Environ. Sci. Technol. 46, 5542– 5549. Wu, J., Meng, X., Liu, X., Liu, X., Zheng, Z., Xu, D., Sheng, G., Yu, S., 2010. Life cycle assessment of a wastewater treatment plant focused on material and energy flows. Environ. Manage. 46, 610–617. Zeng, W., Li, L., Yang, Y.Y., Wang, S.Y., Peng, Y.Z., 2010. Nitritation and denitritation of domestic wastewater using a continuous anaerobic-anoxic-aerobic (A(2)O) process at ambient temperatures. Bioresour. Technol. 101, 8074– 8082. Zhang, Z., Wilson, F., 2000. Life-cycle assessment of a sewage-treatment plant in South-East Asia. J. Chartered Inst. Water Environ. Manage. 14, 51–56. Zhou, Z., Wu, Z.C., Wang, Z.W., Tang, S.J., Gu, G.W., Wang, L.C., Wang, Y.J., Xin, Z.L., 2011. Simulation and performance evaluation of the anoxic/anaerobic/aerobic process for biological nutrient removal. Korean J. Chem. Eng. 28, 1233– 1240.