60
Zhang, W., Ding, W. and Ying, a.W. (2013) Biological Activated Carbon Treatment for Removing BTEX from Groundwater. journal of environmental Engineering 139(10), 1246-1254.
Zhao, Y., Liu, D., Huang, W., Yang, Y., Ji, M., Nghiem, L.D., Trinh, Q.T. and Tran, N.H. (2019) Insights into biofilm carriers for biological wastewater treatment processes: Current state-of-the-art, challenges, and opportunities.
Bioresource Technology 288, 121619.
Presenting Author
Dr. Abdullah Ibrahim Lecturer
Department of Environmental Engineering, University of Mosul, Mosul, Iraq Is the presenting author an IWA Young Water Professional? Y/N N
Bio: I have received my M.Sc. Environmental Engineering from Mosul University, Iraq in 2004 then I worked as a faculty member in Civil Engineering department 2005-2013. As well as, I was one of funders and faculty members of Environmental Engineering department 2011-2013.
In 2014, I have joined Kansas University, USA as a PhD student in Environmental
Engineering field. I received my PhD in Spring 2020 in the field of biological treatment of gas and oilfield produced water with emphasis on manipulation of biofilm technology to treat wastewater under harsh conditions.
Biofilms Control Session
61
Role of dynamic membrane biofilm development on chain elongation for medium chain carboxylic acids production
Shrestha, S., Kitt, D., Song, H. and Raskin, L.
*Department of Civil and Environmental Engineering, University of Michigan, United States
Keywords: medium chain carboxylic acids; chain elongation; anerobic dynamic membrane bioreactor, biofilm
Summary of key findings
The formation of a dynamic membrane biofilm achieved a high solids removal efficiency of ≥ 90% and enhanced medium chain carboxylic acids (MCCAs) production via chain elongation. Good permeate quality with low suspended solids was sustained for a period of 221 days without membrane
replacement or cleaning. The dynamic membrane also provided conducive conditions for the enrichment of select MCCAs producers.
Background and relevance
Chain elongation is an emerging anaerobic biotechnology that uses mixed microbial communities for organic waste stream conversion into medium-chain carboxylic acids (MCCAs). MCCAs are platform chemicals with diverse industrial and agricultural applications [1]. A product recovery system is needed to recover MCCAs in a useful form. Membrane based liquid-liquid extraction (LLX) is the most commonly used approach [2, 3]; however, LLX requires suspended solids removal from the bioreactor effluent to avoid membrane fouling. Most chain elongation papers have used multiple external filters for solid-liquid separation before MCCAs extraction [2, 3]. Alternatively, the use of membrane bioreactors for MCCAs production with simultaneous solids removal can enable direct integration with the downstream separation process. An anaerobic dynamic membrane bioreactor (AnDMBR) integrated with an LLX unit was developed to evaluate MCCAs production from pre- fermented food and ethanol-rich brewery waste and produce a low suspended solids effluent
(permeate). The AnDMBR was equipped with stainless steel meshes that are cheaper than conventional polymeric membranes and allow the development of a biological cake layer [4], also referred to as a
“dynamic membrane biofilm”, on the mesh surface to provide effective filtration.
AnDMBR also has the potential to address some of the shortcomings of conventional anaerobic membrane bioreactors due to its low membrane module cost, high membrane flux, ease of membrane
62
fouling control, and low energy requirements [4]. Furthermore, other studies have shown that biomass retention results in a higher conversion rate for MCCAs production due to high cell density and higher resilience towards upsets [5, 6]. Therefore, besides producing high quality permeate, the formation of a dynamic membrane biofilm also has the potential to improve MCCAs production. 16S rRNA gene and 16S rRNA sequencing were employed to compare the microbial community in the suspended biomass and the dynamic membrane biofilm samples as well as study the contribution of the dynamic
membrane biofilm towards MCCAs production.
Results
The AnDMBR was operated in a semi-continuous filtration mode with frequent backwashing until Day 49 during which time the permeate total suspended solids (TSS) concentration remained high (0.81 ± 0.52 g L-1, Figure 1). The permeate TSS decreased from 0.82 ± 0.02 g L-1 on Day 48 to 0.17 ± 0.01 g L-1 on Day 51 after switching to a continuous filtration mode. The AnDMBR produced a permeate with low suspended solids (0.12 ± 0.06 g TSS L-1) for an extended period of 221 days
63
without membrane cleaning, with the lowest TSS concentration of 0.04 ± 0.01 g L-1 achieved on Day 69 (Figure 1). The average bioreactor TSS concentration during this period was two orders of magnitude higher (21.6 ± 9.9 g L-1) than the permeate TSS. A high TSS removal efficiency of 94.6 ± 5.4% was achieved with a ratio of solid retention time to hydraulic retention time of 12.9 ± 7.4 due to high biomass retention by the dynamic membrane formation.
The microbial community in the biofilm was characterized and compared to that in the suspended bioreactor samples. A few dominant microbial groups such as Methanobrevibacter (56.9%), Pseudoramibacter (15.8%), and unclassified Bacteria (6.1%) represented the majority of the active microbial community in the biofilm on Day 41. However, there was a shift in the microbial
community on Day 89 when the relative activity of Methanobrevibacter, Pseudoramibacter, and unclassified_Bacteria decreased while the relative activity of Lactococcus, Megasphaera, Prevotella, and Pseudomonas increased (Figure 2). Lactobacillus, a lactic acid bacterium, was active in the biofilm samples throughout operation with varying relative abundance (0.6-4.6%) and activity (0.4- 45.0%) levels. Clostridiales_unclassified was found at a higher relative activity of 17.2 ± 1.5% from Day 222-271 in the biofilm samples compared to the relative activity of 4.5 ± 2.8% in the suspended samples (Figure 2). The relative activity of Clostridiales_unclassified was significantly correlated (correlation coefficient=0.54, p=0.02) with the MCCA production in the bioreactor. The permeate MCCA concentrations were significantly higher than in the reactor (p=8.2E-05) during this time period.
Discussion
Despite high suspended solids concentration in the reactor, the AnDMBR was able to produce a permeate with low suspended solids. The continuous filtration mode used in this system resulted in a well-formed dynamic membrane biofilm as confirmed by visual observation (Figure 3) and was responsible for the effective filtration observed in this study (Figure 1). The microbial community shift on Day 89 (Figure 2) corresponded to the formation of a stable dynamic membrane biofilm. Although the microbial community membership of the suspended and biofilm samples was similar, the relative activity of MCCA producers such as Clostridiales_unclassified was higher in the biofilm compared to that in the suspended bioreactor samples. Clostridiales have been reported to be present in other ethanol chain elongation studies [3, 7]. The significantly higher MCCA concentrations in the permeate
compared to those in the bioreactor and higher relative activity of MCCA producers such as
Clostridiales_unclassified in the biofilm indicate that biofilm formation contributed to chain elongation.
High spatial organization and different environmental conditions in the dynamic membrane layer may have led to niche differentiation, providing favorable conditions to one population over another.
Moreover, Lactobacillus was also observed at a high relative activity in the dynamic membrane layer.
Lactobacillus produces extracellular polymeric substances (EPS) as a metabolic product of
carbohydrate degradation. As EPS plays an important role in microbial biofilm formation promoting cell adhesion in the dynamic membrane layer [8], it possibly explains the dominance of Lactobacillus in the dynamic membrane layer.
In conclusion, the AnDMBR system was capable of efficient suspended solids removal, which is necessary for the optimal operation of the downstream extraction unit. In terms of cost and
environmental impacts, this integrated approach is superior to other MCCA systems that use multiple external filtration steps before the extraction unit. The dynamic membrane also harbored a specialized microbial community enriched in active MCCA and EPS producing microbial groups. Therefore, the dynamic membrane formation not only improves permeate quality but also plays a significant role in MCCA production.
Acknowledgements
This work was supported by the U.S. National Science Foundation (Sustainability Research Networks 1444745), University of Michigan Integrated Training in Microbial Systems Fellowship and Rackham Postdoctoral Fellowship, and Water Federation Canham Graduate Studies Scholarship. We would like
64
to thank Steve Donajkowski and Ethan Kennedy for their technical assistance with bioreactor construction.
Figure 1. TSS concentration in the permeate ( ) and reactor ( ) over time. The vertical dashed line represents a switch to a continuous filtration mode on Day 50.
Figure 2. Relative activity of the top 15 most abundant microbial groups in each sample classified to the genus or family level in the bioreactor and biofilm samples. The top 15 microbial groups represent 79-99% of the total relative activity.
The relative activity was determined as percentages normalized to the total number of 16S rRNA sequences.
Figure 3. Dynamic membrane formation over time References
Angenent, L. T., Richter, H., Buckel, W., Spirito, C. M., Steinbusch, K. J., Plugge, C. M., ... & Hamelers, H. V. (2016) Chain elongation with reactor microbiomes: open-culture biotechnology to produce biochemicals, Environmental science &
technology, 50(6), 2796-2810
Kucek, L. A., Spirito, C. M., & Angenent, L. T. (2016) High n-caprylate productivities and specificities from dilute ethanol and acetate: chain elongation with microbiomes to upgrade products from syngas fermentation. Energy &
Environmental Science, 9(11), 3482-3494.
65
Kucek, L. A., Xu, J., Nguyen, M., & Angenent, L. T. (2016) Waste conversion into n-caprylate and n-caproate: resource recovery from wine lees using anaerobic reactor microbiomes and in-line extraction, Frontiers in microbiology, 7, 1892.
Ersahin, M. E., Ozgun, H., Dereli, R. K., Ozturk, I., Roest, K., & van Lier, J. B. (2012). A review on dynamic membrane filtration: materials, applications and future perspectives. Bioresource Technology, 122, 196-206.
Carvajal-Arroyo, J. M., Candry, P., Andersen, S. J., Props, R., Seviour, T., Ganigué, R., & Rabaey, K. (2019) Granular fermentation enables high rate caproic acid production from solid-free thin stillage. Green Chemistry, 21(6), 1330-1339.
Roghair, M., Strik, D. P., Steinbusch, K. J., Weusthuis, R. A., Bruins, M. E., & Buisman, C. J. (2016) Granular sludge formation and characterization in a chain elongation process, Process Biochemistry, 51(10), 1594-1598.
Agler, M. T., Spirito, C. M., Usack, J. G., Werner, J. J., & Angenent, L. T. (2012). Chain elongation with reactor
microbiomes: upgrading dilute ethanol to medium-chain carboxylates. Energy & Environmental Science, 5(8), 8189- 8192.
Ersahin, M. E., Tao, Y., Ozgun, H., Spanjers, H., & van Lier, J. B. (2016). Characteristics and role of dynamic membrane layer in anaerobic membrane bioreactors. Biotechnology and Bioengineering, 113(4), 761-771.
66
USING SHEAR RHEOMETRY AND IMAGE ANALYSIS TO ASSESS ENZYMES FOR BIOFILM CONTROL
Nahum, Y.*, Li, M.* and Nerenberg, R.*
*Department of Civil and Environmental Engineering and Earth Sciences, University of Notre Dame, Notre Dame, Indiana
Keywords: Biofilm disruption, Enzymes, Mechanical Properties Summary of key findings
We used shear rheometry and optical coherence tomography (OCT) to assess the effectiveness of two enzyme mixes, a protease and glucanase, on biofilm removal. The protease mix was more effective than the glucanase mix, removing 70% of the biofilm in 3 hours under fluid flow conditions and weakening the mechanical properties of the biofilm by a 30% under rheometer tests. The effect increased with higher reaction times. The protease mix was effective in reducing the integrity of the EPS matrix and promote the removal of biofilms grown under continuous fluid flow conditions.
Background and relevance
Biofilms are aggregates of bacteria embedded in extracellular polymeric substance (EPS) (Flemming, et al., 2010). The EPS matrix acts as a protective barrier against harsh environments by providing mechanical strength and stability (Seviour, et al., 2019). Current biofilm removal strategies, such as mechanical cleaning or chemical biocides, have limitations for many practical systems. In recent years, the use of enzymes for biofilm control has raised interest as a green alternative. These target and degrade specific EPS constituents, destabilizing the biofilm structure and dispersing bacteria (Flemming , et al., 2007). However, most past studies have focused on young and thin biofilms grown under stagnant conditions and for short amount of time. Also, the mechanisms of removal and time- dependency of enzymatic actin is not known.
A mix of protease enzymes and a mix of glucanase enzymes were selected for testing, given that proteins and polysaccharides are considered primary EPS components (Fulaz, et al., 2019). The two enzymes mixes were characterized based on the effect on biofilm mechanical properties. To further understand the mechanisms of removal, the enzymes were tested on biofilms grown in a flow cell under conditions simulating actual biofilm development. The concentration and exposure times of the enzymes were varied, and the effects were analyzed using rheometry and OCT.
Results
Enzymatic disruptors showed up to a 30% weakening effect of the elastic and viscous behavior, depending on the enzymes used. Rheometry tests, shown in Figure 1, showed that the protease mix had a greater weakening the both storage modulus (G’) and loss modulus (G’’) of the biofilms, while the glucanase mix did not have a measurable effect. The weakening effect increased with time for the protease mix.
67
Figure 1. Changes in mechanical properties while using glucanase and protease enzymes compared to phosphate buffer media as a negative control. A. changes in storage modulus or elastic properties of the biofilm. B. Changes in loss modulus or viscous properties of the biofilm.
In terms of biofilm disruption, biofilms were grown in 3-channels flow cells under continuous flow conditions. 100ppm of enzyme was added continuously at the same flow rate after 5 to 6 days. Figure 1Figure 2 shows the comparison between the protease mix enzyme and the glucanase mix of enzymes.
Consistent with the rheometer tests, the protease mix had a higher efficiency, removing up to 70% of the biofilm in the first 3 hours.
Figure 2. Effect of glucanase and protease enzymes on biofilm disruption using 3- channel flow cells with a flowrate of 3ml/h. The results were compared with the addition of Phosphate buffer media as a negative control.
The flow cell channels were continuously monitored by OCT, and images from the initial and final state are shown in Figure 3.
Figure 3. disruption of biofilm test under Optical Coherence Tomography (OCT) over 5h, using protease mix enzyme (a. and b.) and glucanase mix enzyme (c. and d.)
68 Discussion
The protease mix had 30% efficiency in weakening the biofilm mechanical properties and removed 70% of the biofilm in the flow cell, while the glucanase mix did not have measurable effects on decreasing the strength of the biofilm and also did not remove much biofilm, compared to the control without enzyme addition. Based on the results, the EPS of this biofilm may have had proteins as major structural components, the addition of proteases reduced the stability and strength of the EPS matrix by cleaving protein linkages. On the other hand, it is surprising that the glucanase group of enzymes did not have a measurable effect neither in weakening the mechanical properties nor removing the biofilm, as it is known that polysaccharides are also main EPS components (Fulaz, et al., 2019). However, it is possible they were not the main structural components of the matrix and therefore the mechanical properties were not compromised. Further investigation of the EPS composition of the biofilm and role in biofilm mechanical properties is needed.
Important findings include the fact that compromising the mechanical properties of the biofilm will likely induce biofilm removal, shown by rheometer tests and consistent with the flow cells tests. Additionally, it was shown that enzymes can remove real biofilms, grown under continuous flowing conditions instead of stagnant and thin biofilms. These findings could potentially be used on a mathematical model to predict the effect of different groups of enzymes and their impact on the mechanical properties of biofilms, providing an important tool for designing and predicting biofilm control strategies in engineering systems.
References
Flemming H. C., Neu T. R. and Wozniak D. The EPS matrix: the house of biofilm cells [Journal] // J.
Bacteriol.. - 2007. - Vol. 189.
Flemming H.-C. and Wingender J. 2010. The biofilm matrix. [Journal] // Nat. Rev. Microbiol.. - 2010. - Vol. 8. - pp. 623–633.
Fulaz S., Vitale S. and Quinn L. Nanoparticle–Biofilm Interactions: The Role of the EPS Matrix [Journal]
// Trends in Microbiology. - 2019. - Vols. 27, No. 11. - pp. 915-926.
Seviour T. [et al.] Extracellular polymeric substances of biofilms: Suffering from an identity crisis [Journal] // Water Research. - 2019. - Vol. 151. - pp. 1-7.
Xavier J. B., Picioreanu C. and van Loosdrecht M A general description of detachment for multidimensional modelling of biofilms [Journal] // Microbiology. - 2005. - Vol. 151. - pp. 3817–3832.
Presenting Author
Yanina Nahum
PhD Student, Nerenberg Lab University of Notre Dame
Is the presenting author an IWA Young Water Professional? Y (i.e. an IWA member under 35 years of age)
Bio: Yanina is a 3rd year PhD student working on novel ways to disrupt detrimental biofilms and how the disruptors affect biofilm mechanical properties.
69
Reducing the impacts of biofouling in reverse osmosis membrane systems through low fluence pretreatment employing UVC-LED irradiation
Sperle, P.*, Wurzbacher, C.*, Skibinski, B.*, Drewes, J.E.*
* Chair of Urban Water Systems Engineering, Technical University of Munich, Am Coulombwall 3, 85748 Garching, Germany
Keywords: Biofouling; reverse osmosis membranes; UVC disinfection Summary of key findings
Low fluence UVC pretreatment using LEDs is capable of significantly delaying biofilm formation in reverse osmosis (RO) systems. The treatment seems to have a sustained effect on the biofilm
composition, leading to reduced hydraulic resistance. Some bacteria genera showed a high correlation to hydraulic resistance. These findings provide further insights into changes in microbial community and their impact on membrane biofouling. Those insights can further be used in biofilm morphology engineering. In addition, the potential of low fluence UV pretreatment is elucidated.
Background and relevance
Reverse osmosis (RO) membrane filtration is an important process for the treatment and purification of water. The main drawback of this technology is the reduction of treatment performance by membrane biofouling. Biofouling is caused by the formation of microbial biofilms on the membrane surfaces, leading to major problems in numerous membrane systems (Flemming 2020). These include operational issues like a higher pressure drop between feed and concentrate, a reduced membrane permeability, as well as lower salt rejection. To mitigate biofouling, a RO system requires an adequate pretreatment. Currently, appropriate pre-filtration and/or the dosage of biocides are considered state- of-the-art treatment options. However, as the EU regulation 528 2012 limits the use of biocides (European Parliament and Council 2012), there is a great need to develop biocide-free alternative biofouling mitigation strategies.
An alternative pretreatment strategy for biofouling control could be UV disinfection of the feed stream.
Applying UVC irradiation using medium pressure lamps as a pretreatment has already successfully been implemented as a biofouling control strategy using low or medium pressure lamps, with fluences of ≥ 400 J/m² (Harif et al. 2011; Martino et al. 2011; Marconnet et al. 2011; Otaki et al. 1998). In these studies, it could be proven that the UV disinfection is delaying the buildup of the biofilm in
nanofiltration (NF) or RO membrane systems using river or groundwater as feed in a full- scale application (Harif et al. 2011; Marconnet et al. 2011; Martino et al. 2011). Biofilms grown under a defined period of time exhibited a lower total and active cell number and dry weight, as well as a reduced adenosine triphosphate (ATP), protein and polysaccharides content (Marconnet et al. 2011). In addition, Harif et al. (2011) recognized a lower specific extracellular polymeric substance (EPS) production per biovolume and an altered microbial community. Nevertheless, since UV treatment does not provide a residual disinfection effect, it has seen no widespread application (Harif et al. 2011).
Furthermore, it is not clear, whether the observed changes in the biofilm properties are only linked to a delay of the biofilm formation or if the UV pretreatment indeed has a sustained effect on the biofilm composition, as the experiments were terminated at a defined period of time (Harif et al. 2011;
Marconnet et al. 2011; Martino et al. 2011), not allowing to build the same extent of biofouling.
Particularly with the advancements in UV-LED technology, recently available LEDs in the UV-C wavelength range might offer the possibility to integrate UV pretreatment in existing RO membrane systems. On the one hand, as LEDs are compact, robust and do not contain toxic mercury like
conventional low-pressure UV lamps, they offer new, potentially more energy efficient reactor designs (Song et al. 2016). On the other hand, even though the external quantum efficiency of UVC-LEDs is
70
improving, due to the inhomogeneous irradiation patterns and the low power of UV LEDs (Chen et al.
2017), high fluence rates in UVC-LED reactors might be difficult to achieve. Therefore, in this study we investigated if low fluence irradiation using UVC-LEDs is sufficient to control biofouling and if there is a lasting effect at a severe biofouling state.
Results
To investigate the potential of the UV pretreatment for biofouling control, accelerated biofouling experiments, using drinking water fed with acetate and membrane fouling simulators (12 x 2 cm), were performed. Diamond-shaped 20 mil spacer and the Oltremare (Italy) LOW 1 membrane were employed within the fouling simulators. UV fluences were set to approximately 30 J/m². In each experiment, a treatment train receiving the UV pretreatment was compared to a reference train without UV. The development of feed channel pressure drop (FCPD) and the decline in relative membrane permeability for a representative experiment comparing no UV and UV pretreatment is depicted in Figure 2.1. In summary, the UV pretreatments showed a significant delay of biofilm formation by > 15
%, as well as a > 30 % reduced hydraulic resistance at a severe biofouling state (FCPD of 67 mbar/cm).
Membrane autopsy and biofilm analysis results showed a significant reduction of ATP level, as well as fewer cells with damaged membranes. However, the number of intact cells, total organic carbon and protein and polysaccharide content of the EPS only exhibited a declining trend, but no significant difference could be proven (Figure 2.2). Furthermore, the protein composition (tyrosine- and
tryptophan-like proteins) showed no difference using 3D-fluorescence spectroscopy and parallel factor analysis.
Within the analysis of the microbial community composition (using 16S rRNA amplicon sequencing) of the treated and untreated biofilm, certain differences could be detected. A summary of the observed bacteria families is given in Figure 2.3. Whereas the Acidovorax sp. and Ralstonia sp. strains were present with increased abundance in UV pretreated biofilms, if present, a strain of Delftia sp. was reduced. When correlating the individual taxa to the biofilm resistance, especially Aquabacterium sp.
showed a high Pearson correlation (R²=0.86, p= 0.0003). However, the occurrence of Aquabacterum sp. does not seem to differ significantly between the treated and untreated biofilm.
Discussion
In these biofouling experiments, it could be demonstrated that a rather low UV fluence using LEDs is sufficient for biofouling control. Biofilm buildup is delayed and the hydraulic resistance is reduced. If those results can be translated to up-scaled applications during long term operation, this pretreatment could be an effective biofouling control strategy, allowing easy retrofitting as well as, saving energy and chemicals.
The mechanisms behind the delay of biofilm formation are suspected to be inactivation of the bacteria in the feed flow, as well as changing their adsorption properties or introducing a certain stress factor.
Concerning the reduced hydraulic resistance, the reduced number of damaged cells indicates a more porous biofilm. Lower ATP levels lead to the assumption that the active microbial biomass was diminished due to the UV pretreatment. Furthermore, the observed change in the microbial community seems likely to affect the hydraulic resistance of the formed biofilm. It seems that certain genera affect the hydraulic resistance of biofilms more than others. As the observed differences in the biofilm properties were monitored at a severe biofouling state, it is assumed that at least for the lab-scale experiments, the effect of UV pretreatment on the biofilm is sustained. However, since our experiment worked with an accelerated biofilm formation, these requires further investigation
71
Figure 2.1 Mean pressure drop (FCPD) and permeability of the reverse osmosis (RO) systems with and without UV for a representative experiment; shaded areas represent the 95 % confidence intervals
Figure 2.2 Summary biofilm analysis; bar plots with 95 % confidence T intervals (n=6); biofilm resistance for biofilms with biofilm analysis (n=5)
Figure 2.3 Summary of the biodiversity of treated and not treated biofilms
72 References
Chen, Jian; Loeb, Stephanie; Kim, Jae-Hong (2017): LED revolution: fundamentals and prospects for UV disinfection applications. In: Environ. Sci.: Water Res. Technol. 3 (2), S. 188–202. DOI: 10.1039/c6ew00241b.
European Parliament and Council (2012): Concerning the making available on the market and use of biocidal product. REGULATION (EU) No 528/2012. In: Official Journal of the European Union (L 167/1).
Flemming, Hans-Curt (2020): Biofouling and me: My Stockholm syndrome with biofilms. In: Water research 173, S. 115576. DOI: 10.1016/j.watres.2020.115576.
Harif, Tali; Elifantz, Hila; Margalit, Eli; Herzberg, Moshe; Lichi, Tovit; Minz, Dror (2011): The effect of UV pre- treatment on biofouling of BWRO membranes: A field study. In: Desalination and Water Treatment 31 (1-3), S.
151–163. DOI: 10.5004/dwt.2011.2377.
Marconnet, C.; Houari, A.; Seyer, D.; Djafer, M.; Coriton, G.; Heim, V.; Di Martino, P. (2011): Membrane biofouling control by UV irradiation. In: Desalination 276 (1-3), S. 75–81. DOI: 10.1016/j.desal.2011.03.016.
Martino, Di; Ahmed, Houari; Veronique, Heim; Cyril, Marconnet (2011): Assessment of UV Pre-Treatment to Reduce Fouling of NF Membranes. In: Robert Y. Ning (Hg.): Expanding Issues in Desalination: InTech.
Otaki, M.; Takizawa, S.; Ohgaki, S. (1998): Control and modeling of membrane fouling due to microorganism growth by UV pretreatment. In: Water Science and Technology 38 (4-5). DOI: 10.1016/S0273-1223(98)00539- 3.
Song, Kai; Mohseni, Madjid; Taghipour, Fariborz (2016): Application of ultraviolet light-emitting diodes (UV-LEDs) for water disinfection: A review. In: Water research 94, S. 341–349. DOI: 10.1016/j.watres.2016.03.003.
Presenting Author
M.Sc. Sperle Ph.D. candidate
Chair of Urban Water Systems Engineering, Technical University of Munich, Am Coulombwall 3, 85748 Garching, Germany
Is the presenting author an IWA Young Water Professional? Y/N (i.e. an IWA member under 35 years of age)
N
Bio: After finishing his master degree in environmental engineering, Philipp Sperle started his position as a Ph.D. student and doctoral candidate in December of 2018 at the Chair of Urban Water Systems Engineering at the Technical University of Munich. His research focus covers membrane hybrid processes, in particular those combining a membrane process with UV-LED radiation, as well as biofouling and the characterization of UVC-LED.
73
Biofilm growth control with minor maintenances for an optimized operation of gravity-driven membrane filtration with hollow fiber membranes.
Jacquin, C.*, Stoffel, D.*, Rigo, E.*, Derlon, N.*, Staaks, C.**, Heijnen, M.** and Morgenroth, E.*,***
* Eawag, Swiss Federal Institute of Aquatic Science and Technology, Überlandstrasse 133, 8600 Dübendorf, Switzerland.
** inge-DuPont, Flurstraße 27, 86926 Greifenberg, Germany.
*** ETH Zürich, Institute of Environmental Engineering, 8093 Zürich, Switzerland.
Keywords: Biofilm growth control, gravity-driven membrane filtration, hollow fibers, biopolymer biodegradation.
Summary of key findings
Our study showed that controlling the biofilm growth during the operation of inside-out hollow fiber membranes in gravity-driven membrane filtration mode is crucial to maintain high fluxes. Indeed, matter accumulation within the fiber’s lumen leads to flux drop. The smaller the fiber inner diameter (ID), the more filtration performances were reduced. Forward flush (FF), relaxation + forward flush (R+FF) and backwash were therefore implemented to mechanically remove the excess of biofilm inside the fibers. It was shown that all maintenances helped to control the biofilm growth and particle accumulation, but backwash was the most efficient at draining out the accumulated matter within the fiber and therefore maintaining higher fluxes. Interestingly, removing the biofilm mechanically only influenced the flux, while the biopolymer removal was not impacted for both types of ID.
Finally, we could observe that applying a daily backwash to remove the excess biofilm is a strategical approach to maintain fluxes around 10 L/h/m2 and around 90 % biopolymer removal with HF membranes having a restricted space available for biofilm growth (ID=0.9 mm). This interesting result is a first step towards the implementation of GDM filtration in drinking water treatment plants, where HF membranes operated in inside-out mode are preferred over outside-in HF or flat-sheet membranes.
Background and relevance
Gravity-driven membrane (GDM) filtration tolerates biofilm growth at the surface of ultrafiltration membranes (Peter-Varbanets et al., 2010). This approach is strategical to decrease maintenance costs and to increase biopolymer removal due to biofilm biodegradation (Pronk et al., 2019). Despite these benefits, GDM filtration is currently limited to decentralized systems (Boulestreau et al., 2012; Frechen et al., 2011;
Peter-Varbanets et al., 2017), and most of the studies were performed using flat sheet (FS) membranes.
However, hollow fiber (HF) membranes operated in inside-out mode are more commonly used in water treatment centralized systems because of their higher specific membrane filtration per module unit volume and lower capital cost, in comparison with FS membranes (Tang et al., 2016; Akhondi et al., 2017; Chawla et al., 2017). Nevertheless, while several studies showed that flux stabilization was possible with HF operated in outside-in mode, only one study was performed with HF operated in inside-out mode (Chawla et al., 2017). This research gap is due to the fact that the smaller the space available for biofilm growth, the lower the flux and the higher the clogging risk (Wu et al. 2016, 2017, Pronk et al., 2019), limiting the interest of using inside-out HF in GDM filtration if no biofilm growth control is implemented.
Our study therefore aimed at:
Evaluating the feasibility of GDM filtration with HF membranes of different inner diameters.
Optimizing the biofilm growth control to maintain high fluxes.