5 results
Filters
Settings
Now showing1 - 5 of 5
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Development and performance of Layer-by-Layer modified hollow fiber membranes as capillary nanofiltration(2025); The need for alternative water sources due to water stress or changes in raw water qualities for use as drinking water often requires more advanced treatment techniques than currently in use. Thus, there is a demand for resource and energy efficient processes, with high process stability and reliability. Within this work the potential of Layer-by-Layer (LbL) modified hollow fiber ultrafiltration (UF) membranes was examined as one option in terms of ion rejection, rejection of dissolved organics, molecular weight cut-off (MWCO), fouling, and mechanical and process stability. Furthermore, a model was derived to differentiate between predominant polyelectrolyte (PE) build up during modification in terms of pore vs layer dominating PE multilayer formation. The results also identified the PE layering location as another crucial factor during membrane modification. Filtration in lab scale as well as pilot scale results in water works achieved high rejection rates for divalent ions (> 90 % for sulfate and magnesium, and > 80 % for calcium). However, some external parameters were investigated which influenced the resulting ion rejection. As shown in lab scale, high ionic strength in the feed solutions led to swelling of the PE structure, resulting in a change of separation characteristics. Nevertheless, swelling was highly dependent on the present type of ions and respective concentration. Model supported results clearly identified concentration polarization (CP) through the laminar boundary layer as dominating factor for the removal efficiency for sulfate as a model substance for divalent ions. Besides crossflow velocity and resulting laminar boundary layer thickness, model results showed a severe influence of CP dependent on membrane length. This underscores the importance of additional experiments beyond the lab to pilot scale on industrial length modules. Further investigations in lab scale showed that the modification of the membranes resulted in an MWCO in the lower range of NF membranes, providing high removal rates of dissolved organic substances. SAC254 removal rates of > 90% and TOC removal of > 80 % could be achieved even in Dead End operation for solutions containing natural organic matter (NOM). Fouling and removal rates were dependent on the NOM composition and their molecular weight distribution. Hereby, the solution containing a higher share of larger molecules reached higher rejection, while impact of fouling was lower. It is attributed to the deposition of foulants on top of the membrane surface instead of internally within the modified membrane structure. Membranes could successfully withstand the stress of hydraulic backwash (hydr. BW), though it was limited to a maximum hydr. BW flux of 50 L/(m² h). Combined with the good chemical stability, it would allow the implementation of a regular CEB, which was highly efficient for the removal of NOM foulants off the membrane surface. Overall, the LbL modification of hollow fiber UF membranes was successful on lab scale and on industrial scale membranes and could be operated successfully for several months in two different waterworks. The modified membranes combined exceptionally high NOM removal rates with the possibility for high divalent ion rejection and the stability of regular mechanical and chemical cleaning. Thus, LbL modified hollow fiber UF membranes did show a great potential for treatment of waters with high contents of dissolved organics and particle loads.Publicationtype: Doctoral Thesis124 74 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Automated monitoring of organic and bacterial drinking water constitution via PARAFAC fluorescence spectroscopy and flow cytometry(2025); Ensuring high drinking water quality is of major public health concern, protecting consumers from harmful consequences. One relevant concern about drinking water quality is the content of dissolved organic matter (DOM) which can be related to source water quality and treatment needs. Additionally, the bacterial constitution is crucial to monitor evolving changes in time and respond appropriately. However, DOM characterizing methods are a compromise between analytical effort, time and data quality while bacterial methods such as cultivation are labor-intensive, time-consuming and difficult to adapt to online measurements. Therefore, developing an automated monitoring system that quickly and consistently analyzes water quality data with high detail is vital. This work examines the application of parallel factor analysis (PARAFAC) fluorescence spectroscopy to characterize DOM content and flow cytometry to monitor the presence of bacteria in drinking water. The specific research objectives of this work were mainly to further develop, combine and automate, and test both methods in real-world applications. Using PARAFAC fluorescence spectroscopy, DOM compositions of investigated drinking waters, originating from groundwater in Northern Germany, were described through a respective fluorescence score of six compounds (C1–C6). Additionally, the total cell count (TCC) and the proportion of high nucleic acid cells (%HNA) were determined via flow cytometry describing the bacterial constitution. Scenarios of water quality changes were simulated in the laboratory by spiking one specific drinking water with increasing amounts of water samples of varying water quality. Water quality could be described by generating parameter-specific baselines defining their thresholds. Strongly deviating waters, e.g., wastewater effluent and rainwater, were identified in lower volume proportions than less strongly deviating waters. Regarding sensitivity, C1–C3 and TCC were the most performant for detecting water quality changes, e.g., due to a contamination event. Automation of both methods required hardware and software extensions and development. Sampling, data analysis, evaluation, and visualization were the automation objects addressed in this work. In the context of near real-time drinking water analysis, the time required from sampling to data visualization was reduced to less than 15 minutes. In pilot plant trials utilizing a model drinking water distribution system, combined methods were tested to detect water quality changes in a flowing system. Tests were performed to simulate events of water quality change. The combined system provided characteristic fingerprints of flowing water and detected sudden changes. 300 m behind the injection point, continuous induction of wastewater effluent and rainwater could be detected under appropriate conditions. In a large-scale measurement campaign, a three-component PARAFAC model (c1–c3) was proven to characterize groundwater DOM individually and can be applied to estimate humic substances concentrations in groundwater. In conclusion, it can be stated that PARAFAC fluorescence spectroscopy in conjunction with flow cytometry represents a rapid and powerful system for the comprehensive characterization of DOM and the presence of bacteria in both drinking water and groundwater. The continuous and automatic monitoring of the emphasized parameters enables to recognize even minor deviations in water characteristics.Publicationtype: Doctoral Thesis104 86 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Untersuchung des CARIX™-Verfahrens zur Entfernung anthropogener Sulfat-Emissionen aus einem Trinkwasser mit erhöhter NOM-Konzentration(2024); Some water supply companies in Germany are observing increasing sulfate contamination in their raw water resources and are required to consider mitigation strategies due to the drinking water regulation limit of 250 mg/L sulfate. One method suitable for sulfate removal is the CO2-regenerated CARIX™ ion exchange process. This study investigated the performance, influencing factors, and limitations of the process at both pilot and laboratory scales using drinking water from Berlin-Friedrichshagen. Laboratory experiments were also conducted with drinking water from Hamburg. During the twelve-month pilot testing at the Friedrichshagen waterworks, the process achieved sulfate removal of about 60–100 mg/L (34–59%) from the drinking water at yields of 70–84% without wastewater recovery. Additionally, 54–68% of water hardness, 62–75% of acid capacity, and 44–59% of DOC were removed. A reduction in acid capacity from 3.5 mmol/L to 2.9 mmol/L decreased sulfate removal from 100 mg/L (59%) to 60 mg/L (34%). The results indicate that while the yield influences the effluent water quality, the influent water quality, especially acid capacity, plays a crucial role. In Berlin-Friedrichshagen, sulfate enters surface waters as sulfuric acid due to lignite mining in Lusatia. As the sulfate concentration increases, the acid capacity decreases. Under these conditions, the performance of the CARIX™ process significantly declines with increasing sulfate concentrations. A continuous accumulation of natural organic matter (NOM) on the anion exchanger was observed, leading to organic fouling of the material. After one year of pilot operation, approximately 45.6% of the anion exchanger's ion exchange capacity was no longer available for ion exchange processes. Laboratory experiments showed that Berlin's humic substances and the organic fractions detected via UV254 exhibited a pronounced affinity for the anion exchanger, with ion exchange being their primary adsorption mechanism. For the first time, the displacement of anions by organic matter on the anion exchanger in the CARIX™ process was demonstrated. Carbonic acid proved unsuitable as a regenerant to reverse organic fouling due to its low concentration and pH levels. The operational behavior of the ion exchangers in the pilot plant was successfully reproduced in a laboratory setup. Based on the results of ion removal in the pilot and laboratory systems, an empirical model was developed to estimate the CARIX™ effluent water quality using knowledge of acid capacity and carbonate hardness in the influent. Laboratory experiments and the empirical model can thus complement or replace time- and cost-intensive CARIX™ pilot testing. This work expands the understanding of the fundamental mechanisms of the CARIX™ process and provides guidance on its performance and design without relying on pilot trials.Publicationtype: Doctoral Thesis79 112 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Ozonation-biofiltration for the production of drinking water from organic-rich groundwaters(2023); Groundwater resources present a major reservoir for worldwide drinking water production that is used intensively due to its consistent quality and availability. Organic-rich groundwater resources with high concentration of natural organic matter (NOM) are not used to their full potential due to the negative impact of NOM on treatment processes and on water quality, particularly on water coloration. This work examines the ozonation-biofiltration process to enhance NOM removal in drinking water production from organic-rich groundwater. Objects of research regarding the ozonation-biofiltration of organic-rich groundwater were (i) the transformation of groundwater NOM, (ii) by-product formation and its minimization and (iii) applicability and integration potential into groundwater treatment. An experimental approach was chosen to address these objects of research. Experiments included batch ozonation, small-scale pilot ozonation-biofiltration and membrane ozonation using different natural, organic-rich groundwaters. Ozonation of organic-rich groundwater is characterized by rapid ozone (O3) consumption by NOM, normalized to the dissolved organic carbon (DOC) content of groundwater (approxi-mately 0.5 mgO3/mgDOC in the first 30 s of ozonation). This results in significant reduction of ultraviolet absorbance (UV254) and color (SAC436) but in low disintegration of NOM to smaller molecular size and therefore moderate formation of assimilable organic carbon (AOC). The simultaneous formation of bromate, a presumed human carcinogen, may exceed international drinking water limits under these conditions. This is due to high bromide contents of groundwaters and presumably enhanced by hydroxyl radicals formed by ozonation of phenolic NOM. Biofiltration of ozonated groundwater by granular activated carbon and phonolithe results in significant reduction of AOC and minor reduction of UV254, SAC436 and DOC. Overall SAC436 removal potential by small-scale pilot ozonation-biofiltration was 40–50 % at bromate concentrations << 10 µg L-1, for two different groundwaters. For other groundwaters used in batch ozonation experiments, SAC436 removal potential may be up to 70 %. Decolorization performance of ozonation-biofiltration may be enhanced by the implementation of bromate minimization strategies. Process-oriented strategies (multi-stage ozonation and membrane ozonation) and pH reduction showed the highest potential to improve the trade-off between decolorization and bromate formation. Addition of ammonia or hydrogen peroxide reduced bromate concentration only at high ozone doses or compromised decolorization. Ozonation-biofiltration was performed with different feed water qualities including raw groundwater, aerated groundwater and finished groundwater, indicating high potential for integration into groundwater treatment. Overall, it was shown that ozonation-biofiltration is applicable in different groundwater matrices, for different feed water qualities, at low ozone doses of < 0.5 mgO3/mgDOC. Its potential may be increased by the development of bromate minimization strategies particularly eligible for organic-rich groundwater treatment.Publicationtype: Doctoral Thesis151 228 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Untersuchungen von Layer-by-Layer Beschichtungen für polymere Ultrafiltrationsmembranen und Charakterisierung der Membraneigenschaften(2022); The Layer-by-Layer (LbL) technology offers the possibility to modify ultrafiltration membranes in a way that certain retentions for dissolved substances, such as sulfate, DOC or trace substances are possible. Depending on the polyelectrolyte coating of the membrane, not only the performance parameters but also certain membrane properties change. In order to understand these changes and to gain an understanding of the altered retention mechanisms in the modified membranes, polymeric membranes were investigated in this work with respect to their optimal coating, membrane properties and solute retention behaviour. After initial tests, two polyether sulfone ultrafiltration membranes were found to be suitable for LbL modification, a flat sheet membrane and a multicapillary membrane. When coated with the polyelectrolytes poly-(diallyldimethylammonium chloride) (PDADMAC) and poly (sodium 4-styrenesulfonate) (PSS), it became clear that the number of bilayers (DL) deposited and the additional ionic strength in the coating solution were the dominant factors during coating. Coating at 0.1 mol/L NaCl additional ionic strength and a layer number of 8 DL resulted in 70% sulfate retention for both membranes, with the flat sheet membrane having a slightly higher permeability of 15.6 L∙m-2∙h-1∙bar-1 compared to the capillary membrane. Additionally, a stable backwash for the multicapillary membrane could be achieved when coated under pressure (1 bar). After single layer coating, the so-called odd-even effect was observed, in which the coatings terminated with the polyanion exhibited higher retention and slightly lower permeability than the coatings terminated with the polycation, suggesting that the charge of the top polyelectrolyte has an influence on the separation behaviour of the LbL membranes. This effect could be confirmed by zeta potential measurements which showed that, depending on the terminating polyelectrolyte, a zig-zag effect occurred which corresponded to the odd-even effect observed for retention and permeability; the coatings terminated with PSS had a more negative zeta potential, causing a stronger repulsion of the sulfate anion and thus a higher retention. The separation limit of the membrane, the so-called molecular weight cut-off (MWCO) was found to be between 300 and 400 Da after coating with 8 DL, which corresponds to the MWCO of nanofiltration membranes. Despite this low MWCO, no retention comparable to NF membranes could be achieved for sulfate. Further studies on ion retention showed that size exclusion was initially the dominant factor, caused by the reduction in pore size due to the polyelectrolyte coating. When the PE coatings have reduced the pores to the point where some retention is possible, membrane charge is also an important retention mechanism, and a weakly negative surface charge seems to be sufficient to achieve increased retention for a divalent anion. From the results of this work, it can be concluded that increased solute retention is possible by LbL modification of flat and multicapillary membranes. The retention mechanisms here are similarly complex to those of NF membranes, which LbL membranes resemble not only in separation behavior but also in many membrane properties.Publicationtype: Doctoral Thesis258 527