- Email: [email protected]
Simultaneous use of caustic and oxygen for efficient sulfide control in sewers
Science of the Total Environment 601–602 (2017) 776–783
Contents lists available at ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
Simultaneous use of caustic and oxygen for efficient sulfide control in sewers Hui-Wen Lin a, Yang Lu a, Ramon Ganigué b, Keshab R. Sharma a, Korneel Rabaey a,b, Zhiguo Yuan a, Ilje Pikaar a,c,⁎ a b c
The University of Queensland, Advanced Water Management Centre (AWMC), QLD 4072, Australia Center for Microbial Ecology and Technology (CMET), Ghent University, Coupure Links 653, 9000 Ghent, Belgium The University of Queensland, The School of Civil Engineering, QLD 4072, Australia
H I G H L I G H T S
G R A P H I C A L
A B S T R A C T
• Long-term efficient sulfide control in lab-scale sewer by NaOH-O2 combined dosing • The addition of O2 to caustic shock resulted in prolonged sulfide recovery period. • Synergistic effect of NaOH and O2 led to a reduction in CH4 production by 99% • Intermittent O2 addition reduced the dosing frequency for NaOH by 50%.
a r t i c l e
i n f o
Article history: Received 4 April 2017 Received in revised form 23 May 2017 Accepted 24 May 2017 Available online xxxx Editor: D. Barcelo Keywords: Sewer corrosion Sulfate reducing bacteria Sulfide abatement Dynamic modelling Caustic shock-loading Oxygen
a b s t r a c t Periodic caustic shock-loading is a commonly used method for sulfide control in sewers. Caustic shock-loading relies on the elevation of the sewage pH to ≥10.5 for several hours, thereby removing sewer pipe biofilms as well as deactivating SRB activity in the remaining biofilm. Although a widely used method, SRB activity is often not completely inhibited, and as such sulfide is still being generated. Here, we propose and experimentally demonstrate an innovative approach which combines caustic with oxygen, another commonly used method, as a dosing strategy for overcoming the drawbacks of caustic shock-loading. Six laboratory-scale rising main reactors were subjected to three dosing schemes over a period of three months, namely (i) simultaneous caustic and oxygen addition, (ii) caustic addition and (iii) no chemical addition. Our results showed that the combination of caustic and oxygen achieved efficient sulfide control, leading to a prolonged biofilm recovery period in between caustic shocks. In addition, methane emissions were reduced to a negligible level compared to caustic treatment only. To translate the findings to real-life application, the key parameters obtained during the long-term lab-scale experiments were subjected to extensive simulation studies using the SeweX model under a wide range of conditions commonly found in sewers. Overall, this study highlights the potential of periodic shock-loading and intermittent oxygen injection as combined dosing strategy for efficient sulfide control in sewers. © 2017 Elsevier B.V. All rights reserved.
⁎ Corresponding author at: The School of Civil Engineering, The University of Queensland, St. Lucia, QLD 4072, Australia. E-mail address: [email protected] (I. Pikaar).
http://dx.doi.org/10.1016/j.scitotenv.2017.05.225 0048-9697/© 2017 Elsevier B.V. All rights reserved.
H.-W. Lin et al. / Science of the Total Environment 601–602 (2017) 776–783
1. Introduction Hydrogen sulfide generation by sulfate reducing bacteria (SRB) in sewer pipes is a major issue in sewer management globally (Pikaar et al. 2014). The emission of hydrogen sulfide into the sewer atmosphere causes concrete corrosion as well as the release of obnoxious and toxic odours, posing a threat to sewer workers (Vollertsen et al. 2008). Current sulfide control strategies mainly involve chemical addition to either prevent hydrogen sulfide generation or mitigate its effects after its formation (Zhang et al. 2008). Two of the most commonly used chemicals are periodic caustic shock-loading and oxygen injection (Ganigue et al. 2011). Periodic caustic shock-loading, by raising the pH of the sewage to N10.5 for a short period of time (i.e. 2–6 h), relies on the removal of the biofilm from the sewer pipes as well as deactivating SRB activity in the remaining biofilms responsible for hydrogen sulfide formation (Gutierrez et al. 2014; O'Gorman et al. 2011; WERF 2007). While considered a cost-effective method, the sewer pipe biofilm is not completely removed after a dosing event. As the SRB activity in the remaining biofilm is not completely inhibited, sulfide is still being generated (Gutierrez et al. 2014). In addition, the re-growth of SRB can already commence within 1 to 3 days after a dosing event and may achieve complete SRB recovery within 3 to 14 days, depending on local conditions and sewer pipe design (Gutierrez et al. 2014; WERF 2007). In addition to deactivating SRB, it has been reported that pH elevation of sewage to above 8.6 deactivates methanogenic archaea (MA) (Gutierrez et al. 2009). Methane emissions from sewer systems to the atmosphere are also problematic as methane is a potent greenhouse gas that substantially contributes to the carbon footprint of water utilities (Guisasola et al. 2008a, Guisasola et al. 2009). Hence, inhibiting methane production in sewers is considered to be an additional benefit of using periodic caustic shock-loading as a sulfide control strategy. Oxygen injection suppresses the biological activity of SRB by maintaining oxic conditions as well as oxidizing sulfide either by chemical or biological sulfide oxidation (Gutierrez et al. 2008; Nielsen et al. 2003). While considered as one of the cheapest sulfide control methods, a major limitation of oxygen injection is its poor performance in some systems due to inappropriate choices of dosing locations and/or dosing rates (Ganigue et al. 2011). The diffusion of oxygen into the sewer biofilm largely determines the effectiveness of suppressing the SRB activity (Gutierrez et al. 2008). Under normal sewer conditions, dissolved oxygen does not fully penetrate into the biofilm, and as such the generation of sulfide continues in the anaerobic inner part of the biofilm (Gutierrez et al. 2008). The same holds true for MA, as these grow in the deeper layers of the biofilm and oxygen can only partially inhibit their activity (Ganigué and Yuan 2014). Moreover, the growth of heterotrophic biofilm caused by aerobic conditions results in an increase in oxygen uptake rate. Consequently, aerobic conditions cannot be maintained throughout the entire sewer pipe with sulfide production still occurring (Gutierrez et al. 2008). Although periodic caustic shock-loading and oxygen injection have been widely used as individual sulfide (and methane) control strategies, simultaneous use of these two chemicals has to our best knowledge never been investigated. It may however allow for more efficient sulfide (and methane) control. Periodic caustic shock-loading (partly) removes the sewer pipe biofilm during each dosing event (Gutierrez et al. 2014; O'Gorman et al. 2011; WERF 2007). This results in a thinner biofilm which would allow oxygen to penetrate further into the biofilm, thereby contacting the remaining SRB in the deeper layer of the biofilm. In this way, a better suppression of SRB may be achieved, leading to a prolonged biofilm recovery period in between caustic shocks. Recently, the simultaneous electrochemical generation of caustic and oxygen from sewage for sulfide control was successfully demonstrated (Lin et al. 2015). Using a three-compartment electrochemical cell, sewage was continuously oxygenated in the anode compartment whereas the cathode was operated in batch mode thereby producing a
777
moderate strength caustic solution (Lin et al. 2015). It was hypothesized that in this way, caustic can be periodically applied to the sewer pipe to clean sewer biofilms while sewage is continuously oxygenated and as such could suppress the residual biofilms between caustic shocks (Lin et al. 2015). While the above-mentioned study highlighted the practical and economic feasibility of simultaneous production of caustic and oxygen for sulfide control in sewers, the potential synergistic effect of caustic and oxygen on sewer biofilm was not investigated. In this study, we therefore investigate the potential synergistic effect of caustic and oxygen addition as combined dosing strategy for sulfide control in sewers. Using six laboratory-scale sewer rising main reactors, we conducted long-term experiments using three dosing schemes, namely (i) simultaneous caustic and oxygen addition, (ii) caustic addition and (iii) no chemical addition. Subsequently, to translate the result to real-life applications, the key parameters obtained were subjected to an extensive set of dynamic simulations using the SeweX model under a wide range of conditions commonly found in sewers. 2. Material and methods 2.1. Laboratory sewer reactors Laboratory-scale rising main sewer reactors in three parallel lines were used, as depicted in Fig. 1. The three parallel lines were defined as the caustic and oxygen line (the first reactor RE1 and the second reactor RE2), the caustic line (the first reactor RE3 and the second reactor RE4) and the control line (the first reactor RC1 and the second reactor RC2). Each line comprised of two completely sealed reactors connected in series. The reactors were entirely shielded with aluminum foil to prevent the sewage and biofilms from light exposure. A continuously mixed 1 L schott bottle mimicking a wet-well was placed at the beginning of each line for the addition of caustic and oxygen and caustic, respectively. No chemical addition took place in the control line. A 20 L composite sample container was equipped at the end of each line for collecting 24-h composite effluent. The sewer reactors, each with a volume of 750 mL, an inner diameter of 8 cm and a height of 15 cm, were made of Perspex™ and designed to mimic anaerobic sewer rising mains, as described in detail by Gutierrez et al. (2011) Ten Plastic Kaldnes carriers with dimensions of 9 mm height and 7 mm diameter (Anox Kaldnes, Norway) were equipped on a stainless steel rod inside each reactor for analysis of the microbial composition. The biofilm surface area in each reactor was calculated at 0.05 m2, resulting the area to volume (A/V) ratio of 61.4 m2/m3. Each reactor lid was equipped with a 70 mL container filled with sewage to avoid any vacuum or entry of air during sewage displacement. All reactors were continuously stirred using magnetic stirrers (Heidolph MR 3000), creating turbulent conditions of 0.44 Pa shear stress, mimicking real sewer conditions as previously determined by Sun et al. (2015). Fresh sewage was collected from a local pumping station in Brisbane on a weekly basis and immediately stored at 4 °C to minimize biological transformation of the sewage. All lines were exposed to 18 uneven pumping events daily (i.e. 3 repeated cycles with each cycle consists of 6 uneven pumping events) with a volume of 750 mL sewage fed to each line during each pumping event (see supplementary material, Fig. S1). Sewage was heated up to 24 ± 3 °C using a water bath before being fed to reactors. The typical domestic sewage contained 10 to 25 mg/L sulfate-S, b 3 mg/L dissolved sulfide-S and negligible amount of sulfite and thiosulfate. Methane and dissolved oxygen (DO) concentrations were also negligible. The average pH of the sewage influent used in the experiment was 7.4 ± 0.4. 2.2. Operation conditions The experiments were divided into two phases: the baseline and the experimental phase. During the baseline phase, the system was operated for 9 months without the addition of caustic and oxygen in order to
778
H.-W. Lin et al. / Science of the Total Environment 601–602 (2017) 776–783
Fig. 1. (A) Simplified schematic overview of the experimental setup and (B) cross-sectional view of a sewer rising main reactor.
establish pseudo steady-state conditions and to develop a mature anaerobic biofilm on the reactor walls and carriers in all reactors. The experimental phase was subsequently started and assessed the response of biofilms in each line for 12 weeks. Different lines were exposed to different dosing regimes (see Table 1). As the focus of this study was to investigate the potential synergistic effect of caustic and oxygen on sewer biofilms rather than electrochemically producing caustic and oxygen as demonstrated previously (Lin et al. 2015), purchased caustic soda and aeration stone were employed (see Fig. 1). Caustic shock-loading was manually applied to both caustic and oxygen line and caustic line on a weekly basis. A NaOH solution (12 M) was added to the wet-wells before sewage being fed to reactors for 6 successive pumping events on day 0, resulting in the elevation of the pH in reactors to a value of ~12 for 8 h. Subsequently, a recovery period of 6 days and 16 h (i.e. day 0 to day 6) was applied before the next caustic treatment. In the mimicking wet-well of the caustic and oxygen line, a DO concentration of ~ 6.5 mg/L in the sewage was obtained by means of air addition using an aeration stone for 11 min before sewage being fed into the reactors prior to each pumping event. 2.3. Reactor monitoring and sampling campaign Both reactor performance monitoring and batch tests were carried out during the baseline and experimental phase. During the baseline phase, determination of the reactor performance and the batch tests were carried out on a weekly basis to ensure stable and comparable sulfide (as well as methane) production rates in all lines. During the experimental phase, the reactor performance was measured from day 1 to day 5 after each caustic shock. Batch tests were performed on day 3 after each caustic addition. Two on-line S::CAN sensors (spectro::lyser, s::can Messtechnik GmbH) were used for the continuously monitoring of total dissolved sulfide concentration in the reactors (SutherlandStacey et al. 2008). The reactor performance was determined by monitoring the sulfide concentration throughout the pumping cycle without disturbance of the operational routine. For the batch tests, fresh sewage (without any caustic and oxygen addition) was pumped into each line
for 6 min to replace the sewage in the reactors followed by continuous monitoring of sulfide concentrations throughout a period of 1 h for the determination of sulfide production rate (SPR). Liquid samples were taken through the sampling ports at 15-minute intervals for the determination of methane production rate (MPR). In addition, liquid samples were taken at the end of pumping cycle 3 (HRT of 3 h) for sulfide measurement from reactor RE2, RE4 and RC2 at day 0 (before caustic addition), 24, 48, 72 and 168 h after caustic treatment, giving the highest sulfide concentrations of day 0 (before caustic addition), 1, 2, 3 and 7 (day 0 of the next caustic addition) over the course of two consecutive caustic additions. The methane concentrations of RE1, RE3 and RC1 were also determined at the end of pumping cycle 3 prior to the next caustic treatment in order to monitor the weekly recovery of methane levels. Additionally, 24-hour composite samples were collected for the measurement of volatile suspended solids (VSS) and total suspended solids (TSS). VSS and TSS samples were collected on the day before each caustic addition, as well as on day 0, 1 and 2 in order to determine the amount of biofilm being washed out after each shock loading event. DO contents and pH levels were continuously measured using an on-line DO sensor (Mettler Toledo, Switzerland) and pH sensors (Hanna Instrument, USA), respectively. 2.4. Microbial sampling and analyses Biofilm samples were collected from the carriers equipped in each reactor during the baseline phase as well as during the experimental phase in weeks 8 and 12. The samples in the experimental phase were taken 6 days after chemical dosing (i.e. one day before a next dosing event). Genomic DNA was extracted using FastDNA SPIN Kit for soil (MP Biomedicals, USA). Subsequently, extracted DNA samples were submitted to the Australian Centre for Ecogenomics (the University of Queensland) for 16S Amplicon sequencing by Illumina Miseq Platform using pyroLSSU926F (5′-AAACTYAAAKGAATTGACGG-3′) and pyroLSSU1392R (5′-ACGGGCGGTGWGTRC-3′) primer set (Dove et al. 2013). Sequencing results were analysed according to Hülsen et al. (2016). Filtered Open taxonomy units (OTUs) table were rarefied by
Table 1 Experimental conditions for baseline and experimental phases.
Baseline phase Experimental phase
Line
Reactor
pH
NaOH exposure time (hour)
Dissolved oxygen concentration (mg/L)
All lines Caustic and oxygen
All reactors RE1 RE2 RE3 RE4 RC1 RC2
Not controlled ~12
Not controlled 8 h (weekly)
Anaerobic condition ~6.5 mg/L in sewage in the mimicking wet-well before being fed to reactors
~12
8 h (weekly)
Anaerobic condition
Not controlled
Not controlled
Anaerobic condition
Caustic Control
H.-W. Lin et al. / Science of the Total Environment 601–602 (2017) 776–783
function “rarefy_even_depth” of package phyloseq in R (RCoreTeam) and principle component analysis (PCA) were performed by ampvis (Albertsen et al. 2015). A quantitative polymerase chain reaction (qPCR) analysis was performed to quantify microbial load, as described in Vanwonterghem et al. (2014). 2.5. Modelling of sulfide and methane inhibition A set of dynamic simulations were conducted using the SeweX model (Sharma et al. 2013) to assess the performance of each sulfide control strategy (i.e. no dosing, oxygen injection, caustic shock, and caustic shock and oxygen injection) under different sewer conditions. The addition of oxygen was conducted at the beginning of the sewer pipe in a continuous fashion, reaching initial DO levels of 6.5 mg O2/L in sewage while caustic shock-loading to pH 12 was assumed to have a periodicity of 3 days. Default model parameters were used as in Sharma et al. (2013), except for the sulfide production (surface) rate, methane production (surface) rate and oxygen uptake rate, which were adopted based on the experimental results obtained from the three different dosing regimes. Parameter values were calculated based on the average of all individual tests performed on day 3 after the caustic shock, and corrected to a temperature of 20 °C according to the van't Hoff-Arrhenius equation (Tchobanoglous et al. 2003). Sewage composition was assumed constant and defined based on the average concentrations recorded throughout the experiments. Table 2 shows the sewer conditions and dosing scenarios used in simulation studies. First, we simulated a rising main with a pipe length of 2500 m, pipe diameter of 0.25 m (A/V ratio of 16) and HRT of 3 h at 20 °C, after 3 days of a caustic shock. These parameters were chosen based on the industry survey of chemical dosing for sulfide control as reviewed by Ganigue et al. (2011). Subsequently, three key sewer characteristics, namely, the A/V ratio, HRT and temperature, were changed individually to assess their impact on the effectiveness of sulfide control (see Table 2) by simulating the sulfide profiles along the rising main also 3 days after a caustic shock. 2.6. Analytical methods and calculations Dissolved sulfur species (i.e. sulfide, sulfate, sulfite and thiosulfate) were measured using Ion Chromatography (IC) equipped with a Dionex 2010i system according to Keller-Lehmann et al. (2006). Dissolved methane measurements were determined using the methodology described by Alberto et al. (2000) and Guisasola et al. (2008). 5 mL of
779
sewage samples were filtered (0.22 μm) and injected into 12 mL evacuated Exetainer® tubes (Labco, UK). The tubes allowed equilibrium of gas and liquid phases overnight. Subsequently, the gaseous methane in the headspace of the tube was measured by means of gas chromatography (GC) equipped with a flame ionization detector. The dissolved methane concentration in the sewage sample was calculated using Henry's law (Alberto et al. 2000). SPR and MPR were calculated using linear regression. Oxygen uptake rate was calculated using linear regression based on the real-time DO concentrations measured in RE1. TSS and VSS were analysed using Standard Methods 2540D and 2540E, respectively (APHA 1995). 3. Results and discussion 3.1. Long-term reactor performance Fig. 2 shows the typical performance profiles observed on day 1 after a caustic treatment event. During the course of the experimental period, both caustic and oxygen line and caustic line showed a significant reduction in sulfide levels after caustic treatment. Overall, the caustic and oxygen line showed no sulfide accumulation compared to the caustic line, indicating either oxidation of sulfide took place or the activity of SRB was fully inhibited. The latter was further confirmed by IC analyses which revealed that no sulfate reduction was observed in RE1 (Fig. S2), suggesting intermittent oxygen addition after caustic treatment is likely to inhibit the activity of the remaining SRB. Contrarily, sulfide production activity was still observed in the caustic line after caustic treatment (Fig. 2), showing that caustic treatment did not control the activity of SRB entirely. A complete overview of the weekly results after one day of caustic treatment throughout the experimental phase can be found in the supplementary material (Fig. S3–5). In addition, the typical profiles of pH changes on day 0 in both caustic and oxygen line and caustic line are illustrated in Fig. S6. The daily recovery of sulfide levels in the caustic and oxygen line and the caustic line are shown in Fig. 3. On day 5, the maximum relative sulfide concentrations (the sulfide concentration of control line is considered as 100%) in RE1 and RE2 were 37.0% and 44.5%, respectively. On the contrary, the results showed the maximum relative sulfide concentrations of 98.3% in RE3 and 102.7% in RE4, indicating that a full recovery of sulfide levels in the caustic line was reached on day 5. Although the sulfide level recovered gradually in the caustic and oxygen line, it recovered much slower compared to the caustic line and did not return to the pre-shock level (i.e. control sulfide level). The latter was further
Table 2 Sewer conditions and dosing scenarios used in simulation studies. Pipe length (m)
Pipe diameter (m)
A/V ratio
HRT (hours)
Temperature (°C)
2500
0.25
16
3
20
Modified cases A/V ratio
2500
0.25
3
20
HRT
2500
0.25
8 16 26.66 16
20
Temperature
2500
0.25
16
1.5 3 5 8 3
Proposed standard rising main
10 15 20 25
Dosing scenarios
pH DO (mg O2/L)
Caustic and Oxygen
Caustic
Oxygen
Control
12 6.5
12 –
– 6.5
– –
780
H.-W. Lin et al. / Science of the Total Environment 601–602 (2017) 776–783
Fig. 2. Typical sulfide profiles of the first (A) and second reactors (B) one day after caustic treatment for the caustic and oxygen line, caustic line and control line. No chemical addition was performed in the control line.
supported by the highest sulfide levels of the day over the course of two consecutive caustic treatments (Fig. 4). The shaded areas represent the time that caustic was added to the reactor. On average, the caustic and oxygen line showed lower sulfide levels compared to the caustic line. The relative sulfide level of caustic and oxygen line in the end of the recovery period (i.e. day 7 and day 14) was 58 ± 9% (n = 2; 2 weeks) in comparison with the control line. A full recovery of the sulfide production in the caustic line was observed approximately after day 3, indicating a fully recovery of sulfate-reducing activity (in comparison with the control line) in the caustic line between day 4 and day 6. Daily solid discharge and relative solid discharge from the reactors before, during (Day 0) and after the caustic treatment (Day 1 and 2) can be found in Table S1 and Table S2, respectively. On average, a
significant amount of solid was washed out from the caustic and oxygen line and caustic line on day 0 (during caustic treatment). The VSS/TSS ratios observed during caustic treatment were lower than the ratios observed on other days. This suggested that a fraction of the TSS adsorbed to the biofilms were released through the detachment of biofilm into the sewage. No significant difference in solid discharge between caustic line and caustic and oxygen line was observed, indicating the aggravation of the biofilm detachment caused by the oxygen addition did not take place. Solid discharges then returned to the stable level on day 1 and 2 after caustic treatment, giving similar values in comparison with the control line. The average VSS/TSS ratio in the sewage influent during 12 weeks was 93 ± 2%, which was slightly higher than the typical VSS/ TSS ratios in municipal wastewater (i.e. 40–90%) (Henze et al., 2008).
Fig. 3. The recovery of sulfide levels in the caustic and oxygen line (RE1 and RE2) and the caustic line (RE3 and RE4) after caustic treatment. The monitoring for caustic and oxygen line and caustic line was carried out daily for one complete pumping cycle in week 3 and week 4, respectively.
H.-W. Lin et al. / Science of the Total Environment 601–602 (2017) 776–783
Fig. 4. Measurements of highest dissolved sulfide level at the end of the longest pumping event (HRT: 3 h) over the course of two consecutive caustic additions. The shaded areas represent the additions of caustic.
781
Fig. 6. Average oxygen uptake rate (OUR) measured at 152 to 160 h after caustic addition over the course of 12 weeks in comparison with the OUR reported in Gutierrez et al. (2008).
However, the VSS/TSS content in sewage is not expected to have impacted the experimental results. 3.2. Inhibition of sulfide producing activity Batch activity test were conducted on a weekly basis to determine the sulfide production rate (SPR) in the absence of oxygen on day 3 after caustic treatment. Fig. 5 shows that on average, the relative SPRs of the caustic and oxygen line in comparison to the control line are 29.6 ± 13.1% for RE1 and 24.2 ± 8.5% for RE2 while the caustic line showed higher SPRs of 38.6.6 ± 11.3% for RE3 and 39.6 ± 16.6% for RE4. The result suggests that the additional oxygen supply after caustic treatment effectively further lowered the SPR by ≈ 20–40% on day 3 compared with the stand-alone caustic treatment. The latter is inconsistent with our previous study which has found that short periods of oxygen exposure do not affect the sulfide producing activity of SRB (Gutierrez et al. 2008). It further reveals that intermittent oxygen could potentially lead to a more long-lasting suppression effect on the remaining SRB after caustic treatment. Although certain SRB strains are able to survive to oxygen exposure by defence mechanisms such as aggregation and even oxygen reduction or reactive oxygen species (ROS) detoxification, oxygen tolerance of SRB is species-dependent (Cypionka et al. 1985; Dolla et al. 2006; Sass et al. 1998). Oxygen can have toxic effect on some SRB without oxygen tolerance. In addition, the presence of ROS can damage cell components and could be
responsible for oxygen toxicity (Baumgartner et al. 2006; Dolla et al. 2006). These might contribute to the better SRB suppression observed in the caustic and oxygen line. Also, it is noteworthy that the relative SPRs of treated reactors (i.e. RE1, RE2, RE3 and RE4) slightly increased over the course of the experiment in comparison to the control line. This was likely due to the increase in the sewage temperature over time from 21 ± 1 °C in weeks 1–4 to 26 ± 1 °C in weeks 5–12. This is supported by the fact that biofilm removal was observed in RE1 and RE2 after each caustic shock (see supplementary material, Fig. S7).
3.3. Oxygen uptake rate The typical profiles of DO concentrations in RE1 and RE2 on day 0, 1, 2 and 6 (i.e. 152 to 160 h after caustic addition) are given in Figs. S8 and S9, respectively. Overall, the oxygen uptake rates (OUR) of RE1 and RE2 were highest on day 0 and decreased gradually. Previously, it was found that the repeated exposure of biofilms to oxygen significantly increased the OUR with a four to five-fold increase (Gutierrez et al. 2008). The latter was found to be related to the growth of heterotrophic bacteria. Fig. 6 shows an overview of OUR in RE1 measured at 152 to 160 h after caustic addition over the course of 12 weeks in comparison with the OUR reported in Gutierrez et al. (2008). Here, the OUR remained constant over
Fig. 5. Sulfide production rates (SPR) obtained from batch tests on day 3 in absence of oxygen. Experimental temperatures ranged from 21 ± 1 °C (baseline and week 1 to week 4) to 26 ± 1 °C (week 5 to week 12) over the course of experimental phase. (A) and (B) represent the SPR of first reactors and second reactors, respectively.
782
H.-W. Lin et al. / Science of the Total Environment 601–602 (2017) 776–783
RE2, 8% in RE3 and 12% in RE4 (the total community in each reactor in baseline phase is considered as 100%). Furthermore, the relative abundance of methanogens in the biofilm in week 8 was greatly reduced to b1% and 5% in RE1 and RE2, respectively, but was still present in the caustic line (with 10–15% Methanobacterium in RE3 and 4). Additionally, changes in the biofilm colour and coverage of the walls of the laboratory rising main reactors were visually observed in the caustic and oxygen line (see Fig. S7) while the biofilm colour in the caustic line (see Fig. S11) remained similar to the control line (see Fig. S12). This gradual biofilm colour change in the caustic and oxygen line suggests that the addition of oxygen after caustic shock resulted in a microbial community change. Indeed, a shift from Methanosaeta and Bacteroidales to Acinetobacter, family Porphyomonadaceae and Peptostreptococcaceae was observed, as shown in the detailed principle component analysis of microbial community in all reactors (see Fig. S13). Further research such as metaproteomic profiling is warranted to fundamentally understand the behaviour of SRB, methanogenic archaea or other bacteria under caustic stress with additional oxygen supply.
3.5. Simulation studies
Fig. 7. Total dissolved sulfide, methane and dissolved oxygen profiles. In all three cases, a sewer pipe with a length of 2500 m, A/V ratio of 16, HRT of 3 h and temperature of 20 °C was used.
Fig. 7 depicts the total dissolved sulfide, methane and oxygen profiles along a 2.5 km rising main for 4 different dosing scenarios, 3 days after a caustic shock. Results showed that the combined caustic and oxygen dosing substantially improved the reduction in total dissolved sulfide and methane levels in comparison to the caustic shock, with further reductions in total dissolved sulfide and methane level of 33% and 67%, respectively. Compared to the control line, oxygen dosing only reduced the total dissolved sulfide level by 6% and had little effect on methane control (i.e. 7% reduction). In addition, a number of simulations under various sewer conditions (i.e. pipe diameters, temperatures and HRT) were carried out to further assess the treatment effectiveness in terms of sulfide control, suggesting caustic and oxygen dosing is a more efficient sulfide control strategy based on the results of simulation. More detailed results about these simulations can be found in the supplementary material (Figs. S14–16). 3.6. Implications for practice
12 weeks, suggesting that the development of a heterotrophic biofilm did not take place.
3.4. Inhibition of methane-producing activity Prior to the next caustic dosing (n = 3; day 6 of week 6, 7 and 8), a negligible amount of methane (i.e. 0.2 ± 0.1 mg/L, equal to 0.6 ± 0.3 mg-COD/L) was detected in the caustic and oxygen line (RE1) while 1.4 ± 0.1 mg/L (i.e. 5.7 ± 0.6 mg-COD/L) was detected in the caustic line (RE3). The discharge of methane on day 6 in both caustic and oxygen line and caustic line was reduced notably. Compared these values to the methane level of 16.1 ± 1.6 mg/L (i.e. 64.4 ± 6.4 mg-COD/L) in the control line RC1, the relative methane discharge levels were reduced significantly to 0.9 ± 0.5% for caustic and oxygen line and 8.9 ± 1.4% for caustic line. Additionally, the results of the batch activity tests on day 3 show that both caustic and oxygen line and caustic line had significant lower MPRs compared with the control line. The caustic and oxygen line had lower MPRs on day 3 compared to the caustic line, accounting for 39% for the first reactor and 63% for the second reactor of the caustic line. In addition, the addition of oxygen reduced methane concentrations to a negligible level and slowed down the re-growth of methanogenic archaea compared to caustic treatment (see supplementary material, Fig. S10). The results of qPCR analysis further determined that the total community in week 12 were reduced to 26% in RE1, 8% in
Both long-term laboratory experiments and simulation results suggest that periodic caustic treatment combined with intermittent addition of oxygen substantially enhanced the effectiveness of sulfide control in comparison with caustic treatment alone. Our results show that in order to achieve the same efficiency in terms of sulfide control, the dosing frequency for caustic shock-loading is doubled compared to caustic shock and oxygen addition (e.g. weekly dosing for caustic and oxygen addition; 3.5 day dosing frequency for caustic treatment under the sewer conditions simulated). Equally important, in addition to sulfide control, also a better overall control of methane was achieved. In full-scale applications, this combined dosing method is especially suitable for smaller rising mains (e.g. ≤0.3 m) with a relative low flow rate (average dry weather flow b0.5 MLd−1), conditions known to be suitable for caustic dosing (Ganigue et al. 2011). Large size rising mains with large flows are less suitable due to the large caustic requirements. In practice, the required amounts of caustic and oxygen could also be produced on-site by using an electrochemical system, as previously demonstrated by Lin et al. (2015).
4. Conclusions In this study, we investigated an innovative sulfide control strategy which combines periodic caustic shock-loading with intermittent supply of oxygen for efficient sulfide control in sewers. The key findings of this work are:
H.-W. Lin et al. / Science of the Total Environment 601–602 (2017) 776–783
• Long-term experiments showed that the simultaneous use of periodic caustic shock-loading and intermittent oxygen supply during and after caustic treatment achieves more efficient sulfide control in comparison with caustic treatment alone, leading to a prolonged biofilm recovery period in between caustic shocks. • The addition of oxygen reduced methane emissions to a negligible level and slowed down the re-growth of methanogenic archaea in comparison with caustic shock-loading. The relative abundance of methanogens in the caustic and oxygen line was greatly reduced compared to the caustic line. • Based on the experimentally results obtained and simulation studies was conducted, it was demonstrated that the simultaneous use of periodic caustic shock-loading and intermittent oxygen supply is more effective in sulfide control compared to conventional caustic shockloading and reduced the dosing frequency for caustic treatment by 50%. Acknowledgements H.-W. L thanks the University of Queensland for scholarship support. R.G. gratefully acknowledges support from Ghent University BOF postdoctoral fellowship (BOF15/PDO/068). This work was supported by the Australian Research Council, District of Columbia Water and Sewer Authority, ACTEW Corporation Limited, The City of Gold Coast, Queensland Urban Utilities and Yarra Valley Water through ARC Linkage project LP120200238: “In-situ electrochemical generation of caustic and oxygen from sewage for emission control in sewers”. The authors acknowledge Dr. Beatrice Keller-Lehmann and Nathan Clayton for their helpful assistance with the chemical analyses. We also want to thank Dr. Nicola Angel from ACE for the assistance of 16S rRNA amplicon sequencing. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.scitotenv.2017.05.225. References Alberto, M.C.R., Arah, J.R.M., Neue, H.U., Wassmann, R., Lantin, R.S., Aduna, J.B., Bronson, K.F., 2000. A sampling technique for the determination of dissolved methane in soil solution. Chemosphere Global Change Sci. 2 (1), 57–63. Albertsen, M., Karst, S.M., Ziegler, A.S., Kirkegaard, R.H., Nielsen, P.H., 2015. Back to basics the influence of DNA extraction and primer choice on phylogenetic analysis of activated sludge communities. PLoS One 10 (7), e0132783, e0132783. APHA, 1995. Standard Methods for the Examination of Water and Wastewater. Baumgartner, L.K., Reid, R.P., Dupraz, C., Decho, A.W., Buckley, D., Spear, J., Przekop, K.M., Visscher, P.T., 2006. Sulfate reducing bacteria in microbial mats: changing paradigms, new discoveries. Sediment. Geol. 185 (3-4), 131–145. Cypionka, H., Widdel, F., Pfennig, N., 1985. Survival of sulfate-reducing bacteria after oxygen stress, and growth in sulfate-free oxygen-sulfide gradients. FEMS Microbiol. Lett. 31 (1), 39–45. Dolla, A., Fournier, M., Dermoun, Z., 2006. Oxygen defense in sulfate-reducing bacteria. J. Biotechnol. 126 (1), 87–100.
783
Dove, S.G., Kline, D.I., Pantos, O., Angly, F.E., Tyson, G.W., Hoegh-Guldberg, O., 2013. Future reef decalcification under a business-as-usual CO2 emission scenario. Proc. Natl. Acad. Sci. 110 (38), 15342–15347. Ganigué, R., Yuan, Z., 2014. Impact of oxygen injection on CH4 and N2O emissions from rising main sewers. J. Environ. Manag. 144, 279–285. Ganigue, R., Gutierrez, O., Rootsey, R., Yuan, Z., 2011. Chemical dosing for sulfide control in Australia: an industry survey. Water Res. 45 (19), 6564–6574. Guisasola, A., de Haas, D., Keller, J., Yuan, Z., 2008. Methane formation in sewer systems. Water Res. 42 (6-7), 1421–1430. Guisasola, A., Sharma, K.R., Keller, J., Yuan, Z., 2009. Development of a model for assessing methane formation in rising main sewers. Water Res. 43 (11), 2874–2884. Gutierrez, O., Mohanakrishnan, J., Sharma, K.R., Meyer, R.L., Keller, J., Yuan, Z., 2008. Evaluation of oxygen injection as a means of controlling sulfide production in a sewer system. Water Res. 42 (17), 4549–4561. Gutierrez, O., Park, D., Sharma, K.R., Yuan, Z., 2009. Effects of long-term pH elevation on the sulfate-reducing and methanogenic activities of anaerobic sewer biofilms. Water Res. 43 (9), 2549–2557. Gutierrez, O., Sudarjanto, G., Sharma, K.R., Keller, J., Yuan, Z., 2011. SCORe-CT: a new method for testing effectiveness of sulfide-control chemicals used in sewer systems. Water Sci. Technol. 64 (12), 2381–2388. Gutierrez, O., Sudarjanto, G., Ren, G., Ganigue, R., Jiang, G., Yuan, Z., 2014. Assessment of pH shock as a method for controlling sulfide and methane formation in pressure main sewer systems. Water Res. 48 (1), 569–578. Henze, M., van Loosdrecht, M.C., Ekama, G.A., Brdjanovic, D., 2008. Biological wastewater treatment. IWA publishing. Hülsen, T., Barry, E.M., Lu, Y., Puyol, D., Keller, J., Batstone, D.J., 2016. Domestic wastewater treatment with purple phototrophic bacteria using a novel continuous photo anaerobic membrane bioreactor. Water Res. 100, 486–495. Keller-Lehmann, B., Corrie, S., Ravn, R., Yuan, Z., Keller, J., 2006. Preservation and Simultaneous Analysis of Relevant Soluble Sulfur Species in Sewage Samples. p. 28 (Vienna, Austria). Lin, H.-W., Rabaey, K., Keller, J.r., Yuan, Z., Pikaar, I., 2015. Scaling-free electrochemical production of caustic and oxygen for sulfide control in sewers. Environ. Sci. Technol. 49 (19), 11395–11402. Nielsen, A.H., Vollertsen, J., Hvitved-Jacobsen, T., 2003. Determination of kinetics and stoichiometry of chemical sulfide oxidation in wastewater of sewer networks. Environ. Sci. Technol. 37 (17), 3853–3858. O'Gorman, J., Purssell, I., Iori, G., 2011. Caustic soda washing of a sewer pressure main. Water 38 (2), 83–87. Pikaar, I., Sharma, K.R., Hu, S., Gernjak, W., Keller, J., Yuan, Z., 2014. Reducing sewer corrosion through integrated urban water management. Science 345 (6198), 812–814. Sass, H., Berchtold, M., Branke, J., König, H., Cypionka, H., Babenzien, H.-D., 1998. Psychrotolerant sulfate-reducing bacteria from an oxic freshwater sediment description of Desulfovibrio cuneatus sp. nov. and Desulfovibrio litoralis sp. nov. Syst. Appl. Microbiol. 21 (2), 212–219. Sharma, K., Ganigue, R., Yuan, Z.G., 2013. pH dynamics in sewers and its modeling. Water Res. 47 (16), 6086–6096. Sun, J., Pikaar, I., Sharma, K.R., Keller, J., Yuan, Z., 2015. Feasibility of sulfide control in sewers by reuse of iron rich drinking water treatment sludge. Water Res. 71, 150–159. Sutherland-Stacey, L., Corrie, S., Neethling, A., Johnson, I., Gutierrez, O., Dexter, R., Yuan, Z., Keller, J., Hamilton, G., 2008. Continuous measurement of dissolved sulfide in sewer systems. Water Sci. Technol. 57 (3), 375–381. Tchobanoglous, G., Burton, F.L., Stensel, H.D., Metcalf and Eddy, 2003. Wastewater Engineering: Treatment and Reuse. McGraw-Hill, Boston. Vanwonterghem, I., Jensen, P.D., Dennis, P.G., Hugenholtz, P., Rabaey, K., Tyson, G.W., 2014. Deterministic processes guide long-term synchronised population dynamics in replicate anaerobic digesters. ISME J. 8 (10), 2015–2028. Vollertsen, J., Nielsen, A.H., Jensen, H.S., Wium-Andersen, T., Hvitved-Jacobsen, T., 2008. Corrosion of concrete sewers—the kinetics of hydrogen sulfide oxidation. Sci. Total Environ. 394 (1), 162–170. WERF, 2007. Minimization of Odors and Corrosion in Collection Systems Phase 1. Water Environment Research Foundation, London, UK. Zhang, L., De Schryver, P., De Gusseme, B., De Muynck, W., Boon, N., Verstraete, W., 2008. Chemical and biological technologies for hydrogen sulfide emission control in sewer systems: a review. Water Res. 42 (1-2), 1–12.
Contents lists available at ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
Simultaneous use of caustic and oxygen for efficient sulfide control in sewers Hui-Wen Lin a, Yang Lu a, Ramon Ganigué b, Keshab R. Sharma a, Korneel Rabaey a,b, Zhiguo Yuan a, Ilje Pikaar a,c,⁎ a b c
The University of Queensland, Advanced Water Management Centre (AWMC), QLD 4072, Australia Center for Microbial Ecology and Technology (CMET), Ghent University, Coupure Links 653, 9000 Ghent, Belgium The University of Queensland, The School of Civil Engineering, QLD 4072, Australia
H I G H L I G H T S
G R A P H I C A L
A B S T R A C T
• Long-term efficient sulfide control in lab-scale sewer by NaOH-O2 combined dosing • The addition of O2 to caustic shock resulted in prolonged sulfide recovery period. • Synergistic effect of NaOH and O2 led to a reduction in CH4 production by 99% • Intermittent O2 addition reduced the dosing frequency for NaOH by 50%.
a r t i c l e
i n f o
Article history: Received 4 April 2017 Received in revised form 23 May 2017 Accepted 24 May 2017 Available online xxxx Editor: D. Barcelo Keywords: Sewer corrosion Sulfate reducing bacteria Sulfide abatement Dynamic modelling Caustic shock-loading Oxygen
a b s t r a c t Periodic caustic shock-loading is a commonly used method for sulfide control in sewers. Caustic shock-loading relies on the elevation of the sewage pH to ≥10.5 for several hours, thereby removing sewer pipe biofilms as well as deactivating SRB activity in the remaining biofilm. Although a widely used method, SRB activity is often not completely inhibited, and as such sulfide is still being generated. Here, we propose and experimentally demonstrate an innovative approach which combines caustic with oxygen, another commonly used method, as a dosing strategy for overcoming the drawbacks of caustic shock-loading. Six laboratory-scale rising main reactors were subjected to three dosing schemes over a period of three months, namely (i) simultaneous caustic and oxygen addition, (ii) caustic addition and (iii) no chemical addition. Our results showed that the combination of caustic and oxygen achieved efficient sulfide control, leading to a prolonged biofilm recovery period in between caustic shocks. In addition, methane emissions were reduced to a negligible level compared to caustic treatment only. To translate the findings to real-life application, the key parameters obtained during the long-term lab-scale experiments were subjected to extensive simulation studies using the SeweX model under a wide range of conditions commonly found in sewers. Overall, this study highlights the potential of periodic shock-loading and intermittent oxygen injection as combined dosing strategy for efficient sulfide control in sewers. © 2017 Elsevier B.V. All rights reserved.
⁎ Corresponding author at: The School of Civil Engineering, The University of Queensland, St. Lucia, QLD 4072, Australia. E-mail address: [email protected] (I. Pikaar).
http://dx.doi.org/10.1016/j.scitotenv.2017.05.225 0048-9697/© 2017 Elsevier B.V. All rights reserved.
H.-W. Lin et al. / Science of the Total Environment 601–602 (2017) 776–783
1. Introduction Hydrogen sulfide generation by sulfate reducing bacteria (SRB) in sewer pipes is a major issue in sewer management globally (Pikaar et al. 2014). The emission of hydrogen sulfide into the sewer atmosphere causes concrete corrosion as well as the release of obnoxious and toxic odours, posing a threat to sewer workers (Vollertsen et al. 2008). Current sulfide control strategies mainly involve chemical addition to either prevent hydrogen sulfide generation or mitigate its effects after its formation (Zhang et al. 2008). Two of the most commonly used chemicals are periodic caustic shock-loading and oxygen injection (Ganigue et al. 2011). Periodic caustic shock-loading, by raising the pH of the sewage to N10.5 for a short period of time (i.e. 2–6 h), relies on the removal of the biofilm from the sewer pipes as well as deactivating SRB activity in the remaining biofilms responsible for hydrogen sulfide formation (Gutierrez et al. 2014; O'Gorman et al. 2011; WERF 2007). While considered a cost-effective method, the sewer pipe biofilm is not completely removed after a dosing event. As the SRB activity in the remaining biofilm is not completely inhibited, sulfide is still being generated (Gutierrez et al. 2014). In addition, the re-growth of SRB can already commence within 1 to 3 days after a dosing event and may achieve complete SRB recovery within 3 to 14 days, depending on local conditions and sewer pipe design (Gutierrez et al. 2014; WERF 2007). In addition to deactivating SRB, it has been reported that pH elevation of sewage to above 8.6 deactivates methanogenic archaea (MA) (Gutierrez et al. 2009). Methane emissions from sewer systems to the atmosphere are also problematic as methane is a potent greenhouse gas that substantially contributes to the carbon footprint of water utilities (Guisasola et al. 2008a, Guisasola et al. 2009). Hence, inhibiting methane production in sewers is considered to be an additional benefit of using periodic caustic shock-loading as a sulfide control strategy. Oxygen injection suppresses the biological activity of SRB by maintaining oxic conditions as well as oxidizing sulfide either by chemical or biological sulfide oxidation (Gutierrez et al. 2008; Nielsen et al. 2003). While considered as one of the cheapest sulfide control methods, a major limitation of oxygen injection is its poor performance in some systems due to inappropriate choices of dosing locations and/or dosing rates (Ganigue et al. 2011). The diffusion of oxygen into the sewer biofilm largely determines the effectiveness of suppressing the SRB activity (Gutierrez et al. 2008). Under normal sewer conditions, dissolved oxygen does not fully penetrate into the biofilm, and as such the generation of sulfide continues in the anaerobic inner part of the biofilm (Gutierrez et al. 2008). The same holds true for MA, as these grow in the deeper layers of the biofilm and oxygen can only partially inhibit their activity (Ganigué and Yuan 2014). Moreover, the growth of heterotrophic biofilm caused by aerobic conditions results in an increase in oxygen uptake rate. Consequently, aerobic conditions cannot be maintained throughout the entire sewer pipe with sulfide production still occurring (Gutierrez et al. 2008). Although periodic caustic shock-loading and oxygen injection have been widely used as individual sulfide (and methane) control strategies, simultaneous use of these two chemicals has to our best knowledge never been investigated. It may however allow for more efficient sulfide (and methane) control. Periodic caustic shock-loading (partly) removes the sewer pipe biofilm during each dosing event (Gutierrez et al. 2014; O'Gorman et al. 2011; WERF 2007). This results in a thinner biofilm which would allow oxygen to penetrate further into the biofilm, thereby contacting the remaining SRB in the deeper layer of the biofilm. In this way, a better suppression of SRB may be achieved, leading to a prolonged biofilm recovery period in between caustic shocks. Recently, the simultaneous electrochemical generation of caustic and oxygen from sewage for sulfide control was successfully demonstrated (Lin et al. 2015). Using a three-compartment electrochemical cell, sewage was continuously oxygenated in the anode compartment whereas the cathode was operated in batch mode thereby producing a
777
moderate strength caustic solution (Lin et al. 2015). It was hypothesized that in this way, caustic can be periodically applied to the sewer pipe to clean sewer biofilms while sewage is continuously oxygenated and as such could suppress the residual biofilms between caustic shocks (Lin et al. 2015). While the above-mentioned study highlighted the practical and economic feasibility of simultaneous production of caustic and oxygen for sulfide control in sewers, the potential synergistic effect of caustic and oxygen on sewer biofilm was not investigated. In this study, we therefore investigate the potential synergistic effect of caustic and oxygen addition as combined dosing strategy for sulfide control in sewers. Using six laboratory-scale sewer rising main reactors, we conducted long-term experiments using three dosing schemes, namely (i) simultaneous caustic and oxygen addition, (ii) caustic addition and (iii) no chemical addition. Subsequently, to translate the result to real-life applications, the key parameters obtained were subjected to an extensive set of dynamic simulations using the SeweX model under a wide range of conditions commonly found in sewers. 2. Material and methods 2.1. Laboratory sewer reactors Laboratory-scale rising main sewer reactors in three parallel lines were used, as depicted in Fig. 1. The three parallel lines were defined as the caustic and oxygen line (the first reactor RE1 and the second reactor RE2), the caustic line (the first reactor RE3 and the second reactor RE4) and the control line (the first reactor RC1 and the second reactor RC2). Each line comprised of two completely sealed reactors connected in series. The reactors were entirely shielded with aluminum foil to prevent the sewage and biofilms from light exposure. A continuously mixed 1 L schott bottle mimicking a wet-well was placed at the beginning of each line for the addition of caustic and oxygen and caustic, respectively. No chemical addition took place in the control line. A 20 L composite sample container was equipped at the end of each line for collecting 24-h composite effluent. The sewer reactors, each with a volume of 750 mL, an inner diameter of 8 cm and a height of 15 cm, were made of Perspex™ and designed to mimic anaerobic sewer rising mains, as described in detail by Gutierrez et al. (2011) Ten Plastic Kaldnes carriers with dimensions of 9 mm height and 7 mm diameter (Anox Kaldnes, Norway) were equipped on a stainless steel rod inside each reactor for analysis of the microbial composition. The biofilm surface area in each reactor was calculated at 0.05 m2, resulting the area to volume (A/V) ratio of 61.4 m2/m3. Each reactor lid was equipped with a 70 mL container filled with sewage to avoid any vacuum or entry of air during sewage displacement. All reactors were continuously stirred using magnetic stirrers (Heidolph MR 3000), creating turbulent conditions of 0.44 Pa shear stress, mimicking real sewer conditions as previously determined by Sun et al. (2015). Fresh sewage was collected from a local pumping station in Brisbane on a weekly basis and immediately stored at 4 °C to minimize biological transformation of the sewage. All lines were exposed to 18 uneven pumping events daily (i.e. 3 repeated cycles with each cycle consists of 6 uneven pumping events) with a volume of 750 mL sewage fed to each line during each pumping event (see supplementary material, Fig. S1). Sewage was heated up to 24 ± 3 °C using a water bath before being fed to reactors. The typical domestic sewage contained 10 to 25 mg/L sulfate-S, b 3 mg/L dissolved sulfide-S and negligible amount of sulfite and thiosulfate. Methane and dissolved oxygen (DO) concentrations were also negligible. The average pH of the sewage influent used in the experiment was 7.4 ± 0.4. 2.2. Operation conditions The experiments were divided into two phases: the baseline and the experimental phase. During the baseline phase, the system was operated for 9 months without the addition of caustic and oxygen in order to
778
H.-W. Lin et al. / Science of the Total Environment 601–602 (2017) 776–783
Fig. 1. (A) Simplified schematic overview of the experimental setup and (B) cross-sectional view of a sewer rising main reactor.
establish pseudo steady-state conditions and to develop a mature anaerobic biofilm on the reactor walls and carriers in all reactors. The experimental phase was subsequently started and assessed the response of biofilms in each line for 12 weeks. Different lines were exposed to different dosing regimes (see Table 1). As the focus of this study was to investigate the potential synergistic effect of caustic and oxygen on sewer biofilms rather than electrochemically producing caustic and oxygen as demonstrated previously (Lin et al. 2015), purchased caustic soda and aeration stone were employed (see Fig. 1). Caustic shock-loading was manually applied to both caustic and oxygen line and caustic line on a weekly basis. A NaOH solution (12 M) was added to the wet-wells before sewage being fed to reactors for 6 successive pumping events on day 0, resulting in the elevation of the pH in reactors to a value of ~12 for 8 h. Subsequently, a recovery period of 6 days and 16 h (i.e. day 0 to day 6) was applied before the next caustic treatment. In the mimicking wet-well of the caustic and oxygen line, a DO concentration of ~ 6.5 mg/L in the sewage was obtained by means of air addition using an aeration stone for 11 min before sewage being fed into the reactors prior to each pumping event. 2.3. Reactor monitoring and sampling campaign Both reactor performance monitoring and batch tests were carried out during the baseline and experimental phase. During the baseline phase, determination of the reactor performance and the batch tests were carried out on a weekly basis to ensure stable and comparable sulfide (as well as methane) production rates in all lines. During the experimental phase, the reactor performance was measured from day 1 to day 5 after each caustic shock. Batch tests were performed on day 3 after each caustic addition. Two on-line S::CAN sensors (spectro::lyser, s::can Messtechnik GmbH) were used for the continuously monitoring of total dissolved sulfide concentration in the reactors (SutherlandStacey et al. 2008). The reactor performance was determined by monitoring the sulfide concentration throughout the pumping cycle without disturbance of the operational routine. For the batch tests, fresh sewage (without any caustic and oxygen addition) was pumped into each line
for 6 min to replace the sewage in the reactors followed by continuous monitoring of sulfide concentrations throughout a period of 1 h for the determination of sulfide production rate (SPR). Liquid samples were taken through the sampling ports at 15-minute intervals for the determination of methane production rate (MPR). In addition, liquid samples were taken at the end of pumping cycle 3 (HRT of 3 h) for sulfide measurement from reactor RE2, RE4 and RC2 at day 0 (before caustic addition), 24, 48, 72 and 168 h after caustic treatment, giving the highest sulfide concentrations of day 0 (before caustic addition), 1, 2, 3 and 7 (day 0 of the next caustic addition) over the course of two consecutive caustic additions. The methane concentrations of RE1, RE3 and RC1 were also determined at the end of pumping cycle 3 prior to the next caustic treatment in order to monitor the weekly recovery of methane levels. Additionally, 24-hour composite samples were collected for the measurement of volatile suspended solids (VSS) and total suspended solids (TSS). VSS and TSS samples were collected on the day before each caustic addition, as well as on day 0, 1 and 2 in order to determine the amount of biofilm being washed out after each shock loading event. DO contents and pH levels were continuously measured using an on-line DO sensor (Mettler Toledo, Switzerland) and pH sensors (Hanna Instrument, USA), respectively. 2.4. Microbial sampling and analyses Biofilm samples were collected from the carriers equipped in each reactor during the baseline phase as well as during the experimental phase in weeks 8 and 12. The samples in the experimental phase were taken 6 days after chemical dosing (i.e. one day before a next dosing event). Genomic DNA was extracted using FastDNA SPIN Kit for soil (MP Biomedicals, USA). Subsequently, extracted DNA samples were submitted to the Australian Centre for Ecogenomics (the University of Queensland) for 16S Amplicon sequencing by Illumina Miseq Platform using pyroLSSU926F (5′-AAACTYAAAKGAATTGACGG-3′) and pyroLSSU1392R (5′-ACGGGCGGTGWGTRC-3′) primer set (Dove et al. 2013). Sequencing results were analysed according to Hülsen et al. (2016). Filtered Open taxonomy units (OTUs) table were rarefied by
Table 1 Experimental conditions for baseline and experimental phases.
Baseline phase Experimental phase
Line
Reactor
pH
NaOH exposure time (hour)
Dissolved oxygen concentration (mg/L)
All lines Caustic and oxygen
All reactors RE1 RE2 RE3 RE4 RC1 RC2
Not controlled ~12
Not controlled 8 h (weekly)
Anaerobic condition ~6.5 mg/L in sewage in the mimicking wet-well before being fed to reactors
~12
8 h (weekly)
Anaerobic condition
Not controlled
Not controlled
Anaerobic condition
Caustic Control
H.-W. Lin et al. / Science of the Total Environment 601–602 (2017) 776–783
function “rarefy_even_depth” of package phyloseq in R (RCoreTeam) and principle component analysis (PCA) were performed by ampvis (Albertsen et al. 2015). A quantitative polymerase chain reaction (qPCR) analysis was performed to quantify microbial load, as described in Vanwonterghem et al. (2014). 2.5. Modelling of sulfide and methane inhibition A set of dynamic simulations were conducted using the SeweX model (Sharma et al. 2013) to assess the performance of each sulfide control strategy (i.e. no dosing, oxygen injection, caustic shock, and caustic shock and oxygen injection) under different sewer conditions. The addition of oxygen was conducted at the beginning of the sewer pipe in a continuous fashion, reaching initial DO levels of 6.5 mg O2/L in sewage while caustic shock-loading to pH 12 was assumed to have a periodicity of 3 days. Default model parameters were used as in Sharma et al. (2013), except for the sulfide production (surface) rate, methane production (surface) rate and oxygen uptake rate, which were adopted based on the experimental results obtained from the three different dosing regimes. Parameter values were calculated based on the average of all individual tests performed on day 3 after the caustic shock, and corrected to a temperature of 20 °C according to the van't Hoff-Arrhenius equation (Tchobanoglous et al. 2003). Sewage composition was assumed constant and defined based on the average concentrations recorded throughout the experiments. Table 2 shows the sewer conditions and dosing scenarios used in simulation studies. First, we simulated a rising main with a pipe length of 2500 m, pipe diameter of 0.25 m (A/V ratio of 16) and HRT of 3 h at 20 °C, after 3 days of a caustic shock. These parameters were chosen based on the industry survey of chemical dosing for sulfide control as reviewed by Ganigue et al. (2011). Subsequently, three key sewer characteristics, namely, the A/V ratio, HRT and temperature, were changed individually to assess their impact on the effectiveness of sulfide control (see Table 2) by simulating the sulfide profiles along the rising main also 3 days after a caustic shock. 2.6. Analytical methods and calculations Dissolved sulfur species (i.e. sulfide, sulfate, sulfite and thiosulfate) were measured using Ion Chromatography (IC) equipped with a Dionex 2010i system according to Keller-Lehmann et al. (2006). Dissolved methane measurements were determined using the methodology described by Alberto et al. (2000) and Guisasola et al. (2008). 5 mL of
779
sewage samples were filtered (0.22 μm) and injected into 12 mL evacuated Exetainer® tubes (Labco, UK). The tubes allowed equilibrium of gas and liquid phases overnight. Subsequently, the gaseous methane in the headspace of the tube was measured by means of gas chromatography (GC) equipped with a flame ionization detector. The dissolved methane concentration in the sewage sample was calculated using Henry's law (Alberto et al. 2000). SPR and MPR were calculated using linear regression. Oxygen uptake rate was calculated using linear regression based on the real-time DO concentrations measured in RE1. TSS and VSS were analysed using Standard Methods 2540D and 2540E, respectively (APHA 1995). 3. Results and discussion 3.1. Long-term reactor performance Fig. 2 shows the typical performance profiles observed on day 1 after a caustic treatment event. During the course of the experimental period, both caustic and oxygen line and caustic line showed a significant reduction in sulfide levels after caustic treatment. Overall, the caustic and oxygen line showed no sulfide accumulation compared to the caustic line, indicating either oxidation of sulfide took place or the activity of SRB was fully inhibited. The latter was further confirmed by IC analyses which revealed that no sulfate reduction was observed in RE1 (Fig. S2), suggesting intermittent oxygen addition after caustic treatment is likely to inhibit the activity of the remaining SRB. Contrarily, sulfide production activity was still observed in the caustic line after caustic treatment (Fig. 2), showing that caustic treatment did not control the activity of SRB entirely. A complete overview of the weekly results after one day of caustic treatment throughout the experimental phase can be found in the supplementary material (Fig. S3–5). In addition, the typical profiles of pH changes on day 0 in both caustic and oxygen line and caustic line are illustrated in Fig. S6. The daily recovery of sulfide levels in the caustic and oxygen line and the caustic line are shown in Fig. 3. On day 5, the maximum relative sulfide concentrations (the sulfide concentration of control line is considered as 100%) in RE1 and RE2 were 37.0% and 44.5%, respectively. On the contrary, the results showed the maximum relative sulfide concentrations of 98.3% in RE3 and 102.7% in RE4, indicating that a full recovery of sulfide levels in the caustic line was reached on day 5. Although the sulfide level recovered gradually in the caustic and oxygen line, it recovered much slower compared to the caustic line and did not return to the pre-shock level (i.e. control sulfide level). The latter was further
Table 2 Sewer conditions and dosing scenarios used in simulation studies. Pipe length (m)
Pipe diameter (m)
A/V ratio
HRT (hours)
Temperature (°C)
2500
0.25
16
3
20
Modified cases A/V ratio
2500
0.25
3
20
HRT
2500
0.25
8 16 26.66 16
20
Temperature
2500
0.25
16
1.5 3 5 8 3
Proposed standard rising main
10 15 20 25
Dosing scenarios
pH DO (mg O2/L)
Caustic and Oxygen
Caustic
Oxygen
Control
12 6.5
12 –
– 6.5
– –
780
H.-W. Lin et al. / Science of the Total Environment 601–602 (2017) 776–783
Fig. 2. Typical sulfide profiles of the first (A) and second reactors (B) one day after caustic treatment for the caustic and oxygen line, caustic line and control line. No chemical addition was performed in the control line.
supported by the highest sulfide levels of the day over the course of two consecutive caustic treatments (Fig. 4). The shaded areas represent the time that caustic was added to the reactor. On average, the caustic and oxygen line showed lower sulfide levels compared to the caustic line. The relative sulfide level of caustic and oxygen line in the end of the recovery period (i.e. day 7 and day 14) was 58 ± 9% (n = 2; 2 weeks) in comparison with the control line. A full recovery of the sulfide production in the caustic line was observed approximately after day 3, indicating a fully recovery of sulfate-reducing activity (in comparison with the control line) in the caustic line between day 4 and day 6. Daily solid discharge and relative solid discharge from the reactors before, during (Day 0) and after the caustic treatment (Day 1 and 2) can be found in Table S1 and Table S2, respectively. On average, a
significant amount of solid was washed out from the caustic and oxygen line and caustic line on day 0 (during caustic treatment). The VSS/TSS ratios observed during caustic treatment were lower than the ratios observed on other days. This suggested that a fraction of the TSS adsorbed to the biofilms were released through the detachment of biofilm into the sewage. No significant difference in solid discharge between caustic line and caustic and oxygen line was observed, indicating the aggravation of the biofilm detachment caused by the oxygen addition did not take place. Solid discharges then returned to the stable level on day 1 and 2 after caustic treatment, giving similar values in comparison with the control line. The average VSS/TSS ratio in the sewage influent during 12 weeks was 93 ± 2%, which was slightly higher than the typical VSS/ TSS ratios in municipal wastewater (i.e. 40–90%) (Henze et al., 2008).
Fig. 3. The recovery of sulfide levels in the caustic and oxygen line (RE1 and RE2) and the caustic line (RE3 and RE4) after caustic treatment. The monitoring for caustic and oxygen line and caustic line was carried out daily for one complete pumping cycle in week 3 and week 4, respectively.
H.-W. Lin et al. / Science of the Total Environment 601–602 (2017) 776–783
Fig. 4. Measurements of highest dissolved sulfide level at the end of the longest pumping event (HRT: 3 h) over the course of two consecutive caustic additions. The shaded areas represent the additions of caustic.
781
Fig. 6. Average oxygen uptake rate (OUR) measured at 152 to 160 h after caustic addition over the course of 12 weeks in comparison with the OUR reported in Gutierrez et al. (2008).
However, the VSS/TSS content in sewage is not expected to have impacted the experimental results. 3.2. Inhibition of sulfide producing activity Batch activity test were conducted on a weekly basis to determine the sulfide production rate (SPR) in the absence of oxygen on day 3 after caustic treatment. Fig. 5 shows that on average, the relative SPRs of the caustic and oxygen line in comparison to the control line are 29.6 ± 13.1% for RE1 and 24.2 ± 8.5% for RE2 while the caustic line showed higher SPRs of 38.6.6 ± 11.3% for RE3 and 39.6 ± 16.6% for RE4. The result suggests that the additional oxygen supply after caustic treatment effectively further lowered the SPR by ≈ 20–40% on day 3 compared with the stand-alone caustic treatment. The latter is inconsistent with our previous study which has found that short periods of oxygen exposure do not affect the sulfide producing activity of SRB (Gutierrez et al. 2008). It further reveals that intermittent oxygen could potentially lead to a more long-lasting suppression effect on the remaining SRB after caustic treatment. Although certain SRB strains are able to survive to oxygen exposure by defence mechanisms such as aggregation and even oxygen reduction or reactive oxygen species (ROS) detoxification, oxygen tolerance of SRB is species-dependent (Cypionka et al. 1985; Dolla et al. 2006; Sass et al. 1998). Oxygen can have toxic effect on some SRB without oxygen tolerance. In addition, the presence of ROS can damage cell components and could be
responsible for oxygen toxicity (Baumgartner et al. 2006; Dolla et al. 2006). These might contribute to the better SRB suppression observed in the caustic and oxygen line. Also, it is noteworthy that the relative SPRs of treated reactors (i.e. RE1, RE2, RE3 and RE4) slightly increased over the course of the experiment in comparison to the control line. This was likely due to the increase in the sewage temperature over time from 21 ± 1 °C in weeks 1–4 to 26 ± 1 °C in weeks 5–12. This is supported by the fact that biofilm removal was observed in RE1 and RE2 after each caustic shock (see supplementary material, Fig. S7).
3.3. Oxygen uptake rate The typical profiles of DO concentrations in RE1 and RE2 on day 0, 1, 2 and 6 (i.e. 152 to 160 h after caustic addition) are given in Figs. S8 and S9, respectively. Overall, the oxygen uptake rates (OUR) of RE1 and RE2 were highest on day 0 and decreased gradually. Previously, it was found that the repeated exposure of biofilms to oxygen significantly increased the OUR with a four to five-fold increase (Gutierrez et al. 2008). The latter was found to be related to the growth of heterotrophic bacteria. Fig. 6 shows an overview of OUR in RE1 measured at 152 to 160 h after caustic addition over the course of 12 weeks in comparison with the OUR reported in Gutierrez et al. (2008). Here, the OUR remained constant over
Fig. 5. Sulfide production rates (SPR) obtained from batch tests on day 3 in absence of oxygen. Experimental temperatures ranged from 21 ± 1 °C (baseline and week 1 to week 4) to 26 ± 1 °C (week 5 to week 12) over the course of experimental phase. (A) and (B) represent the SPR of first reactors and second reactors, respectively.
782
H.-W. Lin et al. / Science of the Total Environment 601–602 (2017) 776–783
RE2, 8% in RE3 and 12% in RE4 (the total community in each reactor in baseline phase is considered as 100%). Furthermore, the relative abundance of methanogens in the biofilm in week 8 was greatly reduced to b1% and 5% in RE1 and RE2, respectively, but was still present in the caustic line (with 10–15% Methanobacterium in RE3 and 4). Additionally, changes in the biofilm colour and coverage of the walls of the laboratory rising main reactors were visually observed in the caustic and oxygen line (see Fig. S7) while the biofilm colour in the caustic line (see Fig. S11) remained similar to the control line (see Fig. S12). This gradual biofilm colour change in the caustic and oxygen line suggests that the addition of oxygen after caustic shock resulted in a microbial community change. Indeed, a shift from Methanosaeta and Bacteroidales to Acinetobacter, family Porphyomonadaceae and Peptostreptococcaceae was observed, as shown in the detailed principle component analysis of microbial community in all reactors (see Fig. S13). Further research such as metaproteomic profiling is warranted to fundamentally understand the behaviour of SRB, methanogenic archaea or other bacteria under caustic stress with additional oxygen supply.
3.5. Simulation studies
Fig. 7. Total dissolved sulfide, methane and dissolved oxygen profiles. In all three cases, a sewer pipe with a length of 2500 m, A/V ratio of 16, HRT of 3 h and temperature of 20 °C was used.
Fig. 7 depicts the total dissolved sulfide, methane and oxygen profiles along a 2.5 km rising main for 4 different dosing scenarios, 3 days after a caustic shock. Results showed that the combined caustic and oxygen dosing substantially improved the reduction in total dissolved sulfide and methane levels in comparison to the caustic shock, with further reductions in total dissolved sulfide and methane level of 33% and 67%, respectively. Compared to the control line, oxygen dosing only reduced the total dissolved sulfide level by 6% and had little effect on methane control (i.e. 7% reduction). In addition, a number of simulations under various sewer conditions (i.e. pipe diameters, temperatures and HRT) were carried out to further assess the treatment effectiveness in terms of sulfide control, suggesting caustic and oxygen dosing is a more efficient sulfide control strategy based on the results of simulation. More detailed results about these simulations can be found in the supplementary material (Figs. S14–16). 3.6. Implications for practice
12 weeks, suggesting that the development of a heterotrophic biofilm did not take place.
3.4. Inhibition of methane-producing activity Prior to the next caustic dosing (n = 3; day 6 of week 6, 7 and 8), a negligible amount of methane (i.e. 0.2 ± 0.1 mg/L, equal to 0.6 ± 0.3 mg-COD/L) was detected in the caustic and oxygen line (RE1) while 1.4 ± 0.1 mg/L (i.e. 5.7 ± 0.6 mg-COD/L) was detected in the caustic line (RE3). The discharge of methane on day 6 in both caustic and oxygen line and caustic line was reduced notably. Compared these values to the methane level of 16.1 ± 1.6 mg/L (i.e. 64.4 ± 6.4 mg-COD/L) in the control line RC1, the relative methane discharge levels were reduced significantly to 0.9 ± 0.5% for caustic and oxygen line and 8.9 ± 1.4% for caustic line. Additionally, the results of the batch activity tests on day 3 show that both caustic and oxygen line and caustic line had significant lower MPRs compared with the control line. The caustic and oxygen line had lower MPRs on day 3 compared to the caustic line, accounting for 39% for the first reactor and 63% for the second reactor of the caustic line. In addition, the addition of oxygen reduced methane concentrations to a negligible level and slowed down the re-growth of methanogenic archaea compared to caustic treatment (see supplementary material, Fig. S10). The results of qPCR analysis further determined that the total community in week 12 were reduced to 26% in RE1, 8% in
Both long-term laboratory experiments and simulation results suggest that periodic caustic treatment combined with intermittent addition of oxygen substantially enhanced the effectiveness of sulfide control in comparison with caustic treatment alone. Our results show that in order to achieve the same efficiency in terms of sulfide control, the dosing frequency for caustic shock-loading is doubled compared to caustic shock and oxygen addition (e.g. weekly dosing for caustic and oxygen addition; 3.5 day dosing frequency for caustic treatment under the sewer conditions simulated). Equally important, in addition to sulfide control, also a better overall control of methane was achieved. In full-scale applications, this combined dosing method is especially suitable for smaller rising mains (e.g. ≤0.3 m) with a relative low flow rate (average dry weather flow b0.5 MLd−1), conditions known to be suitable for caustic dosing (Ganigue et al. 2011). Large size rising mains with large flows are less suitable due to the large caustic requirements. In practice, the required amounts of caustic and oxygen could also be produced on-site by using an electrochemical system, as previously demonstrated by Lin et al. (2015).
4. Conclusions In this study, we investigated an innovative sulfide control strategy which combines periodic caustic shock-loading with intermittent supply of oxygen for efficient sulfide control in sewers. The key findings of this work are:
H.-W. Lin et al. / Science of the Total Environment 601–602 (2017) 776–783
• Long-term experiments showed that the simultaneous use of periodic caustic shock-loading and intermittent oxygen supply during and after caustic treatment achieves more efficient sulfide control in comparison with caustic treatment alone, leading to a prolonged biofilm recovery period in between caustic shocks. • The addition of oxygen reduced methane emissions to a negligible level and slowed down the re-growth of methanogenic archaea in comparison with caustic shock-loading. The relative abundance of methanogens in the caustic and oxygen line was greatly reduced compared to the caustic line. • Based on the experimentally results obtained and simulation studies was conducted, it was demonstrated that the simultaneous use of periodic caustic shock-loading and intermittent oxygen supply is more effective in sulfide control compared to conventional caustic shockloading and reduced the dosing frequency for caustic treatment by 50%. Acknowledgements H.-W. L thanks the University of Queensland for scholarship support. R.G. gratefully acknowledges support from Ghent University BOF postdoctoral fellowship (BOF15/PDO/068). This work was supported by the Australian Research Council, District of Columbia Water and Sewer Authority, ACTEW Corporation Limited, The City of Gold Coast, Queensland Urban Utilities and Yarra Valley Water through ARC Linkage project LP120200238: “In-situ electrochemical generation of caustic and oxygen from sewage for emission control in sewers”. The authors acknowledge Dr. Beatrice Keller-Lehmann and Nathan Clayton for their helpful assistance with the chemical analyses. We also want to thank Dr. Nicola Angel from ACE for the assistance of 16S rRNA amplicon sequencing. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.scitotenv.2017.05.225. References Alberto, M.C.R., Arah, J.R.M., Neue, H.U., Wassmann, R., Lantin, R.S., Aduna, J.B., Bronson, K.F., 2000. A sampling technique for the determination of dissolved methane in soil solution. Chemosphere Global Change Sci. 2 (1), 57–63. Albertsen, M., Karst, S.M., Ziegler, A.S., Kirkegaard, R.H., Nielsen, P.H., 2015. Back to basics the influence of DNA extraction and primer choice on phylogenetic analysis of activated sludge communities. PLoS One 10 (7), e0132783, e0132783. APHA, 1995. Standard Methods for the Examination of Water and Wastewater. Baumgartner, L.K., Reid, R.P., Dupraz, C., Decho, A.W., Buckley, D., Spear, J., Przekop, K.M., Visscher, P.T., 2006. Sulfate reducing bacteria in microbial mats: changing paradigms, new discoveries. Sediment. Geol. 185 (3-4), 131–145. Cypionka, H., Widdel, F., Pfennig, N., 1985. Survival of sulfate-reducing bacteria after oxygen stress, and growth in sulfate-free oxygen-sulfide gradients. FEMS Microbiol. Lett. 31 (1), 39–45. Dolla, A., Fournier, M., Dermoun, Z., 2006. Oxygen defense in sulfate-reducing bacteria. J. Biotechnol. 126 (1), 87–100.
783
Dove, S.G., Kline, D.I., Pantos, O., Angly, F.E., Tyson, G.W., Hoegh-Guldberg, O., 2013. Future reef decalcification under a business-as-usual CO2 emission scenario. Proc. Natl. Acad. Sci. 110 (38), 15342–15347. Ganigué, R., Yuan, Z., 2014. Impact of oxygen injection on CH4 and N2O emissions from rising main sewers. J. Environ. Manag. 144, 279–285. Ganigue, R., Gutierrez, O., Rootsey, R., Yuan, Z., 2011. Chemical dosing for sulfide control in Australia: an industry survey. Water Res. 45 (19), 6564–6574. Guisasola, A., de Haas, D., Keller, J., Yuan, Z., 2008. Methane formation in sewer systems. Water Res. 42 (6-7), 1421–1430. Guisasola, A., Sharma, K.R., Keller, J., Yuan, Z., 2009. Development of a model for assessing methane formation in rising main sewers. Water Res. 43 (11), 2874–2884. Gutierrez, O., Mohanakrishnan, J., Sharma, K.R., Meyer, R.L., Keller, J., Yuan, Z., 2008. Evaluation of oxygen injection as a means of controlling sulfide production in a sewer system. Water Res. 42 (17), 4549–4561. Gutierrez, O., Park, D., Sharma, K.R., Yuan, Z., 2009. Effects of long-term pH elevation on the sulfate-reducing and methanogenic activities of anaerobic sewer biofilms. Water Res. 43 (9), 2549–2557. Gutierrez, O., Sudarjanto, G., Sharma, K.R., Keller, J., Yuan, Z., 2011. SCORe-CT: a new method for testing effectiveness of sulfide-control chemicals used in sewer systems. Water Sci. Technol. 64 (12), 2381–2388. Gutierrez, O., Sudarjanto, G., Ren, G., Ganigue, R., Jiang, G., Yuan, Z., 2014. Assessment of pH shock as a method for controlling sulfide and methane formation in pressure main sewer systems. Water Res. 48 (1), 569–578. Henze, M., van Loosdrecht, M.C., Ekama, G.A., Brdjanovic, D., 2008. Biological wastewater treatment. IWA publishing. Hülsen, T., Barry, E.M., Lu, Y., Puyol, D., Keller, J., Batstone, D.J., 2016. Domestic wastewater treatment with purple phototrophic bacteria using a novel continuous photo anaerobic membrane bioreactor. Water Res. 100, 486–495. Keller-Lehmann, B., Corrie, S., Ravn, R., Yuan, Z., Keller, J., 2006. Preservation and Simultaneous Analysis of Relevant Soluble Sulfur Species in Sewage Samples. p. 28 (Vienna, Austria). Lin, H.-W., Rabaey, K., Keller, J.r., Yuan, Z., Pikaar, I., 2015. Scaling-free electrochemical production of caustic and oxygen for sulfide control in sewers. Environ. Sci. Technol. 49 (19), 11395–11402. Nielsen, A.H., Vollertsen, J., Hvitved-Jacobsen, T., 2003. Determination of kinetics and stoichiometry of chemical sulfide oxidation in wastewater of sewer networks. Environ. Sci. Technol. 37 (17), 3853–3858. O'Gorman, J., Purssell, I., Iori, G., 2011. Caustic soda washing of a sewer pressure main. Water 38 (2), 83–87. Pikaar, I., Sharma, K.R., Hu, S., Gernjak, W., Keller, J., Yuan, Z., 2014. Reducing sewer corrosion through integrated urban water management. Science 345 (6198), 812–814. Sass, H., Berchtold, M., Branke, J., König, H., Cypionka, H., Babenzien, H.-D., 1998. Psychrotolerant sulfate-reducing bacteria from an oxic freshwater sediment description of Desulfovibrio cuneatus sp. nov. and Desulfovibrio litoralis sp. nov. Syst. Appl. Microbiol. 21 (2), 212–219. Sharma, K., Ganigue, R., Yuan, Z.G., 2013. pH dynamics in sewers and its modeling. Water Res. 47 (16), 6086–6096. Sun, J., Pikaar, I., Sharma, K.R., Keller, J., Yuan, Z., 2015. Feasibility of sulfide control in sewers by reuse of iron rich drinking water treatment sludge. Water Res. 71, 150–159. Sutherland-Stacey, L., Corrie, S., Neethling, A., Johnson, I., Gutierrez, O., Dexter, R., Yuan, Z., Keller, J., Hamilton, G., 2008. Continuous measurement of dissolved sulfide in sewer systems. Water Sci. Technol. 57 (3), 375–381. Tchobanoglous, G., Burton, F.L., Stensel, H.D., Metcalf and Eddy, 2003. Wastewater Engineering: Treatment and Reuse. McGraw-Hill, Boston. Vanwonterghem, I., Jensen, P.D., Dennis, P.G., Hugenholtz, P., Rabaey, K., Tyson, G.W., 2014. Deterministic processes guide long-term synchronised population dynamics in replicate anaerobic digesters. ISME J. 8 (10), 2015–2028. Vollertsen, J., Nielsen, A.H., Jensen, H.S., Wium-Andersen, T., Hvitved-Jacobsen, T., 2008. Corrosion of concrete sewers—the kinetics of hydrogen sulfide oxidation. Sci. Total Environ. 394 (1), 162–170. WERF, 2007. Minimization of Odors and Corrosion in Collection Systems Phase 1. Water Environment Research Foundation, London, UK. Zhang, L., De Schryver, P., De Gusseme, B., De Muynck, W., Boon, N., Verstraete, W., 2008. Chemical and biological technologies for hydrogen sulfide emission control in sewer systems: a review. Water Res. 42 (1-2), 1–12.