- Email: [email protected]
Preparation, characterization and application of low-cost pyrophyllite-alumina composite ceramic membranes for treating low-strength domestic wastewater
Author’s Accepted Manuscript Preparation, Characterization and Application of Low-Cost Pyrophyllite-Alumina Composite Ceramic Membranes for Treating Low-Strength Domestic Wastewater Yeongmi Jeong, Sanghyup Lee, Seungkwan Hong, Chanhyuk Park www.elsevier.com/locate/memsci
PII: DOI: Reference:
S0376-7388(17)30576-8 http://dx.doi.org/10.1016/j.memsci.2017.04.068 MEMSCI15232
To appear in: Journal of Membrane Science Received date: 28 February 2017 Revised date: 3 April 2017 Accepted date: 30 April 2017 Cite this article as: Yeongmi Jeong, Sanghyup Lee, Seungkwan Hong and Chanhyuk Park, Preparation, Characterization and Application of Low-Cost Pyrophyllite-Alumina Composite Ceramic Membranes for Treating LowStrength Domestic Wastewater, Journal of Membrane Science, http://dx.doi.org/10.1016/j.memsci.2017.04.068 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 galley proof before it is published in its final citable 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.
Preparation, Characterization and Application of Low-Cost Pyrophyllite-Alumina Composite Ceramic Membranes for Treating Low-Strength Domestic Wastewater
Yeongmi Jeonga, Sanghyup Leea,b, Seungkwan Hongc, Chanhyuk Parka,*
a
Center for Water Resource Cycle Research, Korea Institute of Science and Technology, Seoul 02792, South Korea
b
Green School, Korea University, Seoul 02841, South Korea
c
School of Civil, Environmental and Architectural Engineering, Korea University, Seoul 02841, South Korea
Submitted to Journal of Membrane Science
*
Corresponding author
Tel. 82-2-958-6798 E-mail: [email protected] (C. P.)
1
Abstract: To advance cost-effective strategies for developing flat-sheet ceramic microfiltration membranes, the feasibility of using waste mineral-based materials as a ceramic membrane in engineered membrane bioreactor systems (MBRs) for treating low-strength domestic wastewater was evaluated. The ceramic membrane support layers that had effectively been fabricated with pyrophyllite and alumina were sintered at 1,350 °C, and subsequently coated with alumina powder suspension to achieve a narrow pore size distribution. Membrane surface properties were characterized to understand the membrane fouling phenomenon in ceramic MBRs (CMBRs). Consistently high organic removal efficiency could be achieved under all investigated hydraulic retention times (HRTs) in the CMBRs. The physico-chemical sludge properties were measured to evaluate design parameters affecting membrane fouling for the CMBR. We further evaluated the performance of pyrophyllite-alumina composite ceramic membranes in a pilot-scale CMBR plant, with 1 m3/d capacity that was constructed at an actual water resource recovery facility (WRRF), to improve nitrogen removal and produce high quality effluent through a combined process of modified Ludzack-Etiinger (MLE) and CMBR (MLE/CMBR). The MLE/CMBR was operated successfully up to 20 LMH, with the final effluent chemical oxygen demand (COD) being maintained below 16.3 mg/L.
Keywords: Ceramic membrane; Ceramic membrane bioreactor; Pyrophyllite; Pyrophyllitealumina composite membrane
2
1.
Introduction The membrane properties used for MBRs were based on polymeric membranes, due to
the relatively low cost of materials and numerous operational experiences. The demand of the MBRs for the mainstream treatment of domestic wastewater has continued to increase in recent years. Conventional MBR technology has weakness of nutrient removal due to the intensive aeration [1]. Recent studies have been directed towards improved nitrogen removal along with organic compound oxidation by using modified MBRs combined with anaerobic/oxic (AOMBR) [2, 3], anaerobic/anoxic/oxic (A2O-MBR) [4], modified anoxic/oxic (A/O-MBR) [5, 6], MUCT-MBR [7-9], and MLE-MBR systems [10, 11]. Among these advanced MBR configurations, membrane fouling control remains the biggest problem, as it incurs maintenance costs [12-14]. Most of the research in this area has been focused on alleviating membrane fouling by controlling operating conditions, such as membrane relaxation, backwashing, electrochemical oxidation, vibration and the addition of fluidized media or bio-carriers [15-20]. A great deal of effort has been made to understand the physicochemical interactions between various foulants and membrane surfaces [18, 21]. Increased membrane hydrophilicity caused by surface modification, has been reported to help reduce membrane fouling by limiting hydrophobic interactions between foulants and the membrane surface [22-24]. Although inspired surface modification approaches to membrane fouling control, such as a polymer grafting or a polydopamine coating induced by self-polymerization of dopamine, have been proposed [25-27], advancements in membrane production and module design continue to have limitations for widespread practical use. Another emerging application involving antifouling membranes is the use of alternative membrane materials, such as ceramic membranes, as a replacement for polymeric membranes.
3
Interest in ceramic membranes has been increasing in recent years as they offer lower fouling propensity, greater structural integrity and strong chemical resistivity [18, 21, 28, 29]. Ceramic membranes can be subjected to more extreme backwashing, and aggressive chemical cleaning and regeneration methods during application. However, most commercialized ceramic membranes use expensive high-purity ceramic materials, such as alumina, titania and silicon carbide [30, 31]. Therefore, the potential application of mineral-based ceramic membranes have attracted more attention due to the abundance of natural minerals in them and the lower sintering costs associated with them [30, 32]. Pyrophyllite is a natural clay material that is abundant in South Korea. It could be an option for low-cost ceramic membranes. The chemical composition of pyrophyllite materials is aluminum silicate hydroxide (Al2Si4O10(OH)2), which basically consists of one mole of aluminum (III) oxide (Al2O3) and four moles of silicon dioxide (SiO2). Although the flexural strength of these natural mineral-based support layers may not be sufficient, it could be enhanced through the formation of a composite support layer by adding a small amount of alumina, diatomite or kaolin. Currently, a significant amount of waste pyrophyllite, which contains less than 20 % aluminum oxide, has been accumulating in the local mines in South Korea. A practical use of these waste pyrophyllite materials would be necessary to create high-value-added industries, such as membrane-based wastewater treatment industry. This study initially developed flat-sheet ceramic microfiltration membranes by utilizing the waste pyrophyllite materials as a means of obtaining low-cost ceramic membranes. To the best of our knowledge, this work is the first report suggesting that using pyrophyllite-alumina composite ceramic membranes for treating domestic wastewater is a feasible option to reduce membrane-related costs, as well as for the development of polymeric membranes. The goals of
4
this study were to determine whether the novel ceramic membrane is capable of being applied to MBR systems for solid-liquid separation at various HRTs. The effect of various HRTs on the physico-chemical properties of sludge that would be relevant to membrane fouling behavior, such as extracellular polymeric substance (EPS) and surface charge of microbial flocs, were then examined to determine reasonable optimal operating conditions for the CMBR. In addition, the membrane flux and trans-membrane pressure (TMP) of a pilot-scale MLE/CMBR process were investigated to assess the feasibility of the developed pyrophyllite-alumina composite ceramic membranes for applying the MBR systems with improved filtration performance.
2.
Materials and Methods
2.1.
Pyrophyllite-based ceramic membrane preparation Pyrophyllite (Hankook Mineral Powder Co., Ltd., South Korea) and alumina (AM-21,
Sumitomo Chemical Co., Ltd., Japan) powders were used for the preparation of ceramic membrane support layers. The average particle sizes of these powders were 4.80 and 6.95 μm, as measured by a particle size analyzer (LSTM 13 320 MW, Beckman Coulter, Indianapolis, IN). The mixed slurry for the preparation of the ceramic membrane support layers was composed of 62.5 wt% of pyrophyllite powder, 15.6 wt% of alumina powder and 6.3 wt% of methyl cellulose (LOTTE Fine Chemical Co., Ltd., South Korea) as the binder, with 15.6 wt% of distilled water as the solvent. After stirring continuously for 0.5 hours at room temperature, the mixed slurry was extruded using a conventional screw extruder (Y-50E, Ishikawa Toki Iron Works, Co., Ltd., Japan). The dimensions of the support layers were, a width of 7.5 cm, a length of 21.0 cm and a height of 0.4 cm, with 15 inner holes (3.5 mm wide and 2.0 mm high). Standard membrane specimens were dried for 24 h at 80 °C and subsequently sintered at 1,350 °C for 2 h.
5
To apply the coating layers on the pyrophyllite-alumina composite support layers, a coating solution was prepared with 8 wt% of alumina powder (AES-11, Sumitomo Chemical Co., Ltd., Japan), 33.0 wt% of isopropyl alcohol (I0359, Samchun Pure Chemical Co., Ltd., South Korea), 2 wt% of binder (AP-2, Yuken Industry Co., Ltd., Japan) and 57 wt% of distilled water. The pyrophyllite-alumina composite support layers were dip-coated at a descending speed of 1.1 cm/s, maintained for 1 min, before being retracted at a withdrawal speed of 0.5 cm/s. To remove any excess water and alcohol residues, the alumina-coated membrane specimens were first dried in an oven at 80 °C for 12 h, and subsequently heated at 1,350 °C for 2 h.
2.2.
Membrane characterization The pore size distribution and mean pore size of the pyrophyllite-alumina composite
ceramic membranes before and after alumina coating were analyzed using a mercury porosimetry analysis technique (AutoPore IV 9510 Series, Micromeritics Instrument Corp., Norcross, GA) which is based on the intrusion of mercury into a porous structure under stringently controlled pressures. The elemental composition of the pyrophyllite-alumina composite ceramic membranes before and after the alumina coating was measured by X-ray photoelectron spectroscopy (XPS, PHI 5000 VersaProbe, ULVAC PHI, Japan). The surface of the sample was irradiated with a beam of monochromatic Al Kα source operating at 1.486 keV and 24.5 W beam power. The base pressure of the sample analysis chamber was 2.0 × 10-7 Pa. The spectra were collected in hybrid mode using electrostatic and magnetic lenses from a nominal spot size of 100 μm × 100 μm. For the alumina-coated ceramic membrane samples, surface morphology was investigated by scanning electron microscopy (SEM, Hitachi S-5000H, Japan). Before imaging,
6
the membrane surfaces were sputter-coated with a chromium layer. An acceleration voltage of 20.0 kV was applied to capture an image of all the membrane samples. The zeta potential of the prepared ceramic membranes was determined by electrophoretic mobility measurements of finely ground membrane powders in liquid suspension using an optical technique (Zetasizer Nano ZS, Malvern Instruments Ltd., United Kingdom). Membrane suspensions in the background electrolyte, with 50 mg/L of membrane powder, were prepared and injected into a folded capillary cell (DTS1070, Malvern Instruments Ltd., United Kingdom). The pH of the suspensions was adjusted to appropriate values in the range of 3 to 10. The flexural strengths of the prepared ceramic membranes were measured by a three-point bending strength method using a universal testing machine (H10K-S, Tinius Olsen, PA), for which, 8 × 5 × 80 mm specimens were machined and tested on an outer span width of 70 mm.
2.3.
Membrane bioreactors and experimental setup The CMBR used in this study was made of acrylic, with a working volume of 2.4 L,
being operated to evaluate the filtration and treatment performance of the developed ceramic membranes (Fig. 1). For the CMBR, two flat sheet pyrophyllite-alumina composite ceramic membrane modules, with a nominal pore size of 0.15 μm, were mounted on the reactor, with each module located above a diffuser with an overall effective aeration rate of 2 L of airflow per minute. The effective area of the membrane surface was 3.15 × 10-2 m2 for each module. The pyrophyllite-alumina composite ceramic membrane was operated in the inside-out mode during the filtration experiments, without backwashing, physical and/or chemical cleanings. The reactors were equipped with level sensors, pH and ORP probes, while the flow rate was kept at the same specific gas flow rate of 2 m/h. The concentration of oxygen in the CMBR was always
7
monitored by a DO probe with a bench-top multi meter (HQ440d, Hach Company, Loveland, CO) above 2 mg/L, indicating that the air applied for cake scouring is also sufficient for supplying oxygen for aerobic microorganisms.
FIGURE 1
The CMBR was seeded with the sludge obtained from the activated sludge bioreactor of a full-scale domestic wastewater treatment plant in South Korea, respectively. The initial mixed liquor suspended solids (MLSS) concentrations in the CMBR were adjusted to 6,000 mg/L. The CMBR was fed with identical synthetic wastewater containing glucose (294.6 mg/L for CMBR as COD) and the required nutrient sources for microbial growth. The membrane fluxes were kept constant under each operating condition by adopting a suction cycle of 4-min on and 1-min off, to reduce the attachment of foulants to the membrane surface. TMP was recorded by a digital pressure gauge (KELLER PR-21Y, Switzerland), connected between the membranes and the peristaltic permeate pumps (Masterflex L/S, Cole-Parmer, IL). The TMPs were used to assess the membrane fouling rates. The solids retention time (SRT) was kept constant at 45 days in the aerobic CMBR.
2.4.
Pilot-scale MLE/CMBR MLE is the most commonly used biological nutrient removal (BNR) process in MBR,
which is primarily targeting nitrogen removal. In our pilot-scale study, the 1 m3/d capacity MLE/CMBR process coupled with pyrophyllite-alumina composite ceramic membranes was operated at a target SRT of 15 d and a HRT of 4 hr. The actual influent wastewater has low
8
strength organic contents with a total COD concentration of 197.0 ± 62.9 mg/L, a soluble COD of 102.2 ± 52.2 mg/L, a total nitrogen (T-N) of 23.0 ± 5.0 mg/L, and a total phosphorus (T-P) of 7.3 ± 2.6 mg/L. The mixed liquor is recycled from membrane tank to anoxic tank. DO concentration in membrane tank was maintained between 2-4 mg/L because the excess DO transferred from membrane tank to anoxic tank can interrupt de-nitrification process. The mixed liquor was recycled at 2-3 times of influent flowrate (Q) for the purpose of preventing excess MLSS accumulation in the membrane tank, which somewhat limits the process flexibility. MLE process is not primarily for phosphorus removal, however, some extra phosphorus removal can occur beyond the level achievable without anoxic tank depending on the ORP and SRT in anoxic tank. The membrane module in the aerobic tank was constructed with 22 elements (1.28 × 10-1 m2 for each element) of the pyrophyllite-alumina composite ceramic membrane, which have a 2.8 m2 of effective membrane surface area. Ceramic membrane filtration started at a constant flux (15.0 L m-2 h-1 (LMH)) over 30 d of operation. After 30 days, the membrane flux was increased to a constant operating flux of 20.0 LMH by changing the permeate suction pump speed. During this whole experimental period, a significant increase in TMP was not observed without physical and chemical cleaning.
2.5.
Analytical methods Samples were regularly collected in the sequence of influent, permeate and mixed liquor,
from the CMBR for analyses. The suspended biomass concentration was determined by measuring the MLSS and volatile suspended solids (MLVSS) in accordance with the Standard Methods 2540 D/E [33]. Chemical analyses for COD, T-N and T-P were carried out using HACH DR/3900 spectrophotometer, following the testing procedure for each. The total organic
9
carbon (TOC) was analyzed using a TOC analyzer (TOC-L CPH, Shimadzu Corp., Japan) to quantify the concentrations of organics in the influent wastewater, mixed liquor supernatant and membrane permeate. Cations were measured using one ion chromatography (ICS-1000, Dionex Corp., Sunnyvale, CA), while anions were measured using another (883 Basic IC plus, Metrohm., Switzerland). The concentration of EPS was measured as carbohydrate and protein, using a cation exchange resin (CER, Dowex® Marathon® C, Na+ form, Sigma-Aldrich, Bellefonte, PA) extraction method [34]. The exchange resin (70 g of CER/ g VSS) was added to 50 mL of mixed liquor sample and mixed at 600 rpm for 2 h at 4 °C. The mixture (50 mL) was centrifuged for 15 min at 12,000 g to remove suspended solids. The carbohydrate content was measured using the Anthrone method with glucose as the standard reference [14, 35, 36]. The Lowry method was used to determine the protein, using bovine serum albumin (BSA) as the standard [37]. The absorbance of the prepared supernatant solutions was subsequently measured using a spectrophotometer. The surface charge of sludge flocs was determined using the colloidal titration technique [38]. Polybrene (0.002 N) and the potassium salt of polyvinyl sulfate (PVSK, 0.001 N) were used as the cationic and anionic standards, respectively. 2 mL of sludge sample (2,000 mg VSS/L) was mixed with an excess amount of polybrene standard solution. The PVSK standard solution was subsequently used to titrate against the excess amount of polybrene, using toluidine blue as an indicator. An equal volume of polybrene in deionized water was used as the blank. The surface charge of the sludge was calculated using Eq. (1). Charge (meq./L) =
(A - C) ´ B ´100 V ,
where, A = volume of PVSK added to the sludge sample, mL B = normality of PVSK, eq./L 10
(1)
C = volume of PVSK added to the blank, mL V = volume of sludge sample used, mL
3.
Results and Discussion
3.1.
Physico-chemical properties of the pyrophyllite-alumina composite ceramic membrane
3.1.1. Morphology SEM images depicting surface morphologies of the pyrophyllite-alumina composite ceramic membrane are shown in Fig. 2(a) and 2(b). Membrane surface morphologies before and after the alumina coating were markedly different (Fig. 2(a)). The ceramic membrane represented an asymmetric structure, with thin and dense layers, originating from the alumina coating, which play an important role in determining the average pore size of ceramic membranes and the filtration characteristics. The support layers of typical ceramic membranes, with thick and porous layers, provide most of the mechanical integrity and size-exclusion separation characteristics of particles in aqueous solutions. The SEM images of pyrophyllitealumina composite ceramic membranes after the coating showed that alumina powders were randomly distributed on their support layers, without defects, cracks and delamination between the coating and support layers (Fig. 2(b)).
FIGURE 2(a), (b)
3.1.2. Pore structure and distribution The nominal pore size and distribution of the prepared ceramic membranes changed after the application of alumina coating on the pyrophyllite-alumina composite support layers (Fig. 3),
11
indicating that the coated layers were properly bound to the support layers, without significant aggregation and pore blocking. The elimination efficiency of alumina coating layers on the support layers was evaluated by subjecting the ceramic membranes to bath sonication. If the alumina constituents remained after 10 min of bath sonication, it implies a robust binding on the porous support layers (Fig. S1). Although the nominal pore size of the pyrophyllite-alumina composite support layers was 0.95 ± 0.04 μm, the surfaces formed a very small pore size (0.15 μm) on the membrane coated with alumina particles. The prepared coated layer presented a narrow pore size distribution, which is suitable for use as a separation membrane for wastewater treatment. The pore size determination was affected by the sintering temperatures on porous ceramic membranes.
FIGURE 3
3.1.3. Mechanical strength To explain the importance of mechanical stability for practical MBR operations better, we have presented the bending strength of the developed ceramic membrane. Specifically, the bending strength of the pyrophyllite-alumina composite ceramic membrane, sintered at 1,350 °C, was approximately 28.5 ± 1.8 MPa (Table. S1), which represents a reliable value, sufficiently robust for the assembly of ceramic membranes into their module without fracture and severe micro-cracks [30]. Moreover, natural pyrophyllite materials retain their mechanical strength at relatively low sintering temperatures, thereby resulting in improved feasibility and costeffectiveness.
12
3.1.4. Zeta potential The surface charge of the pyrophyllite-alumina composite ceramic membranes increased as a function of pH, within the range of 3 to 10 (Fig. 4). The isoelectric point (IEP) of the ceramic membranes was expected to have a pH value lower than 3. This ceramic membrane consistently showed a higher negative charge at neutral pH than the results from previous studies, which showed the IEP of TiO2-based ceramic membranes in the pH range investigated [39]. This might be attributed to the different sintering and coating processes applied in order to produce ceramic membranes of different base-materials, which likely altered the IEP value and surface charge. More negative surface charge would reduce the adsorption and attachment of foulants with a negative charge in the wastewater sludge. Thus, with regard to domestic wastewater treatment, negatively charged ceramic membranes should be used in order to alleviate membrane fouling due to the strong electrostatic repulsion forces between the membrane surfaces and foulants.
FIGURE 4
3.1.5. XPS analysis Results from the XPS analysis for the ceramic membranes, before and after alumina coating, have been compared in Fig. 5. Carbon (C) and nitrogen (N) are base elements, while silica (Si) and oxygen (O) are two major elements present in the ceramic membrane before alumina coating, as the chemical composition of pyrophyllite minerals is based on silicon dioxide (SiO2) [40]. The ratios of elemental fractions of silica to oxygen remain almost identical (Si/O = 0.3) for the ceramic membrane, which decreased (Si/O = 0.04) after being coated with
13
alumina. A higher elemental fraction of aluminum (Al) could be expected for the coated membrane than the uncoated membrane because the alumina content of the coating solutions adds to its density on the membrane surfaces. This is evident from the Al/Si for uncoated (Al/Si = 0.57) and coated (Al/Si = 12.42) ceramic membranes. The results demonstrate that the coated layers were well deposited by the alumina particles. Al-coated surfaces produced from alumina powders display a narrower pore size as shown in Fig. 2.
FIGURE 5
3.2.
Laboratory- Scale CMBR
3.2.1. Filtration performance The average MLSS concentration for the CMBR was 5.7 g L-1, with average volatile fraction percentages of 71.4 %, under all conditions tested. These results indicate that the pyrophyllite-alumina composite ceramic membrane can retain the biomass present in the reactor to produce a high-quality effluent. Ceramic membrane filtration started at a low constant flux (3.0 L m-2 h-1 (LMH)) over 40 d of operation, except for the periods immediately after temporary foaming incidents. The start-up period for the CMBR was about 10 days for the stabilization of biomass concentration and permeate water quality. During the start-up period, temporary foaming was experienced in the CMBR, which was caused by aeration for membrane fouling control. However, it gradually disappeared within a week, as the foam surfaces became mobile [41, 42]. As severe membrane fouling can be expected for the CMBR, the flux applied was considerably lower than that in typical CMBR operation with polymeric membranes. During this experimental period, a significant increase in TMP was not observed without physical and
14
chemical cleaning. After 40 days, the ceramic membranes were taken out of the reactor and chemically cleaned by soaking in 0.1 % NaOCl for 24 hr. The membrane flux was then increased to a constant operating flux of 5.0 LMH by changing the permeate suction pump speed. Even though the average MLSS concentrations at steady state were observed to be similar at different HRTs, the TMP reached the required cleaning value of 30 kPa faster, as the HRT was reduced from 12 to 8 h. At an HRT of 12 h, membrane fouling occurred within approximately 40 d, while it required 25 d with an HRT of 8 h. This was due to the increase of organic loading rate (OLR) from 0.59 to 0.87 kg COD m-3 d-1, with the reduction in HRT. Although the membrane flux was not increased significantly in this study, the filtration performances of the developed ceramic membranes were sufficiently acceptable, as compared to the results obtained under similar operational conditions. To identify the membrane filtration performance, the average membrane permeability throughout the operation of CMBRs is obtained. It could be one of the indicators for comparison with the performances of other membranes reported in previous literature. The pure water permeability of the pyrophyllite-alumina composite ceramic membrane was measured as 280.9 LMH bar-1, whereas the average membrane permeability in the CMBR was recorded to 145.3 LMH bar-1. In this respect, the pyrophyllite-alumina composite ceramic membrane is a promising, competitive material for the treatment of domestic wastewater.
3.2.2. Reactor performance A consistently high COD and TOC removal efficiency of over 90 % could be achieved without fouling control (except for membrane relaxation) under all the investigated HRTs. The COD and TOC concentrations in the membrane permeate were maintained below 18.0 mg L-1
15
and 3.5 mg L-1, respectively. The corresponding average COD and TOC removals at different HRTs are presented in Fig. 6, which were calculated based on the influent wastewater organic (COD and TOC) concentration and membrane permeate. Increasing the OLR did affect the organic removal efficiencies of the CMBR. As seen in Fig. 7., at an OLR of 0.59 ± 0.04 kg COD m-3 d-1 (HRT = 12 h), the CMBR achieved the highest total organic removal efficiency, with a COD of 95.9 % and TOC of 97.3 %. When the OLR was increased to 0.88 ± 0.04 kg COD m-3 d1
(HRT = 8 h), the COD and TOC rejection rates were slightly reduced to 92.3 % and 90.4 %,
respectively. A decrease in HRT influenced the aerobic oxidation of influent wastewater organic compounds, even though the biomass concentration in the reactor was slightly higher at a shorter HRT. This finding indicates that the microorganisms required sufficient time to degrade the organic compounds in the influent. The effluent quality in this study suggests that the standards of U.S. EPA and the Ministry of Environment in Korea, with regard to secondary effluent, can be met by using pyrophyllite-alumina composite ceramic membrane treatment in CMBR. This suggests that a low-cost ceramic membrane could be a strong contender for a variety of MBR operations.
FIGURE 6
3.2.3. Physico-chemical sludge properties Numerous MBR studies have reported that EPS, which consists of proteins, polysaccharides, lipids and nucleic acids, is one of the main causes of fouling. The amount of EPS in the CMBR increased, when the HRT was decreased (Fig. 7). The protein and carbohydrate concentrations increased in the CMBR with shorter HRT, which required in a high
16
fouling potential due to the positive correlation between EPS and MLSS concentration [43]. A change in hydraulic conditions, such as a permeate flux, had an impact on the EPSp and EPSc concentrations in the CMBRs. However, membrane fouling was not significantly affected. The results were consistent with several researches in the past, which demonstrate that MLSS concentration in the range of 5 to 8 g/L, which is the general practical value in the MBR, does not have a significant effect on membrane fouling [44, 45].
FIGURE 7
A decrease in HRT could significantly reduce the negative charge on the surface of sludge flocs in CMBR (Fig. S2). Surface charge is related to the ionizable groups present on sludge surfaces that induce an increase in the polar interactions of EPS with water molecules. Sludge surfaces with a shorter HRT of 8 h (-1.1 ± 0.5 meq./g VSS) were negatively charged and less hydrophobic than those with a longer HRT of 12 h (-2.5 ± 0.3 meq./g VSS). This was attributed to the ratio of proteins to carbohydrates increasing slightly when the HRT decreased (Fig. 7) because the amino groups in proteins carry positive charges, with the ability to neutralize some of the negative charge of the sludge flocs. The importance of the ratio of proteins to carbohydrates in determining the surface charge of aerobic sludge flocs was reported [38]. It was also evidenced that the hydrophobic fraction of EPS was made up only of proteins, instead of carbohydrates [46].
3.3.
Pilot-scale MLE-CMBR plant
3.3.1. Reactor performance
17
The MLE/CMBR coupled with the developed low-cost ceramic membranes, was configured to improve the nitrogen removal of actual domestic wastewater. Table 1 contains a summary of the overall operating conditions during the last 50 d of operation with an HRT of 4 hr. The influent and effluent COD concentrations in the MLE/CMBR during the study period are presented (Fig. 8). With an average influent COD concentration of 197.0 ± 62.9 mg/L, the effluent COD concentration was 16.3 ± 5.5 mg/L for the MLE/CMBR systems, with an average COD removal rate of 91.6 ± 3.5% and an excellent NH4+-N removal efficiency of 93.2 ± 9.3%, with an average effluent NH4+-N concentration of 2.1 mg/L. The average total N concentration in the MLE/CMBR effluent was 7.3 mg/L N, with a removal rate of 68.8 %, compared to the influent concentration (average value = 23.0 mg/L N).
TABLE 1 FIGURE 8
3.3.2. Membrane flux During short-term MLE/CMBR operations, the permeate flux was set to a constant value of 15 LMH, while the time required for the rapid increase in TMP of MLE/CMBR to cease was approximately 10 days without fouling control, except for membrane relaxation strategies (Fig. 9). The increase in TMP, corresponding to the membrane flux, is commonly attributed to the formation of a cake layer on the membrane surface, in addition to being correlated with biomass characteristics such as EPS, hydrophobicity, and surface charge [17, 47]. After the first 10 days of operation, maximum TMP (0.3 bar) was reached at an MLSS concentration of 6,500 mg/L, making it necessary to initiate physical cleaning strategies, such as back-flushing, in order to
18
keep the filtration process working at less than adequate TMP value. Therefore, the membrane was operated with 300 s cycles (270 s filtration and 30 s relaxation), with constant TMP of 0.28 bar (data not shown). It is a well-known fact that back-flushing affects the economic feasibility of the process [10]. However, higher frequency of back-flush did not improve membrane performance when the MLSS concentration was over 30,000 mg/L [48]. Most of the effort in this chemical-free cleaning strategy was desirable to reduce the time required to reach the maximum TMP in a full-scale system design [49]. The permeate membrane flux was maintained at 20 LMH for the last 20 days of operation, despite the short-term operation (Fig. 9).
4.
Conclusions In this work, a low-cost pyrophyllite-alumina composite ceramic flat-sheet microfiltration
membrane was fabricated and applied to the CMBR for treating the low-strength wastewater. Pyrophyllite, a bulk waste material with limited uses, was used as the main material for the support layers of the ceramic membrane, on which a more negatively-charged coated layer was deposited via dip-coating method, using sub-micron Al2O3 powder suspension. The prepared low-cost pyrophyllite-alumina composite ceramic membrane successfully removed organic matter (COD and TOC) in the CMBR from low-strength domestic wastewater. The physicochemical sludge properties, especially the ratio of proteins to carbohydrates, were affected when the HRT was decreased. The membrane performance was similar to the pyrophyllite-alumina composite ceramic membranes, with a larger effective surface area in the pilot MLE/CMBR study. To summarize, a cost-effective ceramic membrane could be a very competitive candidate for membrane-based treatment of low-strength domestic wastewater. This study is expected to suggest an effective way to recycle waste materials and produce a high-value ceramic membrane
19
for use as the MBR in wastewater treatment with a high effluent water quality. Results offer new options for the recycling of natural waste materials and opportunities for low-cost mineral-based ceramic membranes.
Acknowledgements This work was funded by the Ministry of Environment in Korea as part of ‘The Eco – Innovation Program’ (Grant No. 2014000150021) and in part by the Korea Institute of Science and Technology (KIST) Institutional Research Program (No. 2E27080). The authors are grateful to IB Materials, Co. Ltd., South Korea and Mr. Myoungjoong Kim, for his technical assistance on the ceramic membrane development.
References [1] K. Kimura, R. Nishisako, T. Miyoshi, R. Shimada, Y. Watanabe, Baffled membrane bioreactor (BMBR) for efficient nutrient removal from municipal wastewater, Water Res, 42 (2008) 625-632. [2] X.X. Yan, M.R. Bilad, R. Gerards, L. Vriens, A. Piasecka, I.F.J. Vankelecom, Comparison of MBR performance and membrane cleaning in a single-stage activated sludge system and a twostage anaerobic/aerobic (A/A) system for treating synthetic molasses wastewater, J Membrane Sci, 394 (2012) 49-56. [3] S.J. You, R.A. Damodar, S.C. Hou, Degradation of Reactive Black 5 dye using anaerobic/aerobic membrane bioreactor (MBR) and photochemical membrane reactor, J Hazard Mater, 177 (2010) 1112-1118. [4] J.R. Banu, D.K. Uan, I.T. Yeom, Nutrient removal in an A2O-MBR reactor with sludge reduction, Bioresource Technol, 100 (2009) 3820-3824. [5] J.Y. Shen, R. He, W.Q. Han, X.Y. Sun, J.S. Li, L.J. Wang, Biological denitrification of highnitrate wastewater in a modified anoxic/oxic-membrane bioreactor (A/O-MBR), J Hazard Mater, 172 (2009) 595-600. [6] Z.M. Fu, F.L. Yang, Y.Y. An, Y. Xue, Simultaneous nitrification and denitrification coupled with phosphorus removal in an modified anoxic/oxic-membrane bioreactor (A/O-MBR), Biochem Eng J, 43 (2009) 191-196.
20
[7] H.M. Zhang, X.L. Wang, J.N. Xiao, F.L. Yang, J. Zhang, Enhanced biological nutrient removal using MUCT-MBR system, Bioresource Technol, 100 (2009) 1048-1054. [8] H. Lee, J. Han, Z. Yun, Biological nitrogen and phosphorus removal in UCT-type MBR process, Water Sci Technol, 59 (2009) 2093-2099. [9] H. Monclus, J. Sipma, G. Ferrero, J. Comas, I. Rodriguez-Roda, Optimization of biological nutrient removal in a pilot plant UCT-MBR treating municipal wastewater during start-up, Desalination, 250 (2010) 592-597. [10] Z.H. Liang, A. Das, D. Beerman, Z.Q. Hu, Biomass characteristics of two types of submerged membrane bioreactors for nitrogen removal from wastewater, Water Res, 44 (2010) 3313-3320. [11] C. Abegglen, M. Ospelt, H. Siegrist, Biological nutrient removal in a small-scale MBR treating household wastewater, Water Res, 42 (2008) 338-346. [12] X. Ren, H.K. Shon, N. Jang, Y.G. Lee, M. Bae, J. Lee, K. Cho, I.S. Kim, Novel membrane bioreactor (MBR) coupled with a nonwoven fabric filter for household wastewater treatment, Water Res, 44 (2010) 751-760. [13] R.S. Trussell, R.P. Merlo, S.W. Hermanowicz, D. Jenkins, Influence of mixed liquor properties and aeration intensity on membrane fouling in a submerged membrane bioreactor at high mixed liquor suspended solids concentrations, Water Res, 41 (2007) 947-958. [14] H.Y. Ng, S.W. Hermanowicz, Membrane bioreactor operation at short solids retention times: performance and biomass characteristics, Water Res, 39 (2005) 981-992. [15] L. Jin, S.L. Ong, H.Y. Ng, Fouling control mechanism by suspended biofilm carriers addition in submerged ceramic membrane bioreactors, J Membrane Sci, 427 (2013) 250-258. [16] C.M. Chung, T. Tobino, K. Cho, K. Yamamoto, Alleviation of membrane fouling in a submerged membrane bioreactor with electrochemical oxidation mediated by in-situ free chlorine generation, Water Res, 96 (2016) 52-61. [17] P. Le-Clech, V. Chen, T.A.G. Fane, Fouling in membrane bioreactors used in wastewater treatment, J Membrane Sci, 284 (2006) 17-53. [18] L. Jin, S.L. Ong, H.Y. Ng, Comparison of fouling characteristics in different pore-sized submerged ceramic membrane bioreactors, Water Res, 44 (2010) 5907-5918. [19] S.R. Chae, Y.T. Ahn, S.T. Kang, H.S. Shin, Mitigated membrane fouling in a vertical submerged membrane bioreactor (VSMBR), J Membrane Sci, 280 (2006) 572-581. [20] H.C. Hong, H.J. Lin, R.W. Mei, X.L. Zhou, B.Q. Liao, L.H. Zhao, Membrane fouling in a membrane bioreactor: A novel method for membrane surface morphology construction and its application in interaction energy assessment, J Membrane Sci, 516 (2016) 135-143.
21
[21] S.J. Lee, M. Dilaver, P.K. Park, J.H. Kim, Comparative analysis of fouling characteristics of ceramic and polymeric microfiltration membranes using filtration models, J Membrane Sci, 432 (2013) 97-105. [22] C. Boo, J. Lee, M. Elimelech, Omniphobic Polyvinylidene Fluoride (PVDF) Membrane for Desalination of Shale Gas Produced Water by Membrane Distillation, Environ Sci Technol, 50 (2016) 12275-12282. [23] R.P. Merlo, R.S. Trussell, S.W. Hermanowicz, D. Jenkins, A comparison of the physical, chemical, and biological properties of sludges from a complete-mix activated sludge reactor and a submerged membrane bioreactor, Water Environ Res, 79 (2007) 320-328. [24] T. Nittami, H. Tokunaga, A. Satoh, M. Takeda, K. Matsumoto, Influence of surface hydrophilicity on polytetrafluoroethylene flat sheet membrane fouling in a submerged membrane bioreactor using two activated sludges with different characteristics, J Membrane Sci, 463 (2014) 183-189. [25] C.Q. Zhao, G.Q. Zhang, X.C. Xu, F.L. Yang, Y.S. Yang, Rapidly self-assembled polydopamine coating membranes with polyhexamethylene guanidine: Formation, characterization and antifouling evaluation, Colloid Surface A, 512 (2017) 41-50. [26] A. Asatekin, A. Menniti, S.T. Kang, M. Elimelech, E. Morgenroth, A.M. Mayes, Antifouling nanofiltration membranes for membrane bioreactors from self-assembling graft copolymers, J Membrane Sci, 285 (2006) 81-89. [27] S. Liang, Y. Kang, A. Tiraferri, E.P. Giannelis, X. Huang, M. Elimelech, Highly Hydrophilic Polyvinylidene Fluoride (PVDF) Ultrafiltration Membranes via Postfabrication Grafting of Surface-Tailored Silica Nanoparticles, Acs Appl Mater Inter, 5 (2013) 6694-6703. [28] I.S. Chang, K.H. Choo, C.H. Lee, U.H. Pek, U.C. Koh, S.W. Kim, J.H. Koh, Application of Ceramic Membrane as a Pretreatment in Anaerobic-Digestion of Alcohol-Distillery Wastes, J Membrane Sci, 90 (1994) 131-139. [29] M.P. Zacharof, R.W. Lovitt, The filtration characteristics of anaerobic digester effluents employing cross flow ceramic membrane microfiltration for nutrient recovery, Desalination, 341 (2014) 27-37. [30] J.H. Ha, S.Z.A. Bukhari, J. Lee, I.H. Song, C. Park, Preparation processes and characterizations of alumina-coated alumina support layers and alumina-coated natural materialbased support layers for microfiltration, Ceram Int, 42 (2016) 13796-13804. [31] L. Zhu, M.L. Chen, Y.C. Dong, C.Y.Y. Tang, A.S. Huang, L.L. Li, A low-cost mullitetitania composite ceramic hollow fiber microfiltration membrane for highly efficient separation of oil-in-water emulsion, Water Res, 90 (2016) 277-285. [32] J.H. Ha, Y.H. Park, I.H. Song, The preparation and pore characteristics of an alumina coating on a diatomite-kaolin composite support layer, J Ceram Soc Jpn, 122 (2014) 714-718.
22
[33] APHA, AWWA, WEF, Standard Methods for the Examination of Water and Wastewater, 18th Ed. ed., American Public Health Association, Washington, DC, 1992. [34] B. Frolund, R. Palmgren, K. Keiding, P.H. Nielsen, Extraction of extracellular polymers from activated sludge using a cation exchange resin, Water Res, 30 (1996) 1749-1758. [35] R.S. Trussell, R.P. Merlo, S.W. Hermanowicz, D. Jenkins, The effect of organic loading on process performance and membrane fouling in a submerged membrane bioreactor treating municipal wastewater, Water Res, 40 (2006) 2675-2683. [36] D. Jenkins, From Total Suspended Solids to Molecular Biology Tools-A Personal View of Biological Wastewater Treatment Process Population Dynamics, Water Environ Res, 80 (2008) 677-687. [37] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, Protein measurement with the Folin phenol reagent, J Biol Chem, 193 (1951) 265-275. [38] J.W. Morgan, C.F. Forster, L. Evison, A Comparative-Study of the Nature of Biopolymers Extracted from Anaerobic and Activated Sludges, Water Res, 24 (1990) 743-750. [39] R. Shang, A.R.D. Verliefde, J.Y. Hu, Z.Y. Zeng, J. Lu, A.J.B. Kemperman, H.P. Deng, K. Nijmeijer, S.G.J. Heijman, L.C. Rietveld, Tight ceramic UF membrane as RO pre-treatment: The role of electrostatic interactions on phosphate rejection, Water Res, 48 (2014) 498-507. [40] J.H. Ha, J. Lee, I.H. Song, S.H. Lee, The effects of diatomite addition on the pore characteristics of a pyrophyllite support layer, Ceram Int, 41 (2015) 9542-9548. [41] C. Park, S.W. Hermanowicz, D. Jolis, A novel technique for evaluating foam dynamics in anaerobic digesters, Water Sci Technol, 67 (2013) 2595-2601. [42] C. Park, S.W. Hermanowicz, A Multi-Point Electrical Resistance Measurement System for Characterization of Foam Drainage Regime and Stability, Aiche J, 60 (2014) 3143-3150. [43] F.G. Meng, H.M. Zhang, F.L. Yang, Y.S. Li, J.N. Xiao, X.W. Zhang, Effect of filamentous bacteria on membrane fouling in submerged membrane bioreactor, J Membrane Sci, 272 (2006) 161-168. [44] L. Duan, Y.H. Song, H.B. Yu, S.Q. Xia, S.W. Hermanowicz, The effect of solids retention times on the characterization of extracellular polymeric substances and soluble microbial products in a submerged membrane bioreactor, Bioresource Technol, 163 (2014) 395-398. [45] A. Cosenza, G. Di Bella, G. Mannina, M. Torregrossa, The role of EPS in fouling and foaming phenomena for a membrane bioreactor, Bioresource Technol, 147 (2013) 184-192. [46] F. Jorand, F. Boue-Bigne, J.C. Block, V. Urbain, Hydrophobic/hydrophilic properties of activated sludge exopolymeric substances, Water Sci Technol, 37 (1998) 307-315.
23
[47] H.J. Lin, K. Xie, B. Mahendran, D.M. Bagley, K.T. Leung, S.N. Liss, B.Q. Liao, Factors affecting sludge cake formation in a submerged anaerobic membrane bioreactor, J Membrane Sci, 361 (2010) 126-134. [48] A. Robles, M.V. Ruano, J. Ribes, J. Ferrer, Factors that affect the permeability of commercial hollow-fibre membranes in a submerged anaerobic MBR (HF-SAnMBR) system, Water Res, 47 (2013) 1277-1288. [49] R. Lee, P.L. McCarty, J. Bae, J. Kim, Anaerobic fluidized membrane bioreactor polishing of baffled reactor effluent during treatment of dilute wastewater, J Chem Technol Biot, 90 (2015) 391-397.
24
FIGURE CAPTIONS Figure 1 Figure 2
Figure 3 Figure 4 Figure 5
Figure 6 Figure 7 Figure 8 Figure 9
Schematic diagram of the membrane bioreactor submerged pyrophyllitealumina composite ceramic membranes. SEM images of (a) before the alumina coating and (b) after the alumina coating on the surfaces of the pyrophyllite-alumina composite ceramic membrane support layers. Pore size distributions for the support and coated layers of the pyrophyllitealumina composite ceramic membrane Zeta-potentials of the pyrophyllite-alumina composite ceramic membrane suspensions as a function of pH from 3 to 10 XPS analysis of the surface of the pyrophyllite-alumina composite ceramic membranes. Fractions of carbon, nitrogen, oxygen, silica, and aluminum relative to the sum of elements present at the surface of (a) uncoated ceramic membrane support layer and (b) alumina-coated ceramic membrane Removal efficiencies of COD and TOC in the CMBR at different HRTs Concentrations of protein, carbohydrate and total in EPS for the CMBR at different HRTs Influent and effluent COD concentrations in the MLE/CMBR systems during the period of pilot-scale study Variations in permeate flux for pilot-scale pyrophyllite-alumina composite ceramic membranes during the operation of MLE/CMBR.
25
FIGURE 1
26
FIGURE 2 (a)
(b)
27
FIGURE 3
Differential Intrusion (ml/g)
1.0 support layer coated layer
0.8
0.6
0.02
0.01
0.00
0.4
0.05
0.1
0.2
0.5
0.2
0.0 0.01
0.1
1
Pore Size ( m)
28
10
FIGURE 4
Zeta Potential (mV)
20
0
-20
-40
-60
-80 2
4
6
8
pH
29
10
FIGURE 5
60
(b)
Area Fraction (%)
(a) 40
20
0 C
N
O
Al
Si
C
Element
N
O
Element
30
Al
Si
FIGURE 6
Removal Efficiency (%)
100
COD
TOC
95
90
HRT = 12h
HRT = 8h
31
EPS Concentration (mg/g VSS)
FIGURE 7
40 HRT = 12h HRT = 8h
30
20
10
0 Proteins Carbohydrates Total EPS
32
FIGURE 8
COD Concentration (mg/L)
500 COD influent MLE/CMBR permeate 400
300
200
100
0 0
10
20
30
Days of Operation
33
40
50
FIGURE 9
Flux (Lm2hr-1)
20
15
10 0
10
20
30
Days of Operation
34
40
50
Graphical abstract
35
Table 1. Operating conditions of the pilot-scale MLE/CMBR system at actual domestic water resource recycle facility (WRRF) Factor
MLE/CMBR
Volume (L)
174.0
Flow rate (L/hr)
43.1±7.0 2
-1
Permeate Flux (Lm hr )
15.2±2.6
HRT (h)
4.3±1.5
SRT (d)
15.0
Aeration flow rate (L/min)
2.0 2
Effective membrane area (m )
2.9
pH
6.9±0.2
DO (mg/L)
2.6±0.4 27.8±1.6
Temp. (℃) 3
OLR (kg COD/m ·d)
0.04±0.01 Anoxic
MLSS (mg/L)
Aerobic
4,048±2,599
6,418±1,519
MLVSS (mg/L)
2,848±750
4,700±1,195
MLVSS/MLSS
0.70
0.73
36
Highlights
Waste pyrophyllite material was utilized in ceramic membrane developments
New ceramic membranes applied to laboratory-scale MBR for treating low-strength domestic wastewater
Scaled-up ceramic membranes were operated successfully up to 20 LMH in MLE/CMBR
37
PII: DOI: Reference:
S0376-7388(17)30576-8 http://dx.doi.org/10.1016/j.memsci.2017.04.068 MEMSCI15232
To appear in: Journal of Membrane Science Received date: 28 February 2017 Revised date: 3 April 2017 Accepted date: 30 April 2017 Cite this article as: Yeongmi Jeong, Sanghyup Lee, Seungkwan Hong and Chanhyuk Park, Preparation, Characterization and Application of Low-Cost Pyrophyllite-Alumina Composite Ceramic Membranes for Treating LowStrength Domestic Wastewater, Journal of Membrane Science, http://dx.doi.org/10.1016/j.memsci.2017.04.068 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 galley proof before it is published in its final citable 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.
Preparation, Characterization and Application of Low-Cost Pyrophyllite-Alumina Composite Ceramic Membranes for Treating Low-Strength Domestic Wastewater
Yeongmi Jeonga, Sanghyup Leea,b, Seungkwan Hongc, Chanhyuk Parka,*
a
Center for Water Resource Cycle Research, Korea Institute of Science and Technology, Seoul 02792, South Korea
b
Green School, Korea University, Seoul 02841, South Korea
c
School of Civil, Environmental and Architectural Engineering, Korea University, Seoul 02841, South Korea
Submitted to Journal of Membrane Science
*
Corresponding author
Tel. 82-2-958-6798 E-mail: [email protected] (C. P.)
1
Abstract: To advance cost-effective strategies for developing flat-sheet ceramic microfiltration membranes, the feasibility of using waste mineral-based materials as a ceramic membrane in engineered membrane bioreactor systems (MBRs) for treating low-strength domestic wastewater was evaluated. The ceramic membrane support layers that had effectively been fabricated with pyrophyllite and alumina were sintered at 1,350 °C, and subsequently coated with alumina powder suspension to achieve a narrow pore size distribution. Membrane surface properties were characterized to understand the membrane fouling phenomenon in ceramic MBRs (CMBRs). Consistently high organic removal efficiency could be achieved under all investigated hydraulic retention times (HRTs) in the CMBRs. The physico-chemical sludge properties were measured to evaluate design parameters affecting membrane fouling for the CMBR. We further evaluated the performance of pyrophyllite-alumina composite ceramic membranes in a pilot-scale CMBR plant, with 1 m3/d capacity that was constructed at an actual water resource recovery facility (WRRF), to improve nitrogen removal and produce high quality effluent through a combined process of modified Ludzack-Etiinger (MLE) and CMBR (MLE/CMBR). The MLE/CMBR was operated successfully up to 20 LMH, with the final effluent chemical oxygen demand (COD) being maintained below 16.3 mg/L.
Keywords: Ceramic membrane; Ceramic membrane bioreactor; Pyrophyllite; Pyrophyllitealumina composite membrane
2
1.
Introduction The membrane properties used for MBRs were based on polymeric membranes, due to
the relatively low cost of materials and numerous operational experiences. The demand of the MBRs for the mainstream treatment of domestic wastewater has continued to increase in recent years. Conventional MBR technology has weakness of nutrient removal due to the intensive aeration [1]. Recent studies have been directed towards improved nitrogen removal along with organic compound oxidation by using modified MBRs combined with anaerobic/oxic (AOMBR) [2, 3], anaerobic/anoxic/oxic (A2O-MBR) [4], modified anoxic/oxic (A/O-MBR) [5, 6], MUCT-MBR [7-9], and MLE-MBR systems [10, 11]. Among these advanced MBR configurations, membrane fouling control remains the biggest problem, as it incurs maintenance costs [12-14]. Most of the research in this area has been focused on alleviating membrane fouling by controlling operating conditions, such as membrane relaxation, backwashing, electrochemical oxidation, vibration and the addition of fluidized media or bio-carriers [15-20]. A great deal of effort has been made to understand the physicochemical interactions between various foulants and membrane surfaces [18, 21]. Increased membrane hydrophilicity caused by surface modification, has been reported to help reduce membrane fouling by limiting hydrophobic interactions between foulants and the membrane surface [22-24]. Although inspired surface modification approaches to membrane fouling control, such as a polymer grafting or a polydopamine coating induced by self-polymerization of dopamine, have been proposed [25-27], advancements in membrane production and module design continue to have limitations for widespread practical use. Another emerging application involving antifouling membranes is the use of alternative membrane materials, such as ceramic membranes, as a replacement for polymeric membranes.
3
Interest in ceramic membranes has been increasing in recent years as they offer lower fouling propensity, greater structural integrity and strong chemical resistivity [18, 21, 28, 29]. Ceramic membranes can be subjected to more extreme backwashing, and aggressive chemical cleaning and regeneration methods during application. However, most commercialized ceramic membranes use expensive high-purity ceramic materials, such as alumina, titania and silicon carbide [30, 31]. Therefore, the potential application of mineral-based ceramic membranes have attracted more attention due to the abundance of natural minerals in them and the lower sintering costs associated with them [30, 32]. Pyrophyllite is a natural clay material that is abundant in South Korea. It could be an option for low-cost ceramic membranes. The chemical composition of pyrophyllite materials is aluminum silicate hydroxide (Al2Si4O10(OH)2), which basically consists of one mole of aluminum (III) oxide (Al2O3) and four moles of silicon dioxide (SiO2). Although the flexural strength of these natural mineral-based support layers may not be sufficient, it could be enhanced through the formation of a composite support layer by adding a small amount of alumina, diatomite or kaolin. Currently, a significant amount of waste pyrophyllite, which contains less than 20 % aluminum oxide, has been accumulating in the local mines in South Korea. A practical use of these waste pyrophyllite materials would be necessary to create high-value-added industries, such as membrane-based wastewater treatment industry. This study initially developed flat-sheet ceramic microfiltration membranes by utilizing the waste pyrophyllite materials as a means of obtaining low-cost ceramic membranes. To the best of our knowledge, this work is the first report suggesting that using pyrophyllite-alumina composite ceramic membranes for treating domestic wastewater is a feasible option to reduce membrane-related costs, as well as for the development of polymeric membranes. The goals of
4
this study were to determine whether the novel ceramic membrane is capable of being applied to MBR systems for solid-liquid separation at various HRTs. The effect of various HRTs on the physico-chemical properties of sludge that would be relevant to membrane fouling behavior, such as extracellular polymeric substance (EPS) and surface charge of microbial flocs, were then examined to determine reasonable optimal operating conditions for the CMBR. In addition, the membrane flux and trans-membrane pressure (TMP) of a pilot-scale MLE/CMBR process were investigated to assess the feasibility of the developed pyrophyllite-alumina composite ceramic membranes for applying the MBR systems with improved filtration performance.
2.
Materials and Methods
2.1.
Pyrophyllite-based ceramic membrane preparation Pyrophyllite (Hankook Mineral Powder Co., Ltd., South Korea) and alumina (AM-21,
Sumitomo Chemical Co., Ltd., Japan) powders were used for the preparation of ceramic membrane support layers. The average particle sizes of these powders were 4.80 and 6.95 μm, as measured by a particle size analyzer (LSTM 13 320 MW, Beckman Coulter, Indianapolis, IN). The mixed slurry for the preparation of the ceramic membrane support layers was composed of 62.5 wt% of pyrophyllite powder, 15.6 wt% of alumina powder and 6.3 wt% of methyl cellulose (LOTTE Fine Chemical Co., Ltd., South Korea) as the binder, with 15.6 wt% of distilled water as the solvent. After stirring continuously for 0.5 hours at room temperature, the mixed slurry was extruded using a conventional screw extruder (Y-50E, Ishikawa Toki Iron Works, Co., Ltd., Japan). The dimensions of the support layers were, a width of 7.5 cm, a length of 21.0 cm and a height of 0.4 cm, with 15 inner holes (3.5 mm wide and 2.0 mm high). Standard membrane specimens were dried for 24 h at 80 °C and subsequently sintered at 1,350 °C for 2 h.
5
To apply the coating layers on the pyrophyllite-alumina composite support layers, a coating solution was prepared with 8 wt% of alumina powder (AES-11, Sumitomo Chemical Co., Ltd., Japan), 33.0 wt% of isopropyl alcohol (I0359, Samchun Pure Chemical Co., Ltd., South Korea), 2 wt% of binder (AP-2, Yuken Industry Co., Ltd., Japan) and 57 wt% of distilled water. The pyrophyllite-alumina composite support layers were dip-coated at a descending speed of 1.1 cm/s, maintained for 1 min, before being retracted at a withdrawal speed of 0.5 cm/s. To remove any excess water and alcohol residues, the alumina-coated membrane specimens were first dried in an oven at 80 °C for 12 h, and subsequently heated at 1,350 °C for 2 h.
2.2.
Membrane characterization The pore size distribution and mean pore size of the pyrophyllite-alumina composite
ceramic membranes before and after alumina coating were analyzed using a mercury porosimetry analysis technique (AutoPore IV 9510 Series, Micromeritics Instrument Corp., Norcross, GA) which is based on the intrusion of mercury into a porous structure under stringently controlled pressures. The elemental composition of the pyrophyllite-alumina composite ceramic membranes before and after the alumina coating was measured by X-ray photoelectron spectroscopy (XPS, PHI 5000 VersaProbe, ULVAC PHI, Japan). The surface of the sample was irradiated with a beam of monochromatic Al Kα source operating at 1.486 keV and 24.5 W beam power. The base pressure of the sample analysis chamber was 2.0 × 10-7 Pa. The spectra were collected in hybrid mode using electrostatic and magnetic lenses from a nominal spot size of 100 μm × 100 μm. For the alumina-coated ceramic membrane samples, surface morphology was investigated by scanning electron microscopy (SEM, Hitachi S-5000H, Japan). Before imaging,
6
the membrane surfaces were sputter-coated with a chromium layer. An acceleration voltage of 20.0 kV was applied to capture an image of all the membrane samples. The zeta potential of the prepared ceramic membranes was determined by electrophoretic mobility measurements of finely ground membrane powders in liquid suspension using an optical technique (Zetasizer Nano ZS, Malvern Instruments Ltd., United Kingdom). Membrane suspensions in the background electrolyte, with 50 mg/L of membrane powder, were prepared and injected into a folded capillary cell (DTS1070, Malvern Instruments Ltd., United Kingdom). The pH of the suspensions was adjusted to appropriate values in the range of 3 to 10. The flexural strengths of the prepared ceramic membranes were measured by a three-point bending strength method using a universal testing machine (H10K-S, Tinius Olsen, PA), for which, 8 × 5 × 80 mm specimens were machined and tested on an outer span width of 70 mm.
2.3.
Membrane bioreactors and experimental setup The CMBR used in this study was made of acrylic, with a working volume of 2.4 L,
being operated to evaluate the filtration and treatment performance of the developed ceramic membranes (Fig. 1). For the CMBR, two flat sheet pyrophyllite-alumina composite ceramic membrane modules, with a nominal pore size of 0.15 μm, were mounted on the reactor, with each module located above a diffuser with an overall effective aeration rate of 2 L of airflow per minute. The effective area of the membrane surface was 3.15 × 10-2 m2 for each module. The pyrophyllite-alumina composite ceramic membrane was operated in the inside-out mode during the filtration experiments, without backwashing, physical and/or chemical cleanings. The reactors were equipped with level sensors, pH and ORP probes, while the flow rate was kept at the same specific gas flow rate of 2 m/h. The concentration of oxygen in the CMBR was always
7
monitored by a DO probe with a bench-top multi meter (HQ440d, Hach Company, Loveland, CO) above 2 mg/L, indicating that the air applied for cake scouring is also sufficient for supplying oxygen for aerobic microorganisms.
FIGURE 1
The CMBR was seeded with the sludge obtained from the activated sludge bioreactor of a full-scale domestic wastewater treatment plant in South Korea, respectively. The initial mixed liquor suspended solids (MLSS) concentrations in the CMBR were adjusted to 6,000 mg/L. The CMBR was fed with identical synthetic wastewater containing glucose (294.6 mg/L for CMBR as COD) and the required nutrient sources for microbial growth. The membrane fluxes were kept constant under each operating condition by adopting a suction cycle of 4-min on and 1-min off, to reduce the attachment of foulants to the membrane surface. TMP was recorded by a digital pressure gauge (KELLER PR-21Y, Switzerland), connected between the membranes and the peristaltic permeate pumps (Masterflex L/S, Cole-Parmer, IL). The TMPs were used to assess the membrane fouling rates. The solids retention time (SRT) was kept constant at 45 days in the aerobic CMBR.
2.4.
Pilot-scale MLE/CMBR MLE is the most commonly used biological nutrient removal (BNR) process in MBR,
which is primarily targeting nitrogen removal. In our pilot-scale study, the 1 m3/d capacity MLE/CMBR process coupled with pyrophyllite-alumina composite ceramic membranes was operated at a target SRT of 15 d and a HRT of 4 hr. The actual influent wastewater has low
8
strength organic contents with a total COD concentration of 197.0 ± 62.9 mg/L, a soluble COD of 102.2 ± 52.2 mg/L, a total nitrogen (T-N) of 23.0 ± 5.0 mg/L, and a total phosphorus (T-P) of 7.3 ± 2.6 mg/L. The mixed liquor is recycled from membrane tank to anoxic tank. DO concentration in membrane tank was maintained between 2-4 mg/L because the excess DO transferred from membrane tank to anoxic tank can interrupt de-nitrification process. The mixed liquor was recycled at 2-3 times of influent flowrate (Q) for the purpose of preventing excess MLSS accumulation in the membrane tank, which somewhat limits the process flexibility. MLE process is not primarily for phosphorus removal, however, some extra phosphorus removal can occur beyond the level achievable without anoxic tank depending on the ORP and SRT in anoxic tank. The membrane module in the aerobic tank was constructed with 22 elements (1.28 × 10-1 m2 for each element) of the pyrophyllite-alumina composite ceramic membrane, which have a 2.8 m2 of effective membrane surface area. Ceramic membrane filtration started at a constant flux (15.0 L m-2 h-1 (LMH)) over 30 d of operation. After 30 days, the membrane flux was increased to a constant operating flux of 20.0 LMH by changing the permeate suction pump speed. During this whole experimental period, a significant increase in TMP was not observed without physical and chemical cleaning.
2.5.
Analytical methods Samples were regularly collected in the sequence of influent, permeate and mixed liquor,
from the CMBR for analyses. The suspended biomass concentration was determined by measuring the MLSS and volatile suspended solids (MLVSS) in accordance with the Standard Methods 2540 D/E [33]. Chemical analyses for COD, T-N and T-P were carried out using HACH DR/3900 spectrophotometer, following the testing procedure for each. The total organic
9
carbon (TOC) was analyzed using a TOC analyzer (TOC-L CPH, Shimadzu Corp., Japan) to quantify the concentrations of organics in the influent wastewater, mixed liquor supernatant and membrane permeate. Cations were measured using one ion chromatography (ICS-1000, Dionex Corp., Sunnyvale, CA), while anions were measured using another (883 Basic IC plus, Metrohm., Switzerland). The concentration of EPS was measured as carbohydrate and protein, using a cation exchange resin (CER, Dowex® Marathon® C, Na+ form, Sigma-Aldrich, Bellefonte, PA) extraction method [34]. The exchange resin (70 g of CER/ g VSS) was added to 50 mL of mixed liquor sample and mixed at 600 rpm for 2 h at 4 °C. The mixture (50 mL) was centrifuged for 15 min at 12,000 g to remove suspended solids. The carbohydrate content was measured using the Anthrone method with glucose as the standard reference [14, 35, 36]. The Lowry method was used to determine the protein, using bovine serum albumin (BSA) as the standard [37]. The absorbance of the prepared supernatant solutions was subsequently measured using a spectrophotometer. The surface charge of sludge flocs was determined using the colloidal titration technique [38]. Polybrene (0.002 N) and the potassium salt of polyvinyl sulfate (PVSK, 0.001 N) were used as the cationic and anionic standards, respectively. 2 mL of sludge sample (2,000 mg VSS/L) was mixed with an excess amount of polybrene standard solution. The PVSK standard solution was subsequently used to titrate against the excess amount of polybrene, using toluidine blue as an indicator. An equal volume of polybrene in deionized water was used as the blank. The surface charge of the sludge was calculated using Eq. (1). Charge (meq./L) =
(A - C) ´ B ´100 V ,
where, A = volume of PVSK added to the sludge sample, mL B = normality of PVSK, eq./L 10
(1)
C = volume of PVSK added to the blank, mL V = volume of sludge sample used, mL
3.
Results and Discussion
3.1.
Physico-chemical properties of the pyrophyllite-alumina composite ceramic membrane
3.1.1. Morphology SEM images depicting surface morphologies of the pyrophyllite-alumina composite ceramic membrane are shown in Fig. 2(a) and 2(b). Membrane surface morphologies before and after the alumina coating were markedly different (Fig. 2(a)). The ceramic membrane represented an asymmetric structure, with thin and dense layers, originating from the alumina coating, which play an important role in determining the average pore size of ceramic membranes and the filtration characteristics. The support layers of typical ceramic membranes, with thick and porous layers, provide most of the mechanical integrity and size-exclusion separation characteristics of particles in aqueous solutions. The SEM images of pyrophyllitealumina composite ceramic membranes after the coating showed that alumina powders were randomly distributed on their support layers, without defects, cracks and delamination between the coating and support layers (Fig. 2(b)).
FIGURE 2(a), (b)
3.1.2. Pore structure and distribution The nominal pore size and distribution of the prepared ceramic membranes changed after the application of alumina coating on the pyrophyllite-alumina composite support layers (Fig. 3),
11
indicating that the coated layers were properly bound to the support layers, without significant aggregation and pore blocking. The elimination efficiency of alumina coating layers on the support layers was evaluated by subjecting the ceramic membranes to bath sonication. If the alumina constituents remained after 10 min of bath sonication, it implies a robust binding on the porous support layers (Fig. S1). Although the nominal pore size of the pyrophyllite-alumina composite support layers was 0.95 ± 0.04 μm, the surfaces formed a very small pore size (0.15 μm) on the membrane coated with alumina particles. The prepared coated layer presented a narrow pore size distribution, which is suitable for use as a separation membrane for wastewater treatment. The pore size determination was affected by the sintering temperatures on porous ceramic membranes.
FIGURE 3
3.1.3. Mechanical strength To explain the importance of mechanical stability for practical MBR operations better, we have presented the bending strength of the developed ceramic membrane. Specifically, the bending strength of the pyrophyllite-alumina composite ceramic membrane, sintered at 1,350 °C, was approximately 28.5 ± 1.8 MPa (Table. S1), which represents a reliable value, sufficiently robust for the assembly of ceramic membranes into their module without fracture and severe micro-cracks [30]. Moreover, natural pyrophyllite materials retain their mechanical strength at relatively low sintering temperatures, thereby resulting in improved feasibility and costeffectiveness.
12
3.1.4. Zeta potential The surface charge of the pyrophyllite-alumina composite ceramic membranes increased as a function of pH, within the range of 3 to 10 (Fig. 4). The isoelectric point (IEP) of the ceramic membranes was expected to have a pH value lower than 3. This ceramic membrane consistently showed a higher negative charge at neutral pH than the results from previous studies, which showed the IEP of TiO2-based ceramic membranes in the pH range investigated [39]. This might be attributed to the different sintering and coating processes applied in order to produce ceramic membranes of different base-materials, which likely altered the IEP value and surface charge. More negative surface charge would reduce the adsorption and attachment of foulants with a negative charge in the wastewater sludge. Thus, with regard to domestic wastewater treatment, negatively charged ceramic membranes should be used in order to alleviate membrane fouling due to the strong electrostatic repulsion forces between the membrane surfaces and foulants.
FIGURE 4
3.1.5. XPS analysis Results from the XPS analysis for the ceramic membranes, before and after alumina coating, have been compared in Fig. 5. Carbon (C) and nitrogen (N) are base elements, while silica (Si) and oxygen (O) are two major elements present in the ceramic membrane before alumina coating, as the chemical composition of pyrophyllite minerals is based on silicon dioxide (SiO2) [40]. The ratios of elemental fractions of silica to oxygen remain almost identical (Si/O = 0.3) for the ceramic membrane, which decreased (Si/O = 0.04) after being coated with
13
alumina. A higher elemental fraction of aluminum (Al) could be expected for the coated membrane than the uncoated membrane because the alumina content of the coating solutions adds to its density on the membrane surfaces. This is evident from the Al/Si for uncoated (Al/Si = 0.57) and coated (Al/Si = 12.42) ceramic membranes. The results demonstrate that the coated layers were well deposited by the alumina particles. Al-coated surfaces produced from alumina powders display a narrower pore size as shown in Fig. 2.
FIGURE 5
3.2.
Laboratory- Scale CMBR
3.2.1. Filtration performance The average MLSS concentration for the CMBR was 5.7 g L-1, with average volatile fraction percentages of 71.4 %, under all conditions tested. These results indicate that the pyrophyllite-alumina composite ceramic membrane can retain the biomass present in the reactor to produce a high-quality effluent. Ceramic membrane filtration started at a low constant flux (3.0 L m-2 h-1 (LMH)) over 40 d of operation, except for the periods immediately after temporary foaming incidents. The start-up period for the CMBR was about 10 days for the stabilization of biomass concentration and permeate water quality. During the start-up period, temporary foaming was experienced in the CMBR, which was caused by aeration for membrane fouling control. However, it gradually disappeared within a week, as the foam surfaces became mobile [41, 42]. As severe membrane fouling can be expected for the CMBR, the flux applied was considerably lower than that in typical CMBR operation with polymeric membranes. During this experimental period, a significant increase in TMP was not observed without physical and
14
chemical cleaning. After 40 days, the ceramic membranes were taken out of the reactor and chemically cleaned by soaking in 0.1 % NaOCl for 24 hr. The membrane flux was then increased to a constant operating flux of 5.0 LMH by changing the permeate suction pump speed. Even though the average MLSS concentrations at steady state were observed to be similar at different HRTs, the TMP reached the required cleaning value of 30 kPa faster, as the HRT was reduced from 12 to 8 h. At an HRT of 12 h, membrane fouling occurred within approximately 40 d, while it required 25 d with an HRT of 8 h. This was due to the increase of organic loading rate (OLR) from 0.59 to 0.87 kg COD m-3 d-1, with the reduction in HRT. Although the membrane flux was not increased significantly in this study, the filtration performances of the developed ceramic membranes were sufficiently acceptable, as compared to the results obtained under similar operational conditions. To identify the membrane filtration performance, the average membrane permeability throughout the operation of CMBRs is obtained. It could be one of the indicators for comparison with the performances of other membranes reported in previous literature. The pure water permeability of the pyrophyllite-alumina composite ceramic membrane was measured as 280.9 LMH bar-1, whereas the average membrane permeability in the CMBR was recorded to 145.3 LMH bar-1. In this respect, the pyrophyllite-alumina composite ceramic membrane is a promising, competitive material for the treatment of domestic wastewater.
3.2.2. Reactor performance A consistently high COD and TOC removal efficiency of over 90 % could be achieved without fouling control (except for membrane relaxation) under all the investigated HRTs. The COD and TOC concentrations in the membrane permeate were maintained below 18.0 mg L-1
15
and 3.5 mg L-1, respectively. The corresponding average COD and TOC removals at different HRTs are presented in Fig. 6, which were calculated based on the influent wastewater organic (COD and TOC) concentration and membrane permeate. Increasing the OLR did affect the organic removal efficiencies of the CMBR. As seen in Fig. 7., at an OLR of 0.59 ± 0.04 kg COD m-3 d-1 (HRT = 12 h), the CMBR achieved the highest total organic removal efficiency, with a COD of 95.9 % and TOC of 97.3 %. When the OLR was increased to 0.88 ± 0.04 kg COD m-3 d1
(HRT = 8 h), the COD and TOC rejection rates were slightly reduced to 92.3 % and 90.4 %,
respectively. A decrease in HRT influenced the aerobic oxidation of influent wastewater organic compounds, even though the biomass concentration in the reactor was slightly higher at a shorter HRT. This finding indicates that the microorganisms required sufficient time to degrade the organic compounds in the influent. The effluent quality in this study suggests that the standards of U.S. EPA and the Ministry of Environment in Korea, with regard to secondary effluent, can be met by using pyrophyllite-alumina composite ceramic membrane treatment in CMBR. This suggests that a low-cost ceramic membrane could be a strong contender for a variety of MBR operations.
FIGURE 6
3.2.3. Physico-chemical sludge properties Numerous MBR studies have reported that EPS, which consists of proteins, polysaccharides, lipids and nucleic acids, is one of the main causes of fouling. The amount of EPS in the CMBR increased, when the HRT was decreased (Fig. 7). The protein and carbohydrate concentrations increased in the CMBR with shorter HRT, which required in a high
16
fouling potential due to the positive correlation between EPS and MLSS concentration [43]. A change in hydraulic conditions, such as a permeate flux, had an impact on the EPSp and EPSc concentrations in the CMBRs. However, membrane fouling was not significantly affected. The results were consistent with several researches in the past, which demonstrate that MLSS concentration in the range of 5 to 8 g/L, which is the general practical value in the MBR, does not have a significant effect on membrane fouling [44, 45].
FIGURE 7
A decrease in HRT could significantly reduce the negative charge on the surface of sludge flocs in CMBR (Fig. S2). Surface charge is related to the ionizable groups present on sludge surfaces that induce an increase in the polar interactions of EPS with water molecules. Sludge surfaces with a shorter HRT of 8 h (-1.1 ± 0.5 meq./g VSS) were negatively charged and less hydrophobic than those with a longer HRT of 12 h (-2.5 ± 0.3 meq./g VSS). This was attributed to the ratio of proteins to carbohydrates increasing slightly when the HRT decreased (Fig. 7) because the amino groups in proteins carry positive charges, with the ability to neutralize some of the negative charge of the sludge flocs. The importance of the ratio of proteins to carbohydrates in determining the surface charge of aerobic sludge flocs was reported [38]. It was also evidenced that the hydrophobic fraction of EPS was made up only of proteins, instead of carbohydrates [46].
3.3.
Pilot-scale MLE-CMBR plant
3.3.1. Reactor performance
17
The MLE/CMBR coupled with the developed low-cost ceramic membranes, was configured to improve the nitrogen removal of actual domestic wastewater. Table 1 contains a summary of the overall operating conditions during the last 50 d of operation with an HRT of 4 hr. The influent and effluent COD concentrations in the MLE/CMBR during the study period are presented (Fig. 8). With an average influent COD concentration of 197.0 ± 62.9 mg/L, the effluent COD concentration was 16.3 ± 5.5 mg/L for the MLE/CMBR systems, with an average COD removal rate of 91.6 ± 3.5% and an excellent NH4+-N removal efficiency of 93.2 ± 9.3%, with an average effluent NH4+-N concentration of 2.1 mg/L. The average total N concentration in the MLE/CMBR effluent was 7.3 mg/L N, with a removal rate of 68.8 %, compared to the influent concentration (average value = 23.0 mg/L N).
TABLE 1 FIGURE 8
3.3.2. Membrane flux During short-term MLE/CMBR operations, the permeate flux was set to a constant value of 15 LMH, while the time required for the rapid increase in TMP of MLE/CMBR to cease was approximately 10 days without fouling control, except for membrane relaxation strategies (Fig. 9). The increase in TMP, corresponding to the membrane flux, is commonly attributed to the formation of a cake layer on the membrane surface, in addition to being correlated with biomass characteristics such as EPS, hydrophobicity, and surface charge [17, 47]. After the first 10 days of operation, maximum TMP (0.3 bar) was reached at an MLSS concentration of 6,500 mg/L, making it necessary to initiate physical cleaning strategies, such as back-flushing, in order to
18
keep the filtration process working at less than adequate TMP value. Therefore, the membrane was operated with 300 s cycles (270 s filtration and 30 s relaxation), with constant TMP of 0.28 bar (data not shown). It is a well-known fact that back-flushing affects the economic feasibility of the process [10]. However, higher frequency of back-flush did not improve membrane performance when the MLSS concentration was over 30,000 mg/L [48]. Most of the effort in this chemical-free cleaning strategy was desirable to reduce the time required to reach the maximum TMP in a full-scale system design [49]. The permeate membrane flux was maintained at 20 LMH for the last 20 days of operation, despite the short-term operation (Fig. 9).
4.
Conclusions In this work, a low-cost pyrophyllite-alumina composite ceramic flat-sheet microfiltration
membrane was fabricated and applied to the CMBR for treating the low-strength wastewater. Pyrophyllite, a bulk waste material with limited uses, was used as the main material for the support layers of the ceramic membrane, on which a more negatively-charged coated layer was deposited via dip-coating method, using sub-micron Al2O3 powder suspension. The prepared low-cost pyrophyllite-alumina composite ceramic membrane successfully removed organic matter (COD and TOC) in the CMBR from low-strength domestic wastewater. The physicochemical sludge properties, especially the ratio of proteins to carbohydrates, were affected when the HRT was decreased. The membrane performance was similar to the pyrophyllite-alumina composite ceramic membranes, with a larger effective surface area in the pilot MLE/CMBR study. To summarize, a cost-effective ceramic membrane could be a very competitive candidate for membrane-based treatment of low-strength domestic wastewater. This study is expected to suggest an effective way to recycle waste materials and produce a high-value ceramic membrane
19
for use as the MBR in wastewater treatment with a high effluent water quality. Results offer new options for the recycling of natural waste materials and opportunities for low-cost mineral-based ceramic membranes.
Acknowledgements This work was funded by the Ministry of Environment in Korea as part of ‘The Eco – Innovation Program’ (Grant No. 2014000150021) and in part by the Korea Institute of Science and Technology (KIST) Institutional Research Program (No. 2E27080). The authors are grateful to IB Materials, Co. Ltd., South Korea and Mr. Myoungjoong Kim, for his technical assistance on the ceramic membrane development.
References [1] K. Kimura, R. Nishisako, T. Miyoshi, R. Shimada, Y. Watanabe, Baffled membrane bioreactor (BMBR) for efficient nutrient removal from municipal wastewater, Water Res, 42 (2008) 625-632. [2] X.X. Yan, M.R. Bilad, R. Gerards, L. Vriens, A. Piasecka, I.F.J. Vankelecom, Comparison of MBR performance and membrane cleaning in a single-stage activated sludge system and a twostage anaerobic/aerobic (A/A) system for treating synthetic molasses wastewater, J Membrane Sci, 394 (2012) 49-56. [3] S.J. You, R.A. Damodar, S.C. Hou, Degradation of Reactive Black 5 dye using anaerobic/aerobic membrane bioreactor (MBR) and photochemical membrane reactor, J Hazard Mater, 177 (2010) 1112-1118. [4] J.R. Banu, D.K. Uan, I.T. Yeom, Nutrient removal in an A2O-MBR reactor with sludge reduction, Bioresource Technol, 100 (2009) 3820-3824. [5] J.Y. Shen, R. He, W.Q. Han, X.Y. Sun, J.S. Li, L.J. Wang, Biological denitrification of highnitrate wastewater in a modified anoxic/oxic-membrane bioreactor (A/O-MBR), J Hazard Mater, 172 (2009) 595-600. [6] Z.M. Fu, F.L. Yang, Y.Y. An, Y. Xue, Simultaneous nitrification and denitrification coupled with phosphorus removal in an modified anoxic/oxic-membrane bioreactor (A/O-MBR), Biochem Eng J, 43 (2009) 191-196.
20
[7] H.M. Zhang, X.L. Wang, J.N. Xiao, F.L. Yang, J. Zhang, Enhanced biological nutrient removal using MUCT-MBR system, Bioresource Technol, 100 (2009) 1048-1054. [8] H. Lee, J. Han, Z. Yun, Biological nitrogen and phosphorus removal in UCT-type MBR process, Water Sci Technol, 59 (2009) 2093-2099. [9] H. Monclus, J. Sipma, G. Ferrero, J. Comas, I. Rodriguez-Roda, Optimization of biological nutrient removal in a pilot plant UCT-MBR treating municipal wastewater during start-up, Desalination, 250 (2010) 592-597. [10] Z.H. Liang, A. Das, D. Beerman, Z.Q. Hu, Biomass characteristics of two types of submerged membrane bioreactors for nitrogen removal from wastewater, Water Res, 44 (2010) 3313-3320. [11] C. Abegglen, M. Ospelt, H. Siegrist, Biological nutrient removal in a small-scale MBR treating household wastewater, Water Res, 42 (2008) 338-346. [12] X. Ren, H.K. Shon, N. Jang, Y.G. Lee, M. Bae, J. Lee, K. Cho, I.S. Kim, Novel membrane bioreactor (MBR) coupled with a nonwoven fabric filter for household wastewater treatment, Water Res, 44 (2010) 751-760. [13] R.S. Trussell, R.P. Merlo, S.W. Hermanowicz, D. Jenkins, Influence of mixed liquor properties and aeration intensity on membrane fouling in a submerged membrane bioreactor at high mixed liquor suspended solids concentrations, Water Res, 41 (2007) 947-958. [14] H.Y. Ng, S.W. Hermanowicz, Membrane bioreactor operation at short solids retention times: performance and biomass characteristics, Water Res, 39 (2005) 981-992. [15] L. Jin, S.L. Ong, H.Y. Ng, Fouling control mechanism by suspended biofilm carriers addition in submerged ceramic membrane bioreactors, J Membrane Sci, 427 (2013) 250-258. [16] C.M. Chung, T. Tobino, K. Cho, K. Yamamoto, Alleviation of membrane fouling in a submerged membrane bioreactor with electrochemical oxidation mediated by in-situ free chlorine generation, Water Res, 96 (2016) 52-61. [17] P. Le-Clech, V. Chen, T.A.G. Fane, Fouling in membrane bioreactors used in wastewater treatment, J Membrane Sci, 284 (2006) 17-53. [18] L. Jin, S.L. Ong, H.Y. Ng, Comparison of fouling characteristics in different pore-sized submerged ceramic membrane bioreactors, Water Res, 44 (2010) 5907-5918. [19] S.R. Chae, Y.T. Ahn, S.T. Kang, H.S. Shin, Mitigated membrane fouling in a vertical submerged membrane bioreactor (VSMBR), J Membrane Sci, 280 (2006) 572-581. [20] H.C. Hong, H.J. Lin, R.W. Mei, X.L. Zhou, B.Q. Liao, L.H. Zhao, Membrane fouling in a membrane bioreactor: A novel method for membrane surface morphology construction and its application in interaction energy assessment, J Membrane Sci, 516 (2016) 135-143.
21
[21] S.J. Lee, M. Dilaver, P.K. Park, J.H. Kim, Comparative analysis of fouling characteristics of ceramic and polymeric microfiltration membranes using filtration models, J Membrane Sci, 432 (2013) 97-105. [22] C. Boo, J. Lee, M. Elimelech, Omniphobic Polyvinylidene Fluoride (PVDF) Membrane for Desalination of Shale Gas Produced Water by Membrane Distillation, Environ Sci Technol, 50 (2016) 12275-12282. [23] R.P. Merlo, R.S. Trussell, S.W. Hermanowicz, D. Jenkins, A comparison of the physical, chemical, and biological properties of sludges from a complete-mix activated sludge reactor and a submerged membrane bioreactor, Water Environ Res, 79 (2007) 320-328. [24] T. Nittami, H. Tokunaga, A. Satoh, M. Takeda, K. Matsumoto, Influence of surface hydrophilicity on polytetrafluoroethylene flat sheet membrane fouling in a submerged membrane bioreactor using two activated sludges with different characteristics, J Membrane Sci, 463 (2014) 183-189. [25] C.Q. Zhao, G.Q. Zhang, X.C. Xu, F.L. Yang, Y.S. Yang, Rapidly self-assembled polydopamine coating membranes with polyhexamethylene guanidine: Formation, characterization and antifouling evaluation, Colloid Surface A, 512 (2017) 41-50. [26] A. Asatekin, A. Menniti, S.T. Kang, M. Elimelech, E. Morgenroth, A.M. Mayes, Antifouling nanofiltration membranes for membrane bioreactors from self-assembling graft copolymers, J Membrane Sci, 285 (2006) 81-89. [27] S. Liang, Y. Kang, A. Tiraferri, E.P. Giannelis, X. Huang, M. Elimelech, Highly Hydrophilic Polyvinylidene Fluoride (PVDF) Ultrafiltration Membranes via Postfabrication Grafting of Surface-Tailored Silica Nanoparticles, Acs Appl Mater Inter, 5 (2013) 6694-6703. [28] I.S. Chang, K.H. Choo, C.H. Lee, U.H. Pek, U.C. Koh, S.W. Kim, J.H. Koh, Application of Ceramic Membrane as a Pretreatment in Anaerobic-Digestion of Alcohol-Distillery Wastes, J Membrane Sci, 90 (1994) 131-139. [29] M.P. Zacharof, R.W. Lovitt, The filtration characteristics of anaerobic digester effluents employing cross flow ceramic membrane microfiltration for nutrient recovery, Desalination, 341 (2014) 27-37. [30] J.H. Ha, S.Z.A. Bukhari, J. Lee, I.H. Song, C. Park, Preparation processes and characterizations of alumina-coated alumina support layers and alumina-coated natural materialbased support layers for microfiltration, Ceram Int, 42 (2016) 13796-13804. [31] L. Zhu, M.L. Chen, Y.C. Dong, C.Y.Y. Tang, A.S. Huang, L.L. Li, A low-cost mullitetitania composite ceramic hollow fiber microfiltration membrane for highly efficient separation of oil-in-water emulsion, Water Res, 90 (2016) 277-285. [32] J.H. Ha, Y.H. Park, I.H. Song, The preparation and pore characteristics of an alumina coating on a diatomite-kaolin composite support layer, J Ceram Soc Jpn, 122 (2014) 714-718.
22
[33] APHA, AWWA, WEF, Standard Methods for the Examination of Water and Wastewater, 18th Ed. ed., American Public Health Association, Washington, DC, 1992. [34] B. Frolund, R. Palmgren, K. Keiding, P.H. Nielsen, Extraction of extracellular polymers from activated sludge using a cation exchange resin, Water Res, 30 (1996) 1749-1758. [35] R.S. Trussell, R.P. Merlo, S.W. Hermanowicz, D. Jenkins, The effect of organic loading on process performance and membrane fouling in a submerged membrane bioreactor treating municipal wastewater, Water Res, 40 (2006) 2675-2683. [36] D. Jenkins, From Total Suspended Solids to Molecular Biology Tools-A Personal View of Biological Wastewater Treatment Process Population Dynamics, Water Environ Res, 80 (2008) 677-687. [37] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, Protein measurement with the Folin phenol reagent, J Biol Chem, 193 (1951) 265-275. [38] J.W. Morgan, C.F. Forster, L. Evison, A Comparative-Study of the Nature of Biopolymers Extracted from Anaerobic and Activated Sludges, Water Res, 24 (1990) 743-750. [39] R. Shang, A.R.D. Verliefde, J.Y. Hu, Z.Y. Zeng, J. Lu, A.J.B. Kemperman, H.P. Deng, K. Nijmeijer, S.G.J. Heijman, L.C. Rietveld, Tight ceramic UF membrane as RO pre-treatment: The role of electrostatic interactions on phosphate rejection, Water Res, 48 (2014) 498-507. [40] J.H. Ha, J. Lee, I.H. Song, S.H. Lee, The effects of diatomite addition on the pore characteristics of a pyrophyllite support layer, Ceram Int, 41 (2015) 9542-9548. [41] C. Park, S.W. Hermanowicz, D. Jolis, A novel technique for evaluating foam dynamics in anaerobic digesters, Water Sci Technol, 67 (2013) 2595-2601. [42] C. Park, S.W. Hermanowicz, A Multi-Point Electrical Resistance Measurement System for Characterization of Foam Drainage Regime and Stability, Aiche J, 60 (2014) 3143-3150. [43] F.G. Meng, H.M. Zhang, F.L. Yang, Y.S. Li, J.N. Xiao, X.W. Zhang, Effect of filamentous bacteria on membrane fouling in submerged membrane bioreactor, J Membrane Sci, 272 (2006) 161-168. [44] L. Duan, Y.H. Song, H.B. Yu, S.Q. Xia, S.W. Hermanowicz, The effect of solids retention times on the characterization of extracellular polymeric substances and soluble microbial products in a submerged membrane bioreactor, Bioresource Technol, 163 (2014) 395-398. [45] A. Cosenza, G. Di Bella, G. Mannina, M. Torregrossa, The role of EPS in fouling and foaming phenomena for a membrane bioreactor, Bioresource Technol, 147 (2013) 184-192. [46] F. Jorand, F. Boue-Bigne, J.C. Block, V. Urbain, Hydrophobic/hydrophilic properties of activated sludge exopolymeric substances, Water Sci Technol, 37 (1998) 307-315.
23
[47] H.J. Lin, K. Xie, B. Mahendran, D.M. Bagley, K.T. Leung, S.N. Liss, B.Q. Liao, Factors affecting sludge cake formation in a submerged anaerobic membrane bioreactor, J Membrane Sci, 361 (2010) 126-134. [48] A. Robles, M.V. Ruano, J. Ribes, J. Ferrer, Factors that affect the permeability of commercial hollow-fibre membranes in a submerged anaerobic MBR (HF-SAnMBR) system, Water Res, 47 (2013) 1277-1288. [49] R. Lee, P.L. McCarty, J. Bae, J. Kim, Anaerobic fluidized membrane bioreactor polishing of baffled reactor effluent during treatment of dilute wastewater, J Chem Technol Biot, 90 (2015) 391-397.
24
FIGURE CAPTIONS Figure 1 Figure 2
Figure 3 Figure 4 Figure 5
Figure 6 Figure 7 Figure 8 Figure 9
Schematic diagram of the membrane bioreactor submerged pyrophyllitealumina composite ceramic membranes. SEM images of (a) before the alumina coating and (b) after the alumina coating on the surfaces of the pyrophyllite-alumina composite ceramic membrane support layers. Pore size distributions for the support and coated layers of the pyrophyllitealumina composite ceramic membrane Zeta-potentials of the pyrophyllite-alumina composite ceramic membrane suspensions as a function of pH from 3 to 10 XPS analysis of the surface of the pyrophyllite-alumina composite ceramic membranes. Fractions of carbon, nitrogen, oxygen, silica, and aluminum relative to the sum of elements present at the surface of (a) uncoated ceramic membrane support layer and (b) alumina-coated ceramic membrane Removal efficiencies of COD and TOC in the CMBR at different HRTs Concentrations of protein, carbohydrate and total in EPS for the CMBR at different HRTs Influent and effluent COD concentrations in the MLE/CMBR systems during the period of pilot-scale study Variations in permeate flux for pilot-scale pyrophyllite-alumina composite ceramic membranes during the operation of MLE/CMBR.
25
FIGURE 1
26
FIGURE 2 (a)
(b)
27
FIGURE 3
Differential Intrusion (ml/g)
1.0 support layer coated layer
0.8
0.6
0.02
0.01
0.00
0.4
0.05
0.1
0.2
0.5
0.2
0.0 0.01
0.1
1
Pore Size ( m)
28
10
FIGURE 4
Zeta Potential (mV)
20
0
-20
-40
-60
-80 2
4
6
8
pH
29
10
FIGURE 5
60
(b)
Area Fraction (%)
(a) 40
20
0 C
N
O
Al
Si
C
Element
N
O
Element
30
Al
Si
FIGURE 6
Removal Efficiency (%)
100
COD
TOC
95
90
HRT = 12h
HRT = 8h
31
EPS Concentration (mg/g VSS)
FIGURE 7
40 HRT = 12h HRT = 8h
30
20
10
0 Proteins Carbohydrates Total EPS
32
FIGURE 8
COD Concentration (mg/L)
500 COD influent MLE/CMBR permeate 400
300
200
100
0 0
10
20
30
Days of Operation
33
40
50
FIGURE 9
Flux (Lm2hr-1)
20
15
10 0
10
20
30
Days of Operation
34
40
50
Graphical abstract
35
Table 1. Operating conditions of the pilot-scale MLE/CMBR system at actual domestic water resource recycle facility (WRRF) Factor
MLE/CMBR
Volume (L)
174.0
Flow rate (L/hr)
43.1±7.0 2
-1
Permeate Flux (Lm hr )
15.2±2.6
HRT (h)
4.3±1.5
SRT (d)
15.0
Aeration flow rate (L/min)
2.0 2
Effective membrane area (m )
2.9
pH
6.9±0.2
DO (mg/L)
2.6±0.4 27.8±1.6
Temp. (℃) 3
OLR (kg COD/m ·d)
0.04±0.01 Anoxic
MLSS (mg/L)
Aerobic
4,048±2,599
6,418±1,519
MLVSS (mg/L)
2,848±750
4,700±1,195
MLVSS/MLSS
0.70
0.73
36
Highlights
Waste pyrophyllite material was utilized in ceramic membrane developments
New ceramic membranes applied to laboratory-scale MBR for treating low-strength domestic wastewater
Scaled-up ceramic membranes were operated successfully up to 20 LMH in MLE/CMBR
37