- Email: [email protected]
Organics and nitrogen recovery from sewage via membrane-based pre-concentration combined with ion exchange process
Accepted Manuscript Organics and nitrogen recovery from sewage via membrane-based pre-concentration combined with ion exchange process Hui Gong, Zhijie Wang, Xue Zhang, Zhengyu Jin, Cuiping Wang, Liping Zhang, Kaijun Wang PII: DOI: Reference:
S1385-8947(16)31630-8 http://dx.doi.org/10.1016/j.cej.2016.11.068 CEJ 16069
To appear in:
Chemical Engineering Journal
Received Date: Revised Date: Accepted Date:
20 September 2016 7 November 2016 8 November 2016
Please cite this article as: H. Gong, Z. Wang, X. Zhang, Z. Jin, C. Wang, L. Zhang, K. Wang, Organics and nitrogen recovery from sewage via membrane-based pre-concentration combined with ion exchange process, Chemical Engineering Journal (2016), doi: http://dx.doi.org/10.1016/j.cej.2016.11.068
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Organics and nitrogen recovery from sewage via membrane-based pre-concentration combined with ion exchange process
Hui Gong1, Zhijie Wang1, Xue Zhang2, Zhengyu Jin1, Cuiping Wang1, Liping Zhang2, Kaijun Wang1,* 1．State Key Joint Laboratory of Environment Simulation and Pollution Control, School of Environment, Tsinghua University, Beijing 100084, China 2．School of Chemical & Environmental Engineering, China University of Mining & Technology, Beijing 100083, China
Submitted to Chemical Engineering Journal
* Corresponding Author Email:
[email protected]
Phone: (+86) 010-62789411
1
Abstract: This study proposes the recovery of organics and nitrogen from sewage through membrane-based pre-concentration (MPC) combined with ion exchange (IE) process. Unlike conventional activated sludge process, MPC–IE redesigned organic carbon flow to increase chemical oxygen demand (COD) conversion for energy recovery via anaerobic digestion (AD). This process also achieved nitrogen recovery instead of destruction. Membrane-based pre-concentration of COD was conducted using actual sewage for one month. As high as 65% COD was recovered. Batch IE adsorption and regeneration experiments were conducted. However, these experiments recovered only 37.5% NH4-N because of the selectivity of IE for hard ions over ammonium. Despite the low rate of recovery, the process could achieve a total of 0.38 kWh/m3 energy recovery by combining energy production with anaerobic digestion of pre-concentrated organics (0.26 kWh/m3) and energy saving via nitrogen reuse (0.12 kWh/m3). Highly efficient energy and nitrogen recovery from sewage via MPC–IE process was expected after optimization.
Keywords: Sewage; Nitrogen recovery; Energy recovery; Membrane-based pre-concentration; Ion exchange
2
1. Introduction In response to global resource shortage, domestic wastewater should not be treated as waste anymore, but a resource of water, energy, and plant fertilizing nutrients (nitrogen and phosphorus). However, the objective of sustainable resource recovery from domestic wastewater is far from being achieved at present, which is limited by the disadvantages of conventionally activated sludge (CAS). CAS and its modification processes require high-energy consumption during aeration to supply oxygen for organic matter removal by aerobic microorganism in activated sludge. The anaerobic digestion of sludge can offset part of energy requirement, but only a portion of energy potential in organics is captured. Wastewater treatment now accounts for 0.3% and 3% of the total electrical energy load in China and the U.S., respectively [1, 2]. Nitrogen, mostly in form of N2 , is removed and not recovered during CAS process. Water reuse is widely practiced in water shortage regions, but this also increases energy consumption because of long process to satisfy high-quality requirements for water reuse. Therefore, novel short-procedure technologies/processes for resource recovery are in demand. Novel concepts and processes have been proposed in recent years to promote paradigm shift to recovering resources from domestic wastewater. McCarty et al. discussed the potential of wastewater treatment plant (WWTP) as net energy producer [3]. Rulkens emphasized increasing significance of advanced physical/chemical processes to develop sustainable treatment systems[4]. Verstraete et al. proposed a zero-wastewater concept with up-concentration process, followed by organic anaerobic digestion to achieve maximum energy recovery in domestic used water [5]. Kartal et al. proposed a novel sewage treatment process with increased energy recovery by organics bio-adsorption and low energy consumption by anammox [6]. Batstone et al. suggested partition–release–recover concept in which carbon and nutrients are partitioned to solids through either heterotrophic or phototrophic microbes followed by anaerobic digestion[7]. This study proposed the concept of membrane-based pre-concentration (MPC) combined with ion exchange (IE) process to achieve complete recovery of water, energy, and nutrients (nitrogen). Sewage enters MPC reactor where organic matter is flocculated and separated by membrane filtration. Adsorbent (powder-activated carbon) is also added to enhance soluble organic matter removal and to alleviate membrane fouling issue. After MPC, most organic matter is separated from the mainstream and presented in concentrated state, which could obtain energy recovery in form of biogas (methane) generation by organic anaerobic digestion. The MPC effluent (alone or combined with digester effluent) enters the IE 3
process to recover nitrogen in the form of NH4-N. The chemical oxygen demand (COD) pre-concentration is essential to improve energy recovery and to achieve energy neutrality. A WWTP in Strass, Austria demonstrates large-scale feasibility in energy self-sufficiency by increasing the transfer of less-stabilized organics (as high as 60% COD) from the mainstream sewage to anaerobic digesters in form of sludge [8]. Research on highly efficient CODpre-concentration was needed to improve proportion of concentrated organic matters for enhanced energy recovery in anaerobic digestion. Mels and Van Nieuwenhuijzen evaluated physical−chemical technologies including pre-precipitation and flotation, whose efficiency on soluble COD concentration, however, is low [9, 10]. MPC was proposed in our previous studies [11-14]. The high-separation efficiency suggested that membrane technology as a promising tool for organic recovery. Akanyeti demonstrated high-load membrane biological reactor (MBR) for efficient organic concentration in grey water through bio-flocculation[15]. Mezohegyi evaluated vibrating membranes for sewage concentration utilizing vibration to overcome fouling issue [16] . Since
Haber-Bosch process invention in 1908 enabling human society producing reactive nitrogen
in large amounts and low price, overloaded anthropogenic reactive nitrogen has end up in environments. Nitrogen recovery could significantly benefit from reduced loading in local environments and nitrogen cycle. However, nitrogen recovery is ignored in conventional wastewater treatment process (e.g., CAS) whose nitrogen control technologies are based on destructive technologies. Actually, nitrogen recovery potentialis remarkable given the increasing domestic wastewater production caused by population growth and economic development. Nitrogen in domestic wastewater accounts for 2% to 8% of total nitrogen input in China (estimated by national sewage treatment capacity of 1.5 × 108 m3/d [1]). IE is commonly used to recover nitrogen, but mostly for high-ammonium wastewater (200 to 8000 mg/L TN) rather than low-strength domestic wastewater (20 to 80 mg/L TN) [17-19]. IE performance for nitrogen recovery, as well as MPC, has not been investigated in previous studies. The MPC–IE process was proposed and investigated for the first time. In this study, we operated PC reactor with actual sewage for one month and conducted batch IE adsorption and regeneration experiments with membrane reactor effluent. The aim of this paper is to investigate the feasibility of the MPC−IE process by determining (i) organic recovery by MPC, (ii) the advantages and disadvantages of nitrogen recovery by IE resins, and (iii) performance of MPC–IE interaction. 2． ．Materials and methods 4
2.1 Membrane-based COD pre-concentration 2.1.1 Raw sewage Raw sewage for MPC experiment was collected after grid in Xiaojiahe municipal WWTP in Beijing, China. The general characteristics were listed in Table 1. The average COD and NH4-N of fresh sewage for testing period were 260 mg/L and 27.4 mg/L, respectively. Table 1. Average raw sewage quality of COD, TN, TP, NH4-N, and turbidity for MPC
Sewage
COD
TN
TP
NH4-N
Turbidity
(mg/L)
(mg/L)
(mg/L)
(mg/L)
(NTU)
260±32
34.5±1.7
2.78±0.38
27.4±1.3
114±13
2.1.2 Membrane reactor and experimental setup The membrane reactor was designed for sewage preconcentration. Figure 1 showed the schematic diagram of the setup of membrane reactor. The effect volume of filtration tank was 3.5 L, and the dimensions were 430 mm long, 250 mm wide, and 60 mm high. The filtration tank contained one-membrane module with a total of 0.33 m2 membrane surface. Hollow-fiber module was selected because of its large-membrane-surface area-volume ratio. Commercial hollow-fiber polyvinylidene fluoride (PVDF) microfiltration membrane (Origin Water Co. Ltd. China) with average pore size of 0.3 µm was applied in this study. The inner and outer diameters of the membrane fiber were 0.95 and 2.45 mm, respectively. The PVDF membrane module was stored in tap water before operation. Membrane filtration experiment was conducted in outside-in mode. Pressure gauge, flow meter, and paperless recorder were used for automatic monitoring and recording of flux and transmembrane pressure (TMP). Raw sewage was fed into the reactor. Polyaluminum chloride (Al2(OH)nCl6-n]m) measuring 30 mg/L and 10 mg/L power active carbon (iodine value = above 1,000; 0.45 to 0.55 g/cm3 apparent density) were added according to previous study on coagulant selection. A piston pump (Seko, Italy) was used to draw out permeate from the membrane module. Concentrates were retained in the reactor. Initial filtration flux was set to 20 L/m2 h. Liquid level in the reactor was kept constant by peristaltic pump and raw sewage was fed automatically. To avoid severe membrane fouling, the membrane module underwent permeation for 10 mins followed by a 2-min relaxation. During the 2-min relaxation, back-flushing aeration was activated to remove cakes on membrane surface. Back-flushing lasted for 50 s at a pressure of 50 kPa. The permeated and concentrated COD and NH4-N of influent raw sewage were sampled for analysis every 48 h. Given the settling mechanism of particulate COD in the reactor tank, short aeration was provided (1 to 2 mins) to mix 5
the tanks when sampling for COD concentration performance evaluation. The MPC process with two stages lasted for 600 h. The first stage (first 360 h) did not exhibit concentrates discharge. This result enabled the organic matters in sewage to be concentrated quickly in the reactor (except for the 200 ml concentrates which was sampled every 48 h). The concentrate started to be discharged with sludge retention time (SRT) for 3.5 d in the second stage (360 to 600 h) when COD increased to over 9,500 mg/L. Chemical cleaning was not applied during the experiment.
Figure 1. Schematic diagram of MPC (left) and IE resin reactor for nitrogen recovery (right) 2.2 IE for nitrogen recovery 2.2.1 Feed for IE Three wastewater samples were applied to evaluate the effects of the different compositions (hard ions, including Ca2+, Mg2+, and other impurities) of sewage in IE for nitrogen recovery. Sample 1 was the permeate from the membrane reactor. Sample 2 was synthetic wastewater with 40 mg/L NH4-N. Sample 3 was synthetic wastewater with 40 mg/L NH4-N, 20 mg/L Ca2+ and 20 mg/L Mg2+. 2.2.2 IE bed-column setup As shown in Figure 1 (right), IE for nitrogen recovery was investigated by organic glass-packed bed column with an inner diameter of 24 mm and length of 120 mm. The column was filled with 54 mL strong acid cation-exchange resin (SAC; 001×7 resin; particle size 0.5 to 1 mm, XinKe Co. Ltd, Jinzhou, China). Before experiments, SAC resin was transformed to Na-form by pumping 10 g/L of NaCl solution at flow rate of 5.6 mL/min for 4 hours. Two processes of resin adsorption and regeneration were conducted. During resin adsorption, wastewater containing NH4-N was pumped into the column from the bottom. Treated wastewater in the outlet was sampled regularly. During resin regeneration, exhausted resin was regenerated by 10 g/L NaCl solution. Nitrogen recovery was achieved during regeneration. Concentrated 6
NH4-N content in spent regenerant was also measured regularly. 2.3 Analysis methods COD and NH4-N were determined by colorimetric techniques using HACH spectrophotometer (DR 5000, HACH, USA). SS was determined according to the Chinese NEPA Standard Methods [20]; pH was measured using a pH meter (Sension1, HACH, USA); and turbidity was measured using a turbidity meter (2100Q, HACH, USA). 3 Results and discussion 3.1 Performance of membrane-based COD pre-concentration 3.1.1 Pre-concentrated sewage by membrane process for improved energy recovery Organic matters (measured as COD) in sewage were presented in soluble, colloid and suspended state. To recover organics for energy recovery, MPC utilized membrane process separating flocculated colloid and suspended organics. Considering the separation of soluble organic matters from mainstream sewage, powder-activated carbon was added as adsorbent, which also was expected to alleviate membrane fouling issue. Enhanced COD removal was observed during sewage pre-concentration. Permeate COD was 19±3 mg/L with 260±32 mg/L raw sewage COD. The rate of COD removal was as high as 92.9±1.2 %. MPC has higher rate of COD removal than that in processes with direct membrane filtration (36% to 75%) [21] or coagulation alone (e.g., chemically enhanced primary treatment with COD removal of 46% to 50%) [22, 23]. This finding was attributed to combined membrane filtration and coagulation, which was beneficial for efficient sewage organics concentration.
7
Figure 2. COD of influent, permeate and concentrate of MPC reactor over time Figure 2 showed that efficient sewage pre-concentration was achieved. The COD of influent raw sewage was not high (260±32mg/L), but concentrated COD increased to 9640 mg/L in 360 h in the first stage. Concentrate started to be discharged with SRT at 3.5 d in the second stage (360 to 600 h). Concentrated COD was stable in the range of 9,000 to 10,000 mg/L. The increased level of concentrated COD in the second stage was as high as, even higher than that of black water [24, 25]. Black water was domestic wastewater originally distinguished from toilets (faeces, urine and flushing water) which contained concentrated high organic matters (COD) and was usually anaerobically treated. Thus, the pre-concentrated COD in this research can be anaerobically treated with energy recovery by converting organic matters into CH4. Moreover, the more concentrated organics the wastewater contained, the smaller the needed reactor size and space requirements was, and accordingly, the lower the manufacturing cost was[26].
8
Figure 3. COD balance of different stages during MPC (Left: first stage, 0 to 360 h; Right: second stage, 360 to 600 h) As shown in Figure 3, mass balance was calculated to quantify the efficiency of pre-concentration of sewage COD. Three categories, namely, permeated, concentrated, and others, were applied to represent COD discharged in permeate, concentrated in concentrate, and those that do not belong to the first two categories. Others were calculated based on total mass balance of COD. This category contained two parts. One part was mineralized COD, which was converted to CO2 through biochemical process, whereas the other part was those deposited in the membrane surface or in reactor corners (hydraulically dead zones). Permeated + mineralized represented the unrecoverable part, which was lost during pre-concentration. Concentrated + deposited denoted the recoverable part, which could be applied for energy recovery in post treatment. Concentrated COD in concentrate is limited in the first stage because large amount of concentrate was not discharged continuously. The majority of COD were concentrated in others, which was retained in the reactor. The amount of Mineralized, which was proportional to SRT, could be high. Faust found that the proportion of mineralized could increase to as high as 32% at 5 d SRT [27, 28]. This result was likely caused by the accumulated heterotrophic microorganisms in the reactor, which easily utilized biodegradable COD supplied with raw sewage. The first stage period should be controlled to a minimum to limit loss by mineralization and to obtain concentrate continuously. In the second stage, 65% COD was collected in the concentrate. The proportion of recoverable COD would increase if the COD retained in the reactor was considered. The proportion of Mineralized decreased with SRT. Faust demonstrated that the proportion of mineralized COD decreased rapidly with short SRT. Estimated COD loss by mineralization was controlled to 1% to 2% of the total amount by setting SRT at 9
0.125 to 0.25 d [27]. This result indicates that optimization by shortening SRT should be carefully considered in future research to improve the efficiency of COD pre-concentration. 3.1.2 Fouling issue during MPC Membrane fouling was important in determining the efficiency of sewage organics pre-concentration. High water flux would shorten the first stage and guarantee high level of concentrated COD when SRT was reduced. However, water flux in pre-concentration was undermined by membrane fouling. Figure 4 (A) showed that flux was initially set at 20 LHM and kept constant in the first 300 h. Flux started to decrease at the end of the first stage and finally stabilized at 5 to 10 LHM. TMP jumped from 10 to 20 kPa in the first stage to 70 to 80 kPa in the second stage.
Figure 4. Fouling issue during MPC. Flux and TMP of MPC reactor over time (A). MLSS and 10
viscosity of pre-concentrated sewage over time MPC (B) High solid concentration, which resulted in fast cake layer formation on the membrane surface, contributed to severe fouling during sewage pre-concentration. Mixed liquid suspended solids (MLSS) continuously increased from 2,000 mg/L to over 12,000 mg/L. Faust confirmed that fouling is caused by heavy solid concentration; this study proposed the use of submicron (45 to 450 nm) particles as specific fouling indicator [27]. The benefit of cake layer as pre-filter was also reported in previous studies [29, 30]. Cake layer in long-term membrane operation should be further investigated. To evaluate potential issues of MPC, chemical cleaning was not applied in the present study. Efficient membrane cleaning method (e.g., in-situ chemical cleaning [31]) can used for fouling control to facilitate further optimization. Figure 4 (B) showed that high MLSS induced high viscosity. The relationship between MLSS and viscosity was described by an exponential model. The rheological behaviors of concentrate were essential in the design and optimization of anaerobic digestion for energy recovery. High viscosity impeded the efficient mixing of concentrate feed during anaerobic biogas production. 3.2 Nitrogen recovery following COD pre-concentration by IE process Nitrogen removal and organic carbon removal were highly interrelated in CAS process. Nitrification and organic carbon removal consumed oxygen, which was provided by energy-intensive aeration process. De-nitrification consumed organic carbon for metabolism. The combined nitrification and denitrification process converts active nitrogen into N2 at the cost of consuming oxygen and organic carbon resources. In contrast, the proposed MPC–IE process was designed to treat organic carbon and nitrogen to achieve energy and nitrogen recovery.
11
Figure 5. MPC effects on NH4-N and TN removal (A); nitrogen in concentrate and removal rate during COD pre-concentration (B) The impact of COD pre-concentration on nitrogen flow was investigated. Figure 5 (A) showed that the average NH4-N before and after COD pre-concentration was 27.4±1.3 mg/L and 16.5±2.0 mg/L. This result indicated that only 60% NH4-N entered the nitrogen recovery process. The majority of nitrogen was concentrated during COD pre-concentration. This result was confirmed by increasing TN and NH4-N in the 12
concentrate. Figure 5 (B) showed that the concentrate TN and NH4-N increased from 34 mg/L and 24 mg/L to 198 mg/L and 75 mg/L, respectively. The effect of nitrification, which possibly occurred during COD pre-concentration, was not directly investigated in this study. According to Faust, nitrification was not detected when SRT was controlled at 1 to 5 d[27]. Although quite a large proportion of nitrogen was retained in the concentrate, the amount of recovered nitrogen could be improved through additional approaches. This part nitrogen could flow back to the membrane reactor in the form of rejected liquid digestate after the recovery of anaerobic energy. The removal rates of TN and NH4-N concentrates stabilized at 87.4±9.8 % and 74.4±9.1 %. This result indicated that additional nitrogen could be recovered after flow back of high levels of NH4-N liquid digestate. Another potential benefit of COD pre-concentration on nitrogen recovery was the nitrogen state conversion. During MPC, ammonification occurred, which converted organic nitrogen to NH4-N. Influent NH4 -N/TN was 0.80±0.05, whereas permeate NH4-N/TN was 0.88±0.08 after membrane-based COD pre-concentration. This process improved the efficiency of nitrogen recovery, since nitrogen recovery in this study was mainly achieved by cation-exchange resin.
13
Figure 6. Breakthrough curves (A) and resin regeneration for nitrogen recovery (B) □: Sample 1 was the actual permeate from membrane reactor. ○: Sample 2 was synthetic wastewater with 40 mg/L NH4-N. △:Sample 3 was synthetic wastewater with 40 mg/L NH4-N, 20 △: 2+ mg/L Ca , and 20 mg/L Mg2+ Ion exchange occurred with some ions trapping and some other ions releasing, which was widely used in water softening and purification. In this study, strong acid cation-exchange resin was used for trapping NH4-N and 10 g/L NaCl solution was used to regenerate resins by replacing NH4-N with Na+. Nitrogen recovery was achieved through obtaining high concentration NH4-N regenerant. Further process e.g. air stripping or crystallization could be applied according to nitrogen recovery requirements. IE adsorption and regeneration experiments were conducted. Both the actual sewage after COD pre-concentration and synthetic wastewater were applied in IE adsorption and regeneration experiments. This approach was adopted to evaluate the different composition effects (hard ions, including Ca2+, Mg2+, and other impurities) of sewage on IE for nitrogen recovery. Figure 6 (A) showed a remarkable competition between NH4-N and impure ions. Breakthrough occurred at 300 BV for synthetic NH4-N wastewater without hard ions in contrast to only 110 BV for actual sewage (even though its NH4-N concentration was low). Ca2+ and Mg2+ were the two main ions present in sewage at relatively high concentrations. The impacts of these particles on IE should be studied. Breakthrough at 300 BV decreased to about 200 BV when Ca2+ and Mg2+ were added to synthetic wastewater. This result indicates that Ca2+, Mg2+, and other impurities hindered IE for nitrogen. The characteristics of actual sewage before and after 14
IE were sampled and identified to confirm the influence of hard ion. As shown in Table 2, 101.9 mg/L Ca2+ and 33.87 mg/L Mg2+ were measured in actual sewage and almost completely removed by SAC resin. This result showed that SAC resins were more selective for hard ions than for ammonium. Al3+ was also measured because of coagulation during COD pre-concentration. Table 2. Characteristics of actual sewage before and after IE by SAC resin Cation
Actual sewage after COD pre-concentration (mg/L)
Actual sewage after IE by SAC resin (mg/L)
NH4+ Al3+ Ca2+ K+ Mg2+ Na+
27.3 0.0152 101.9 10.14 33.87 72
ND ND 0.0289 ND ND 303
Note: ND stands for no detection.
Similar results caused by selectivity for hard ions over ammonium were also observed in resin regeneration. Figure 6 (B) showed that the NH4-N concentration of regenerant was only 140 mg/L for actual sewage, which is lower than those for synthetic wastewater (1200 mg/L for Sample 2 and 1600 for Sample 3). Based on integral calculation, only 37.5% NH4 -N was recovered. Decreased efficiency of NH4-N IE removal was observed when hard ions were present. The result was consistent with the selectivity rows of reported SAC resin (Fe3+>Al3+>Pb2+>Ca2+>Cu2+>Zn2+>Mg2+>Mn2+>Ag+>NH4+>Na+>H+) [19, 32]. The results indicated existence of hard ions, especially Ca2+ and Mg2+, were prior to be exchanged by resins than NH4-N. Ca2+ and Mg2+ would occupy much of the ion exchange capacity of resins, thus lowering the adsorption amount of NH4 -N during IE process. Therefore, the influence of hard ions should be carefully avoided, and highly efficient N recovery technologies should be developed. Two strategies could be considered to eliminate this disadvantage. One strategy is to choose some NH4-N selective ion exchange materials which are prior to exchange NH4-N. Natural zeolite is an alternative because it is more selective for ammonium than for hard ions[33, 34]. However, the IE capacity of zeolite is lower than that of resin. The other strategy is to develop some pretreatment to remove hard ions before IE process. For example, capacitive deionization technology or selective ion exchange membrane could be an option to selectively remove hard ions (divalent ions)[35, 36]. Further research on such pretreatment should be carried out to improve nitrogen recovery efficiency. 15
3.3 Potential energy and nitrogen recovery by MPC–IE process Two scenarios based on the data obtained from this study and 100% ideal recovery were analyzed to evaluate the potential energy recovery of MPC–IE (Table 3). Energy recovery of 0.38 kWh/m3 can be achieved by combining the energy produced by AD (anaerobic digestion) process (0.26 kWh/m3) and that saved via N reuse (0.12 kWh/m3). A total of 0.89 kWh/m3 could be expected as the maximum in the ideal situation of 100% recovery. MPC-IE has high potential for energy recovery from sewage, which can address the average electrical consumption of 0.29 KWh/m3 in China [1]. Energy production by AD (0.26 kWh/m3) could nearly cover energy consumption. The influent sewage examined in this study has lower strength (260±32mg/L of COD) than that of typical wastewater (around 430 mg/L of COD). Increased energy recovery can be expected when high COD sewage is treated. Table 3. Potential energy recovery by MPC-IE process Scenario 1a
Scenario 2b
Recovered COD
65%
100%
Energy production by AD (kWh/m3) c
0.26
0.39
Recovered NH4 -N
37.5%
100%
Energy saved via N reuse (kWh/m3) d
0.12
0.31e/0.50 f
Total energy recovery (kWh/m3)
0.38
0.70/0.89
a
Scenario 1 was based on data in this study
b
Scenario 2 was based on 100% recovery
c
Estimated by 1 g of COD producing 350 mL of CH4[37]; 40% of methane energy can be converted into electrical energy [38]; energy yield
through 890 kJ/mol methane combustion d
Based on 19.3 kWh/kg N production energy by Haber−Bosch process
e
Estimated by 100% recovery of permeate NH4-N after COD pre-concentration
f
Estimated by 100% recovery of influent NH4-N
The MPC–IE process redesigned carbon flow during sewage treatment. More organic CODs were collected by MPC−IE than by the traditional process. Organic CODs in the form of concentrated state for AD energy production was essential for pursuing energy neutrality in WWTPs. The Strass WWTP in Austria improved its COD for AD from 40% to 60% through a high-load activated sludge system [8] and successfully achieved energy self-sufficiency in full-scale. High COD pre-concentration (higher than 65% ) was achieved in the present study. This result indicates the potential of COD pre-concentration as a 16
self-sufficient energy process. The MPC–IE process increased energy savings via N reuse, which was not considered in the current CAS process. The amount of energy saved could reach a remarkable level (0.50 kWh/m3 for sewage in this study assuming 100% influent NH4-N recovery). Given its potential, efficient N recovery technologies should be studied in future research. Energy consumption for MPC-IE, especially the membrane process, was decisive to achieve net energy production. The energy consumption for the lab-scale MPC–IE system was not measured. However, full-scale MPC–IE process is likely to be operated with limited energy consumption. MBRs (0.4 to 1.0 kWh/m3 specific energy consumption [14, 31]) shared many aspects with the membrane process of MPC–IE. The MPC–IE process can be operated with lesser energy than MBRs because the former does not require intensive aeration for membrane cleaning and oxygen supply for aerobic organic removal, thus showing it promise as an effective method for energy recovery from sewage. In the integrated MPC–IE process, MPC generally has the following advantages. 1) High proportion of organic matter is concentrated for improved energy recovery. 2) Suspended and colloidal solids are removed, which reduces the propensity for clogging in operation of fixed bed granular IE resins and thus, increasing nitrogen recovery effectiveness and longevity. 3) High water quality fit for reuse is guaranteed by combined membrane and IE process. 3. Conclusions This study proposed a novel concept for organics and nitrogen recovery from sewage through MPC–IE process. The feasibility of the concept was evaluated by MPC reactor for one month using actual sewage. Batch IE adsorption and regeneration experiments were conducted with membrane reactor effluent. Results showed that COD as high as 65% was recovered during steady membrane-based pre-concentration. MPC also removed suspended/colloidal solids and improved NH4-N/TN, which benefited IE nitrogen recovery. However, existence of hard ions decreased NH4-N IE efficiency, and only 37.5% NH4-N recovery was attained. Despite the low rate of recovery, the process could achieve a total of 0.38 kWh/m3 energy recovery by combining energy production with concentrates anaerobic digestion (0.26 kWh/m3) and saving via nitrogen reuse (0.12 kWh/m3). Optimization of the MPC-IE process is required to achieve highly efficient energy and nitrogen recovery from sewage in future.
Acknowledgements This work was supported by the Joint program of Beijing Natural Science Foundation and Beijing 17
Academy of Science and Technology (No.L140011) & Beijing Major transformation projects of scientific achievement (Key technology industrialization of Mega-ton desalination project).
18
References [1] X. Hao, R. Liu, X. Huang, Evaluation of the potential for operating carbon neutral WWTPs in China, Water research 87 (2015) 424-431. [2] USEPA, Wastewater Management Fact Sheet: Energy Conservation, Office of Water 2006, pp. EPA 832-F-806-024; U.S. Environmental Protection Agency: Washington DC, 2006; p 2007. [3] P.L. McCarty, J. Bae, J. Kim, Domestic Wastewater Treatment as a Net Energy Producer–Can This be Achieved?, Environmental science & technology 45 (2011) 7100-7106. [4] W. Rulkens, Increasing significance of advanced physical/chemical processes in the development and application of sustainable wastewater treatment systems, Frontiers of Environmental Science & Engineering in China 2 (2008) 385-396. [5] W. Verstraete, S.E. Vlaeminck, ZeroWasteWater: Short-cycling of wastewater resources for sustainable cities of the future, International Journal of Sustainable Development and World Ecology 18 (2011) 253-264. [6] B. Kartal, J.G. Kuenen, M.C.M.V. Loosdrecht, Sewage Treatment with Anammox, Science 328 (2010) 702-703. [7] D. Batstone, T. Hülsen, C. Mehta, J. Keller, Platforms for energy and nutrient recovery from domestic wastewater: A review, Chemosphere 140 (2015) 2-11. [8] B. Wett, K. Buchauer, C. Fimml, Energy self-sufficiency as a feasible concept for wastewater treatment systems,
IWA Leading Edge Technology Conference, Asian Water Singapore, 2007, pp. 21-24.
[9] A.R. Mels, A.F. van Nieuwenhuijzen, J.H.J.M. van der Graaf, B. Klapwijk, J. Koning, W.H. Rulkens, Sustainability criteria as a tool in the development of new sewage treatment methods, Water Science and Technology 39 (1999) 243-250. [10] A. Van Nieuwenhuijzen, J. van der Graaf, A. Mels, Direct influent filtration as a pretreatment step for more sustainable wastewater treatment systems, Water science and technology: a journal of the International Association on Water Pollution Research 43 (2001) 91. [11] H. Gong, Z. Jin, X. Wang, K. Wang, Membrane fouling controlled by coagulation/adsorption during direct sewage membrane filtration (DSMF) for organic matter concentration, Journal of Environmental Sciences 32 (2015) 1-7. [12] Z. Jin, H. Gong, K. Wang, Application of hybrid coagulation microfiltration with air backflushing to direct sewage concentration for organic matter recovery, Journal of Hazardous Materials 283 (2015) 824-831. [13] H. Gong, X. Wang, M. Zheng, Z. Jin, K. Wang, Direct sewage filtration for concentration of organic matters by dynamic membrane, Water Science & Technology 70 (2014) 1434-1440. [14] Z. Jin, H. Gong, H. Temmink, H. Nie, J. Wu, J. Zuo, K. Wang, Efficient sewage pre-concentration with combined coagulation microfiltration for organic matter recovery, Chemical Engineering Journal 292 (2016) 130–138. [15] I. Akanyeti, H. Temmink, M. Remy, A. Zwijnenburg, Feasibility of bioflocculation in a high-loaded membrane bioreactor for improved energy recovery from sewage, Water Science & Technology 61 (2010). [16] G. Mezohegyi, M.R. Bilad, I.F.J. Vankelecom, Direct sewage up-concentration by submerged aerated and vibrated membranes, Bioresource Technology 118 (2012) 1-7. [17] S. Leakovic, I. Mijatovic, S. Cerjan-Stefanovic, E. Hodzic, Nitrogen removal from fertilizer wastewater by ion exchange, Water research 34 (2000) 185-190. [18] H. Zhang, A. Li, W. Zhang, C. Shuang, Combination of Na-modified zeolite and anion exchange resin for advanced treatment of a high ammonia-nitrogen content municipal effluent, Journal of Colloid and Interface 19
Science 468 (2016) 128-135. [19] A. Malovanyy, H. Sakalova, Y. Yatchyshyn, E. Plaza, M. Malovanyy, Concentration of ammonium from municipal wastewater using ion exchange process, Desalination 329 (2013) 93-102. [20] C. NEPA, Water and Wastewater Monitoring Methods, 4th ed., Chinese Environmental Science Publishing House, Beijing, China, 2002. [21] A. Ravazzini, A. Van Nieuwenhuijzen, J. Van Der Graaf, Direct ultrafiltration of municipal wastewater: comparison between filtration of raw sewage and primary clarifier effluent, Desalination 178 (2005) 51-62. [22] Q. Zhao, H. Zhong, K. Wang, L. Wei, J. Liu, Y. Liu, Removal and transformation of organic matters in domestic wastewater during lab-scale chemically enhanced primary treatment and a trickling filter treatment, Journal of Environmental Sciences 25 (2013) 59-68. [23] G. Mouri, S. Takizawa, K. Fukushi, T. Oki, Estimation of the effects of chemically-enhanced treatment of urban sewage system based on life-cycle management, Sustainable Cities and Society 9 (2013) 23-31. [24] S. Luostarinen, W. Sanders, K. Kujawa-Roeleveld, G. Zeeman, Effect of temperature on anaerobic treatment of black water in UASB-septic tank systems, Bioresource Technology 98 (2007) 980-986. [25] M.K. Sharma, A.A. Kazmi, Anaerobic onsite treatment of black water using filter-based packaged system as an alternative of conventional septic tank, Ecological Engineering 75 (2015) 457-461. [26] G. Lettinga, J.B. Van Lier, J.C.L. Van Buuren, G. Zeeman, Sustainable development in pollution control and the role of anaerobic treatment, Water Science and Technology 44 (2001) 181-188. [27] L. Faust, H. Temmink, A. Zwijnenburg, A. Kemperman, H. Rijnaarts, High loaded MBRs for organic matter recovery from sewage: effect of solids retention time on bioflocculation and on the role of extracellular polymers, Water research 56 (2014) 258-266. [28] L. Faust, H. Temmink, A. Zwijnenburg, A. Kemperman, H. Rijnaarts, Effect of dissolved oxygen concentration on the bioflocculation process in high loaded MBRs, Water research 66 (2014) 199-207. [29] Z. Cai, M.M. Benjamin, NOM fractionation and fouling of low-pressure membranes in microgranular adsorptive filtration, Environmental science & technology 45 (2011) 8935-8940. [30] Z. Cai, C. Wee, M.M. Benjamin, Fouling mechanisms in low-pressure membrane filtration in the presence of an adsorbent cake layer, Journal of Membrane Science 433 (2013) 32-38. [31] S.K. Lateef, B.Z. Soh, K. Kimura, Direct membrane filtration of municipal wastewater with chemically enhanced backwash for recovery of organic matter, Bioresource Technology 150 (2013) 149-155. [32] D. Alchin, Ion exchange resins, Chemical Processes in New Zealand, New Zealand Institute of Chemistry Education, pp. XIII-D-1–XIII-D-71998. [33] H. Huang, D. Xiao, R. Pang, C. Han, D. Li, Simultaneous removal of nutrients from simulated swine wastewater by adsorption of modified zeolite combined with struvite crystallization, Chemical Engineering Journal 256 (2014) 431–438. [34] A. Farkas, M. Rozić, Z. Barbarić-Mikocević, Ammonium exchange in leakage waters of waste dumps using natural zeolite from the Krapina region, Croatia, Journal of Hazardous Materials 117 (2005) 25-33. [35] S. Seokjun, J. Hongrae, L. Jaekwang, K. Ghayoung, P. Daewook, H. Nojima, L. Jaeyoung, M. Seunghyeon, Investigation on removal of hardness ions by capacitive deionization (CDI) for water softening applications, Water research 44 (2010) 2267-2275. [36] Z. Chen, H. Zhang, C. Wu, Y. Wang, W. Li, A study of electrosorption selectivity of anions by activated carbon electrodes in capacitive deionization, Desalination 369 (2015) 46-50. [37] I. Angelidaki, W. Sanders, Assessment of the anaerobic biodegradability of macropollutants, Re/Views in Environmental Science & Bio/Technology 3 (2004) 117-129. [38] J.B. Van Lier, High-rate anaerobic wastewater treatment: diversifying from end-of-the-pipe treatment to 20
resource-oriented conversion techniques, Water Science and Technology 57 (2008) 1137-1148.
21
22
Highlights: 1. Concept for simultaneous recovery of energy and nitrogen from sewage was proposed. 2. Membrane-based preconcentration(MPC) was combined with ion exchange(IE) process. 3. COD was recovered by MPC to achieve energy recovery. 4. Presentation of hard ions (Ca2+ & Mg2+) decreased NH4-N IE recovery efficiency.
23
S1385-8947(16)31630-8 http://dx.doi.org/10.1016/j.cej.2016.11.068 CEJ 16069
To appear in:
Chemical Engineering Journal
Received Date: Revised Date: Accepted Date:
20 September 2016 7 November 2016 8 November 2016
Please cite this article as: H. Gong, Z. Wang, X. Zhang, Z. Jin, C. Wang, L. Zhang, K. Wang, Organics and nitrogen recovery from sewage via membrane-based pre-concentration combined with ion exchange process, Chemical Engineering Journal (2016), doi: http://dx.doi.org/10.1016/j.cej.2016.11.068
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Organics and nitrogen recovery from sewage via membrane-based pre-concentration combined with ion exchange process
Hui Gong1, Zhijie Wang1, Xue Zhang2, Zhengyu Jin1, Cuiping Wang1, Liping Zhang2, Kaijun Wang1,* 1．State Key Joint Laboratory of Environment Simulation and Pollution Control, School of Environment, Tsinghua University, Beijing 100084, China 2．School of Chemical & Environmental Engineering, China University of Mining & Technology, Beijing 100083, China
Submitted to Chemical Engineering Journal
* Corresponding Author Email:
[email protected]
Phone: (+86) 010-62789411
1
Abstract: This study proposes the recovery of organics and nitrogen from sewage through membrane-based pre-concentration (MPC) combined with ion exchange (IE) process. Unlike conventional activated sludge process, MPC–IE redesigned organic carbon flow to increase chemical oxygen demand (COD) conversion for energy recovery via anaerobic digestion (AD). This process also achieved nitrogen recovery instead of destruction. Membrane-based pre-concentration of COD was conducted using actual sewage for one month. As high as 65% COD was recovered. Batch IE adsorption and regeneration experiments were conducted. However, these experiments recovered only 37.5% NH4-N because of the selectivity of IE for hard ions over ammonium. Despite the low rate of recovery, the process could achieve a total of 0.38 kWh/m3 energy recovery by combining energy production with anaerobic digestion of pre-concentrated organics (0.26 kWh/m3) and energy saving via nitrogen reuse (0.12 kWh/m3). Highly efficient energy and nitrogen recovery from sewage via MPC–IE process was expected after optimization.
Keywords: Sewage; Nitrogen recovery; Energy recovery; Membrane-based pre-concentration; Ion exchange
2
1. Introduction In response to global resource shortage, domestic wastewater should not be treated as waste anymore, but a resource of water, energy, and plant fertilizing nutrients (nitrogen and phosphorus). However, the objective of sustainable resource recovery from domestic wastewater is far from being achieved at present, which is limited by the disadvantages of conventionally activated sludge (CAS). CAS and its modification processes require high-energy consumption during aeration to supply oxygen for organic matter removal by aerobic microorganism in activated sludge. The anaerobic digestion of sludge can offset part of energy requirement, but only a portion of energy potential in organics is captured. Wastewater treatment now accounts for 0.3% and 3% of the total electrical energy load in China and the U.S., respectively [1, 2]. Nitrogen, mostly in form of N2 , is removed and not recovered during CAS process. Water reuse is widely practiced in water shortage regions, but this also increases energy consumption because of long process to satisfy high-quality requirements for water reuse. Therefore, novel short-procedure technologies/processes for resource recovery are in demand. Novel concepts and processes have been proposed in recent years to promote paradigm shift to recovering resources from domestic wastewater. McCarty et al. discussed the potential of wastewater treatment plant (WWTP) as net energy producer [3]. Rulkens emphasized increasing significance of advanced physical/chemical processes to develop sustainable treatment systems[4]. Verstraete et al. proposed a zero-wastewater concept with up-concentration process, followed by organic anaerobic digestion to achieve maximum energy recovery in domestic used water [5]. Kartal et al. proposed a novel sewage treatment process with increased energy recovery by organics bio-adsorption and low energy consumption by anammox [6]. Batstone et al. suggested partition–release–recover concept in which carbon and nutrients are partitioned to solids through either heterotrophic or phototrophic microbes followed by anaerobic digestion[7]. This study proposed the concept of membrane-based pre-concentration (MPC) combined with ion exchange (IE) process to achieve complete recovery of water, energy, and nutrients (nitrogen). Sewage enters MPC reactor where organic matter is flocculated and separated by membrane filtration. Adsorbent (powder-activated carbon) is also added to enhance soluble organic matter removal and to alleviate membrane fouling issue. After MPC, most organic matter is separated from the mainstream and presented in concentrated state, which could obtain energy recovery in form of biogas (methane) generation by organic anaerobic digestion. The MPC effluent (alone or combined with digester effluent) enters the IE 3
process to recover nitrogen in the form of NH4-N. The chemical oxygen demand (COD) pre-concentration is essential to improve energy recovery and to achieve energy neutrality. A WWTP in Strass, Austria demonstrates large-scale feasibility in energy self-sufficiency by increasing the transfer of less-stabilized organics (as high as 60% COD) from the mainstream sewage to anaerobic digesters in form of sludge [8]. Research on highly efficient CODpre-concentration was needed to improve proportion of concentrated organic matters for enhanced energy recovery in anaerobic digestion. Mels and Van Nieuwenhuijzen evaluated physical−chemical technologies including pre-precipitation and flotation, whose efficiency on soluble COD concentration, however, is low [9, 10]. MPC was proposed in our previous studies [11-14]. The high-separation efficiency suggested that membrane technology as a promising tool for organic recovery. Akanyeti demonstrated high-load membrane biological reactor (MBR) for efficient organic concentration in grey water through bio-flocculation[15]. Mezohegyi evaluated vibrating membranes for sewage concentration utilizing vibration to overcome fouling issue [16] . Since
Haber-Bosch process invention in 1908 enabling human society producing reactive nitrogen
in large amounts and low price, overloaded anthropogenic reactive nitrogen has end up in environments. Nitrogen recovery could significantly benefit from reduced loading in local environments and nitrogen cycle. However, nitrogen recovery is ignored in conventional wastewater treatment process (e.g., CAS) whose nitrogen control technologies are based on destructive technologies. Actually, nitrogen recovery potentialis remarkable given the increasing domestic wastewater production caused by population growth and economic development. Nitrogen in domestic wastewater accounts for 2% to 8% of total nitrogen input in China (estimated by national sewage treatment capacity of 1.5 × 108 m3/d [1]). IE is commonly used to recover nitrogen, but mostly for high-ammonium wastewater (200 to 8000 mg/L TN) rather than low-strength domestic wastewater (20 to 80 mg/L TN) [17-19]. IE performance for nitrogen recovery, as well as MPC, has not been investigated in previous studies. The MPC–IE process was proposed and investigated for the first time. In this study, we operated PC reactor with actual sewage for one month and conducted batch IE adsorption and regeneration experiments with membrane reactor effluent. The aim of this paper is to investigate the feasibility of the MPC−IE process by determining (i) organic recovery by MPC, (ii) the advantages and disadvantages of nitrogen recovery by IE resins, and (iii) performance of MPC–IE interaction. 2． ．Materials and methods 4
2.1 Membrane-based COD pre-concentration 2.1.1 Raw sewage Raw sewage for MPC experiment was collected after grid in Xiaojiahe municipal WWTP in Beijing, China. The general characteristics were listed in Table 1. The average COD and NH4-N of fresh sewage for testing period were 260 mg/L and 27.4 mg/L, respectively. Table 1. Average raw sewage quality of COD, TN, TP, NH4-N, and turbidity for MPC
Sewage
COD
TN
TP
NH4-N
Turbidity
(mg/L)
(mg/L)
(mg/L)
(mg/L)
(NTU)
260±32
34.5±1.7
2.78±0.38
27.4±1.3
114±13
2.1.2 Membrane reactor and experimental setup The membrane reactor was designed for sewage preconcentration. Figure 1 showed the schematic diagram of the setup of membrane reactor. The effect volume of filtration tank was 3.5 L, and the dimensions were 430 mm long, 250 mm wide, and 60 mm high. The filtration tank contained one-membrane module with a total of 0.33 m2 membrane surface. Hollow-fiber module was selected because of its large-membrane-surface area-volume ratio. Commercial hollow-fiber polyvinylidene fluoride (PVDF) microfiltration membrane (Origin Water Co. Ltd. China) with average pore size of 0.3 µm was applied in this study. The inner and outer diameters of the membrane fiber were 0.95 and 2.45 mm, respectively. The PVDF membrane module was stored in tap water before operation. Membrane filtration experiment was conducted in outside-in mode. Pressure gauge, flow meter, and paperless recorder were used for automatic monitoring and recording of flux and transmembrane pressure (TMP). Raw sewage was fed into the reactor. Polyaluminum chloride (Al2(OH)nCl6-n]m) measuring 30 mg/L and 10 mg/L power active carbon (iodine value = above 1,000; 0.45 to 0.55 g/cm3 apparent density) were added according to previous study on coagulant selection. A piston pump (Seko, Italy) was used to draw out permeate from the membrane module. Concentrates were retained in the reactor. Initial filtration flux was set to 20 L/m2 h. Liquid level in the reactor was kept constant by peristaltic pump and raw sewage was fed automatically. To avoid severe membrane fouling, the membrane module underwent permeation for 10 mins followed by a 2-min relaxation. During the 2-min relaxation, back-flushing aeration was activated to remove cakes on membrane surface. Back-flushing lasted for 50 s at a pressure of 50 kPa. The permeated and concentrated COD and NH4-N of influent raw sewage were sampled for analysis every 48 h. Given the settling mechanism of particulate COD in the reactor tank, short aeration was provided (1 to 2 mins) to mix 5
the tanks when sampling for COD concentration performance evaluation. The MPC process with two stages lasted for 600 h. The first stage (first 360 h) did not exhibit concentrates discharge. This result enabled the organic matters in sewage to be concentrated quickly in the reactor (except for the 200 ml concentrates which was sampled every 48 h). The concentrate started to be discharged with sludge retention time (SRT) for 3.5 d in the second stage (360 to 600 h) when COD increased to over 9,500 mg/L. Chemical cleaning was not applied during the experiment.
Figure 1. Schematic diagram of MPC (left) and IE resin reactor for nitrogen recovery (right) 2.2 IE for nitrogen recovery 2.2.1 Feed for IE Three wastewater samples were applied to evaluate the effects of the different compositions (hard ions, including Ca2+, Mg2+, and other impurities) of sewage in IE for nitrogen recovery. Sample 1 was the permeate from the membrane reactor. Sample 2 was synthetic wastewater with 40 mg/L NH4-N. Sample 3 was synthetic wastewater with 40 mg/L NH4-N, 20 mg/L Ca2+ and 20 mg/L Mg2+. 2.2.2 IE bed-column setup As shown in Figure 1 (right), IE for nitrogen recovery was investigated by organic glass-packed bed column with an inner diameter of 24 mm and length of 120 mm. The column was filled with 54 mL strong acid cation-exchange resin (SAC; 001×7 resin; particle size 0.5 to 1 mm, XinKe Co. Ltd, Jinzhou, China). Before experiments, SAC resin was transformed to Na-form by pumping 10 g/L of NaCl solution at flow rate of 5.6 mL/min for 4 hours. Two processes of resin adsorption and regeneration were conducted. During resin adsorption, wastewater containing NH4-N was pumped into the column from the bottom. Treated wastewater in the outlet was sampled regularly. During resin regeneration, exhausted resin was regenerated by 10 g/L NaCl solution. Nitrogen recovery was achieved during regeneration. Concentrated 6
NH4-N content in spent regenerant was also measured regularly. 2.3 Analysis methods COD and NH4-N were determined by colorimetric techniques using HACH spectrophotometer (DR 5000, HACH, USA). SS was determined according to the Chinese NEPA Standard Methods [20]; pH was measured using a pH meter (Sension1, HACH, USA); and turbidity was measured using a turbidity meter (2100Q, HACH, USA). 3 Results and discussion 3.1 Performance of membrane-based COD pre-concentration 3.1.1 Pre-concentrated sewage by membrane process for improved energy recovery Organic matters (measured as COD) in sewage were presented in soluble, colloid and suspended state. To recover organics for energy recovery, MPC utilized membrane process separating flocculated colloid and suspended organics. Considering the separation of soluble organic matters from mainstream sewage, powder-activated carbon was added as adsorbent, which also was expected to alleviate membrane fouling issue. Enhanced COD removal was observed during sewage pre-concentration. Permeate COD was 19±3 mg/L with 260±32 mg/L raw sewage COD. The rate of COD removal was as high as 92.9±1.2 %. MPC has higher rate of COD removal than that in processes with direct membrane filtration (36% to 75%) [21] or coagulation alone (e.g., chemically enhanced primary treatment with COD removal of 46% to 50%) [22, 23]. This finding was attributed to combined membrane filtration and coagulation, which was beneficial for efficient sewage organics concentration.
7
Figure 2. COD of influent, permeate and concentrate of MPC reactor over time Figure 2 showed that efficient sewage pre-concentration was achieved. The COD of influent raw sewage was not high (260±32mg/L), but concentrated COD increased to 9640 mg/L in 360 h in the first stage. Concentrate started to be discharged with SRT at 3.5 d in the second stage (360 to 600 h). Concentrated COD was stable in the range of 9,000 to 10,000 mg/L. The increased level of concentrated COD in the second stage was as high as, even higher than that of black water [24, 25]. Black water was domestic wastewater originally distinguished from toilets (faeces, urine and flushing water) which contained concentrated high organic matters (COD) and was usually anaerobically treated. Thus, the pre-concentrated COD in this research can be anaerobically treated with energy recovery by converting organic matters into CH4. Moreover, the more concentrated organics the wastewater contained, the smaller the needed reactor size and space requirements was, and accordingly, the lower the manufacturing cost was[26].
8
Figure 3. COD balance of different stages during MPC (Left: first stage, 0 to 360 h; Right: second stage, 360 to 600 h) As shown in Figure 3, mass balance was calculated to quantify the efficiency of pre-concentration of sewage COD. Three categories, namely, permeated, concentrated, and others, were applied to represent COD discharged in permeate, concentrated in concentrate, and those that do not belong to the first two categories. Others were calculated based on total mass balance of COD. This category contained two parts. One part was mineralized COD, which was converted to CO2 through biochemical process, whereas the other part was those deposited in the membrane surface or in reactor corners (hydraulically dead zones). Permeated + mineralized represented the unrecoverable part, which was lost during pre-concentration. Concentrated + deposited denoted the recoverable part, which could be applied for energy recovery in post treatment. Concentrated COD in concentrate is limited in the first stage because large amount of concentrate was not discharged continuously. The majority of COD were concentrated in others, which was retained in the reactor. The amount of Mineralized, which was proportional to SRT, could be high. Faust found that the proportion of mineralized could increase to as high as 32% at 5 d SRT [27, 28]. This result was likely caused by the accumulated heterotrophic microorganisms in the reactor, which easily utilized biodegradable COD supplied with raw sewage. The first stage period should be controlled to a minimum to limit loss by mineralization and to obtain concentrate continuously. In the second stage, 65% COD was collected in the concentrate. The proportion of recoverable COD would increase if the COD retained in the reactor was considered. The proportion of Mineralized decreased with SRT. Faust demonstrated that the proportion of mineralized COD decreased rapidly with short SRT. Estimated COD loss by mineralization was controlled to 1% to 2% of the total amount by setting SRT at 9
0.125 to 0.25 d [27]. This result indicates that optimization by shortening SRT should be carefully considered in future research to improve the efficiency of COD pre-concentration. 3.1.2 Fouling issue during MPC Membrane fouling was important in determining the efficiency of sewage organics pre-concentration. High water flux would shorten the first stage and guarantee high level of concentrated COD when SRT was reduced. However, water flux in pre-concentration was undermined by membrane fouling. Figure 4 (A) showed that flux was initially set at 20 LHM and kept constant in the first 300 h. Flux started to decrease at the end of the first stage and finally stabilized at 5 to 10 LHM. TMP jumped from 10 to 20 kPa in the first stage to 70 to 80 kPa in the second stage.
Figure 4. Fouling issue during MPC. Flux and TMP of MPC reactor over time (A). MLSS and 10
viscosity of pre-concentrated sewage over time MPC (B) High solid concentration, which resulted in fast cake layer formation on the membrane surface, contributed to severe fouling during sewage pre-concentration. Mixed liquid suspended solids (MLSS) continuously increased from 2,000 mg/L to over 12,000 mg/L. Faust confirmed that fouling is caused by heavy solid concentration; this study proposed the use of submicron (45 to 450 nm) particles as specific fouling indicator [27]. The benefit of cake layer as pre-filter was also reported in previous studies [29, 30]. Cake layer in long-term membrane operation should be further investigated. To evaluate potential issues of MPC, chemical cleaning was not applied in the present study. Efficient membrane cleaning method (e.g., in-situ chemical cleaning [31]) can used for fouling control to facilitate further optimization. Figure 4 (B) showed that high MLSS induced high viscosity. The relationship between MLSS and viscosity was described by an exponential model. The rheological behaviors of concentrate were essential in the design and optimization of anaerobic digestion for energy recovery. High viscosity impeded the efficient mixing of concentrate feed during anaerobic biogas production. 3.2 Nitrogen recovery following COD pre-concentration by IE process Nitrogen removal and organic carbon removal were highly interrelated in CAS process. Nitrification and organic carbon removal consumed oxygen, which was provided by energy-intensive aeration process. De-nitrification consumed organic carbon for metabolism. The combined nitrification and denitrification process converts active nitrogen into N2 at the cost of consuming oxygen and organic carbon resources. In contrast, the proposed MPC–IE process was designed to treat organic carbon and nitrogen to achieve energy and nitrogen recovery.
11
Figure 5. MPC effects on NH4-N and TN removal (A); nitrogen in concentrate and removal rate during COD pre-concentration (B) The impact of COD pre-concentration on nitrogen flow was investigated. Figure 5 (A) showed that the average NH4-N before and after COD pre-concentration was 27.4±1.3 mg/L and 16.5±2.0 mg/L. This result indicated that only 60% NH4-N entered the nitrogen recovery process. The majority of nitrogen was concentrated during COD pre-concentration. This result was confirmed by increasing TN and NH4-N in the 12
concentrate. Figure 5 (B) showed that the concentrate TN and NH4-N increased from 34 mg/L and 24 mg/L to 198 mg/L and 75 mg/L, respectively. The effect of nitrification, which possibly occurred during COD pre-concentration, was not directly investigated in this study. According to Faust, nitrification was not detected when SRT was controlled at 1 to 5 d[27]. Although quite a large proportion of nitrogen was retained in the concentrate, the amount of recovered nitrogen could be improved through additional approaches. This part nitrogen could flow back to the membrane reactor in the form of rejected liquid digestate after the recovery of anaerobic energy. The removal rates of TN and NH4-N concentrates stabilized at 87.4±9.8 % and 74.4±9.1 %. This result indicated that additional nitrogen could be recovered after flow back of high levels of NH4-N liquid digestate. Another potential benefit of COD pre-concentration on nitrogen recovery was the nitrogen state conversion. During MPC, ammonification occurred, which converted organic nitrogen to NH4-N. Influent NH4 -N/TN was 0.80±0.05, whereas permeate NH4-N/TN was 0.88±0.08 after membrane-based COD pre-concentration. This process improved the efficiency of nitrogen recovery, since nitrogen recovery in this study was mainly achieved by cation-exchange resin.
13
Figure 6. Breakthrough curves (A) and resin regeneration for nitrogen recovery (B) □: Sample 1 was the actual permeate from membrane reactor. ○: Sample 2 was synthetic wastewater with 40 mg/L NH4-N. △:Sample 3 was synthetic wastewater with 40 mg/L NH4-N, 20 △: 2+ mg/L Ca , and 20 mg/L Mg2+ Ion exchange occurred with some ions trapping and some other ions releasing, which was widely used in water softening and purification. In this study, strong acid cation-exchange resin was used for trapping NH4-N and 10 g/L NaCl solution was used to regenerate resins by replacing NH4-N with Na+. Nitrogen recovery was achieved through obtaining high concentration NH4-N regenerant. Further process e.g. air stripping or crystallization could be applied according to nitrogen recovery requirements. IE adsorption and regeneration experiments were conducted. Both the actual sewage after COD pre-concentration and synthetic wastewater were applied in IE adsorption and regeneration experiments. This approach was adopted to evaluate the different composition effects (hard ions, including Ca2+, Mg2+, and other impurities) of sewage on IE for nitrogen recovery. Figure 6 (A) showed a remarkable competition between NH4-N and impure ions. Breakthrough occurred at 300 BV for synthetic NH4-N wastewater without hard ions in contrast to only 110 BV for actual sewage (even though its NH4-N concentration was low). Ca2+ and Mg2+ were the two main ions present in sewage at relatively high concentrations. The impacts of these particles on IE should be studied. Breakthrough at 300 BV decreased to about 200 BV when Ca2+ and Mg2+ were added to synthetic wastewater. This result indicates that Ca2+, Mg2+, and other impurities hindered IE for nitrogen. The characteristics of actual sewage before and after 14
IE were sampled and identified to confirm the influence of hard ion. As shown in Table 2, 101.9 mg/L Ca2+ and 33.87 mg/L Mg2+ were measured in actual sewage and almost completely removed by SAC resin. This result showed that SAC resins were more selective for hard ions than for ammonium. Al3+ was also measured because of coagulation during COD pre-concentration. Table 2. Characteristics of actual sewage before and after IE by SAC resin Cation
Actual sewage after COD pre-concentration (mg/L)
Actual sewage after IE by SAC resin (mg/L)
NH4+ Al3+ Ca2+ K+ Mg2+ Na+
27.3 0.0152 101.9 10.14 33.87 72
ND ND 0.0289 ND ND 303
Note: ND stands for no detection.
Similar results caused by selectivity for hard ions over ammonium were also observed in resin regeneration. Figure 6 (B) showed that the NH4-N concentration of regenerant was only 140 mg/L for actual sewage, which is lower than those for synthetic wastewater (1200 mg/L for Sample 2 and 1600 for Sample 3). Based on integral calculation, only 37.5% NH4 -N was recovered. Decreased efficiency of NH4-N IE removal was observed when hard ions were present. The result was consistent with the selectivity rows of reported SAC resin (Fe3+>Al3+>Pb2+>Ca2+>Cu2+>Zn2+>Mg2+>Mn2+>Ag+>NH4+>Na+>H+) [19, 32]. The results indicated existence of hard ions, especially Ca2+ and Mg2+, were prior to be exchanged by resins than NH4-N. Ca2+ and Mg2+ would occupy much of the ion exchange capacity of resins, thus lowering the adsorption amount of NH4 -N during IE process. Therefore, the influence of hard ions should be carefully avoided, and highly efficient N recovery technologies should be developed. Two strategies could be considered to eliminate this disadvantage. One strategy is to choose some NH4-N selective ion exchange materials which are prior to exchange NH4-N. Natural zeolite is an alternative because it is more selective for ammonium than for hard ions[33, 34]. However, the IE capacity of zeolite is lower than that of resin. The other strategy is to develop some pretreatment to remove hard ions before IE process. For example, capacitive deionization technology or selective ion exchange membrane could be an option to selectively remove hard ions (divalent ions)[35, 36]. Further research on such pretreatment should be carried out to improve nitrogen recovery efficiency. 15
3.3 Potential energy and nitrogen recovery by MPC–IE process Two scenarios based on the data obtained from this study and 100% ideal recovery were analyzed to evaluate the potential energy recovery of MPC–IE (Table 3). Energy recovery of 0.38 kWh/m3 can be achieved by combining the energy produced by AD (anaerobic digestion) process (0.26 kWh/m3) and that saved via N reuse (0.12 kWh/m3). A total of 0.89 kWh/m3 could be expected as the maximum in the ideal situation of 100% recovery. MPC-IE has high potential for energy recovery from sewage, which can address the average electrical consumption of 0.29 KWh/m3 in China [1]. Energy production by AD (0.26 kWh/m3) could nearly cover energy consumption. The influent sewage examined in this study has lower strength (260±32mg/L of COD) than that of typical wastewater (around 430 mg/L of COD). Increased energy recovery can be expected when high COD sewage is treated. Table 3. Potential energy recovery by MPC-IE process Scenario 1a
Scenario 2b
Recovered COD
65%
100%
Energy production by AD (kWh/m3) c
0.26
0.39
Recovered NH4 -N
37.5%
100%
Energy saved via N reuse (kWh/m3) d
0.12
0.31e/0.50 f
Total energy recovery (kWh/m3)
0.38
0.70/0.89
a
Scenario 1 was based on data in this study
b
Scenario 2 was based on 100% recovery
c
Estimated by 1 g of COD producing 350 mL of CH4[37]; 40% of methane energy can be converted into electrical energy [38]; energy yield
through 890 kJ/mol methane combustion d
Based on 19.3 kWh/kg N production energy by Haber−Bosch process
e
Estimated by 100% recovery of permeate NH4-N after COD pre-concentration
f
Estimated by 100% recovery of influent NH4-N
The MPC–IE process redesigned carbon flow during sewage treatment. More organic CODs were collected by MPC−IE than by the traditional process. Organic CODs in the form of concentrated state for AD energy production was essential for pursuing energy neutrality in WWTPs. The Strass WWTP in Austria improved its COD for AD from 40% to 60% through a high-load activated sludge system [8] and successfully achieved energy self-sufficiency in full-scale. High COD pre-concentration (higher than 65% ) was achieved in the present study. This result indicates the potential of COD pre-concentration as a 16
self-sufficient energy process. The MPC–IE process increased energy savings via N reuse, which was not considered in the current CAS process. The amount of energy saved could reach a remarkable level (0.50 kWh/m3 for sewage in this study assuming 100% influent NH4-N recovery). Given its potential, efficient N recovery technologies should be studied in future research. Energy consumption for MPC-IE, especially the membrane process, was decisive to achieve net energy production. The energy consumption for the lab-scale MPC–IE system was not measured. However, full-scale MPC–IE process is likely to be operated with limited energy consumption. MBRs (0.4 to 1.0 kWh/m3 specific energy consumption [14, 31]) shared many aspects with the membrane process of MPC–IE. The MPC–IE process can be operated with lesser energy than MBRs because the former does not require intensive aeration for membrane cleaning and oxygen supply for aerobic organic removal, thus showing it promise as an effective method for energy recovery from sewage. In the integrated MPC–IE process, MPC generally has the following advantages. 1) High proportion of organic matter is concentrated for improved energy recovery. 2) Suspended and colloidal solids are removed, which reduces the propensity for clogging in operation of fixed bed granular IE resins and thus, increasing nitrogen recovery effectiveness and longevity. 3) High water quality fit for reuse is guaranteed by combined membrane and IE process. 3. Conclusions This study proposed a novel concept for organics and nitrogen recovery from sewage through MPC–IE process. The feasibility of the concept was evaluated by MPC reactor for one month using actual sewage. Batch IE adsorption and regeneration experiments were conducted with membrane reactor effluent. Results showed that COD as high as 65% was recovered during steady membrane-based pre-concentration. MPC also removed suspended/colloidal solids and improved NH4-N/TN, which benefited IE nitrogen recovery. However, existence of hard ions decreased NH4-N IE efficiency, and only 37.5% NH4-N recovery was attained. Despite the low rate of recovery, the process could achieve a total of 0.38 kWh/m3 energy recovery by combining energy production with concentrates anaerobic digestion (0.26 kWh/m3) and saving via nitrogen reuse (0.12 kWh/m3). Optimization of the MPC-IE process is required to achieve highly efficient energy and nitrogen recovery from sewage in future.
Acknowledgements This work was supported by the Joint program of Beijing Natural Science Foundation and Beijing 17
Academy of Science and Technology (No.L140011) & Beijing Major transformation projects of scientific achievement (Key technology industrialization of Mega-ton desalination project).
18
References [1] X. Hao, R. Liu, X. Huang, Evaluation of the potential for operating carbon neutral WWTPs in China, Water research 87 (2015) 424-431. [2] USEPA, Wastewater Management Fact Sheet: Energy Conservation, Office of Water 2006, pp. EPA 832-F-806-024; U.S. Environmental Protection Agency: Washington DC, 2006; p 2007. [3] P.L. McCarty, J. Bae, J. Kim, Domestic Wastewater Treatment as a Net Energy Producer–Can This be Achieved?, Environmental science & technology 45 (2011) 7100-7106. [4] W. Rulkens, Increasing significance of advanced physical/chemical processes in the development and application of sustainable wastewater treatment systems, Frontiers of Environmental Science & Engineering in China 2 (2008) 385-396. [5] W. Verstraete, S.E. Vlaeminck, ZeroWasteWater: Short-cycling of wastewater resources for sustainable cities of the future, International Journal of Sustainable Development and World Ecology 18 (2011) 253-264. [6] B. Kartal, J.G. Kuenen, M.C.M.V. Loosdrecht, Sewage Treatment with Anammox, Science 328 (2010) 702-703. [7] D. Batstone, T. Hülsen, C. Mehta, J. Keller, Platforms for energy and nutrient recovery from domestic wastewater: A review, Chemosphere 140 (2015) 2-11. [8] B. Wett, K. Buchauer, C. Fimml, Energy self-sufficiency as a feasible concept for wastewater treatment systems,
IWA Leading Edge Technology Conference, Asian Water Singapore, 2007, pp. 21-24.
[9] A.R. Mels, A.F. van Nieuwenhuijzen, J.H.J.M. van der Graaf, B. Klapwijk, J. Koning, W.H. Rulkens, Sustainability criteria as a tool in the development of new sewage treatment methods, Water Science and Technology 39 (1999) 243-250. [10] A. Van Nieuwenhuijzen, J. van der Graaf, A. Mels, Direct influent filtration as a pretreatment step for more sustainable wastewater treatment systems, Water science and technology: a journal of the International Association on Water Pollution Research 43 (2001) 91. [11] H. Gong, Z. Jin, X. Wang, K. Wang, Membrane fouling controlled by coagulation/adsorption during direct sewage membrane filtration (DSMF) for organic matter concentration, Journal of Environmental Sciences 32 (2015) 1-7. [12] Z. Jin, H. Gong, K. Wang, Application of hybrid coagulation microfiltration with air backflushing to direct sewage concentration for organic matter recovery, Journal of Hazardous Materials 283 (2015) 824-831. [13] H. Gong, X. Wang, M. Zheng, Z. Jin, K. Wang, Direct sewage filtration for concentration of organic matters by dynamic membrane, Water Science & Technology 70 (2014) 1434-1440. [14] Z. Jin, H. Gong, H. Temmink, H. Nie, J. Wu, J. Zuo, K. Wang, Efficient sewage pre-concentration with combined coagulation microfiltration for organic matter recovery, Chemical Engineering Journal 292 (2016) 130–138. [15] I. Akanyeti, H. Temmink, M. Remy, A. Zwijnenburg, Feasibility of bioflocculation in a high-loaded membrane bioreactor for improved energy recovery from sewage, Water Science & Technology 61 (2010). [16] G. Mezohegyi, M.R. Bilad, I.F.J. Vankelecom, Direct sewage up-concentration by submerged aerated and vibrated membranes, Bioresource Technology 118 (2012) 1-7. [17] S. Leakovic, I. Mijatovic, S. Cerjan-Stefanovic, E. Hodzic, Nitrogen removal from fertilizer wastewater by ion exchange, Water research 34 (2000) 185-190. [18] H. Zhang, A. Li, W. Zhang, C. Shuang, Combination of Na-modified zeolite and anion exchange resin for advanced treatment of a high ammonia-nitrogen content municipal effluent, Journal of Colloid and Interface 19
Science 468 (2016) 128-135. [19] A. Malovanyy, H. Sakalova, Y. Yatchyshyn, E. Plaza, M. Malovanyy, Concentration of ammonium from municipal wastewater using ion exchange process, Desalination 329 (2013) 93-102. [20] C. NEPA, Water and Wastewater Monitoring Methods, 4th ed., Chinese Environmental Science Publishing House, Beijing, China, 2002. [21] A. Ravazzini, A. Van Nieuwenhuijzen, J. Van Der Graaf, Direct ultrafiltration of municipal wastewater: comparison between filtration of raw sewage and primary clarifier effluent, Desalination 178 (2005) 51-62. [22] Q. Zhao, H. Zhong, K. Wang, L. Wei, J. Liu, Y. Liu, Removal and transformation of organic matters in domestic wastewater during lab-scale chemically enhanced primary treatment and a trickling filter treatment, Journal of Environmental Sciences 25 (2013) 59-68. [23] G. Mouri, S. Takizawa, K. Fukushi, T. Oki, Estimation of the effects of chemically-enhanced treatment of urban sewage system based on life-cycle management, Sustainable Cities and Society 9 (2013) 23-31. [24] S. Luostarinen, W. Sanders, K. Kujawa-Roeleveld, G. Zeeman, Effect of temperature on anaerobic treatment of black water in UASB-septic tank systems, Bioresource Technology 98 (2007) 980-986. [25] M.K. Sharma, A.A. Kazmi, Anaerobic onsite treatment of black water using filter-based packaged system as an alternative of conventional septic tank, Ecological Engineering 75 (2015) 457-461. [26] G. Lettinga, J.B. Van Lier, J.C.L. Van Buuren, G. Zeeman, Sustainable development in pollution control and the role of anaerobic treatment, Water Science and Technology 44 (2001) 181-188. [27] L. Faust, H. Temmink, A. Zwijnenburg, A. Kemperman, H. Rijnaarts, High loaded MBRs for organic matter recovery from sewage: effect of solids retention time on bioflocculation and on the role of extracellular polymers, Water research 56 (2014) 258-266. [28] L. Faust, H. Temmink, A. Zwijnenburg, A. Kemperman, H. Rijnaarts, Effect of dissolved oxygen concentration on the bioflocculation process in high loaded MBRs, Water research 66 (2014) 199-207. [29] Z. Cai, M.M. Benjamin, NOM fractionation and fouling of low-pressure membranes in microgranular adsorptive filtration, Environmental science & technology 45 (2011) 8935-8940. [30] Z. Cai, C. Wee, M.M. Benjamin, Fouling mechanisms in low-pressure membrane filtration in the presence of an adsorbent cake layer, Journal of Membrane Science 433 (2013) 32-38. [31] S.K. Lateef, B.Z. Soh, K. Kimura, Direct membrane filtration of municipal wastewater with chemically enhanced backwash for recovery of organic matter, Bioresource Technology 150 (2013) 149-155. [32] D. Alchin, Ion exchange resins, Chemical Processes in New Zealand, New Zealand Institute of Chemistry Education, pp. XIII-D-1–XIII-D-71998. [33] H. Huang, D. Xiao, R. Pang, C. Han, D. Li, Simultaneous removal of nutrients from simulated swine wastewater by adsorption of modified zeolite combined with struvite crystallization, Chemical Engineering Journal 256 (2014) 431–438. [34] A. Farkas, M. Rozić, Z. Barbarić-Mikocević, Ammonium exchange in leakage waters of waste dumps using natural zeolite from the Krapina region, Croatia, Journal of Hazardous Materials 117 (2005) 25-33. [35] S. Seokjun, J. Hongrae, L. Jaekwang, K. Ghayoung, P. Daewook, H. Nojima, L. Jaeyoung, M. Seunghyeon, Investigation on removal of hardness ions by capacitive deionization (CDI) for water softening applications, Water research 44 (2010) 2267-2275. [36] Z. Chen, H. Zhang, C. Wu, Y. Wang, W. Li, A study of electrosorption selectivity of anions by activated carbon electrodes in capacitive deionization, Desalination 369 (2015) 46-50. [37] I. Angelidaki, W. Sanders, Assessment of the anaerobic biodegradability of macropollutants, Re/Views in Environmental Science & Bio/Technology 3 (2004) 117-129. [38] J.B. Van Lier, High-rate anaerobic wastewater treatment: diversifying from end-of-the-pipe treatment to 20
resource-oriented conversion techniques, Water Science and Technology 57 (2008) 1137-1148.
21
22
Highlights: 1. Concept for simultaneous recovery of energy and nitrogen from sewage was proposed. 2. Membrane-based preconcentration(MPC) was combined with ion exchange(IE) process. 3. COD was recovered by MPC to achieve energy recovery. 4. Presentation of hard ions (Ca2+ & Mg2+) decreased NH4-N IE recovery efficiency.
23