dc.description.abstract | The application of membrane system in Turkey has started to become widespread. Especially, the desalination process using reverse osmosis membranes extends the field of usage every day from industry to human consumption. In addition, membrane bioreactors have started to find a place in the market due to the advantages they provide. For these reasons, R&D in membrane technologies in our country has gained great importance. This doctoral study is one of the steps taken to meet this R&D requirement. Firstly, autopsy results of 8040 sized spiral wound modules which conducted in Turkey have been examined. It is aimed to develop novel reverse osmosis membranes having high antibiofouling and chlorine resistance capability using the data obtained from autopsy results and using common problems reported in literature. Firstly, the support layer optimization was done on a pilot scale, in order to reach the target membranes and the active layer properties. Novel active layer production step was started after support layer optimization.It was seen that 40% of total pollutant elements on the membrane surfaces were composed of silica. Silica is a serious pollutant and, if necessary precautions are not taken, may cause the membrane to run out of service life. It is normal to see that silica is the most common pollutant, because usually membranes which ended service life are sent to autopsy in Turkey. This is because silica fouling is not removable by chemical cleaning, and will certainly cause the membrane to run out of service life. Other foulant elements following the silica were aluminum, iron and calcium with 14%, 12% and 8% percents respectively. Aluminum generally precipitates as aluminum silicates with silica. It is estimated that the presence of 3-valent cations such as Al3+ and Fe3+ supports silica precipitation. It is known that Al3+ and Fe3+ are used as coagulants especially in large applications. High amount of aluminium and iron fouling can be connected to silica. Calcium and iron scalings can be prevented by the use of suitable antiscalant. Silica scalings can be prevented by optimized process conditions and pH adjustment. In addition to all these common foulants, specific foulants have been observed in some autopsies. For example, C coded autopsy results revealed high levels of manganaese foulant. Therefore, inlet water characteristics should be examined particularly before reverse osmosis system design and appropriate pre-treatment should be selected.Six out of ten different fujiwara tests performed in autopsies were positive. In other words, it can be said that there is 60% oxidation of membranes coming for autopsy. The most common reason for the oxidation of the membrane surface is the free active chlorine ions present in the water. In addition, the presence of high content of iron and bromine may also lead to oxidation. However, if there is free chlorine in the water fed to the treatment system, may result oxidation generally. In order to prevent oxidation, free chlorine should be inhibited by SMBS dosing at the entrance of the treatment system. If oxidation is seen despite dosing of SMBS, it is likely that there is a problem with the dosing system or the ORP probe. If there is no fault in these systems, the content of bromine may be suspected. The most common problem is reported as biofouling in literature based on membrane autopsy. However, autopsies performed in Turkey did not show any problem with biofouling except for special cases. However, organic pollutants were found on the surface of some membranes which were applied FTIR analysis. FTIR analysis was applied to both fouled and cleaned membranes. Carbohydrates and proteins were found on membrane surfaces. Although there is pre-treatment before reverse osmosis membranes and organic pollution is not expected, soluble microbial materials and natural organic substances can accumulate on membrane surface after passing UF filter. Sometimes bacteria or unicellular organisms can cross the membrane filter through a pin hole defect that can be found on the membrane, and can accumulate on the reverse osmosis membrane.2 membrane modules showed module deformation caused by pressure drop. Foulants accumulate on membrane surface and form an extra barrier layer which lead to higher drive pressure to achieve desired flow. In general, operators in plant directly increase the pressure to increase water flow. But this is not always right solution. If effective chemical cleaning is not applied, foulant layer becomes more persistent due to applied pressure. These foulant layers lead to increase the resistance to cross flow. The increased resistance causes a pressure drop between the inlet and outlet of the pressure vessel. This pressure drop is absorbed by the membrane module and converted to an axial pressure force on the membrane module. If the pressure drop is high enough, it will break the membrane module and damage it. Membrane manufacturers provide the maximum amount of pressure drop between inlet and output in their catalog. It is necessary to operate the plant by paying attention to these values. When the results of the autopsy analysis performed in Turkey are evaluated, it is seen that 3 different ways of improvement are needed for a better reverse osmosis process. These are;•Good system design, detailed consideration of raw water quality and proper selection of pre-treatment•The technicians or engineers who operate the plant should know the process well, understand that membrane systems are living systems and right interventions should be applied at the right time.•Development of more advanced membrane, modules, process chemicals or process equipmentIn the context of this doctoral study, it has been decided to develop more advanced special membranes and innovative active layers by considering these problems in the reverse osmosis membrane processes in our country and the common problems in the literature.Polysulfone (PSf) membranes are used to support the active layer in thin film composite membranes. Although salt retention and performance parameters are mainly dependent on the active layer, the support membranes also have significant effects on flux, salt retention, morphology and strength against compression by pressure. In this study, it was aimed to produce membranes with sponge-like morphology and with pore size below 20 micrometer in pilot scale system.There are no studies on the production of pilot scale membranes even though there are many laboratory scale studies in the literature on the production of polysulfone support layer.This thesis provides a detailed overview of the pilot scale production of asymmetric membranes used as support layers in thin film composite membranes. The most important result in this study was the differences found between lab scale production and pilot scale production. The conditions in the lab scale are not always applicable to the pilot scale. In the studies, it was revealed that various parameters affecting the membrane structure were seen together on a pilot scale and had a complex effect. In this study, it has been revealed that increased temperature of the coagulation bath increases the humidity around the bath and this moisture affects the membrane structure. When the coagulation bath temperature rises from 15ºC to 30ºC, the relative humidity on the coagulation bath increases from 35% to 49%. So, more water molecules in air contact with the polymer film and affects the final structure of the membrane.Normally, it is expected slower phase inversion and more spong-like structure in cold coagulation baths in lab scale studies. However, in pilot scale studies, it was observed that as the temperature increased, sponge-like structure and pore diameters were increased too. Because thin gel layer formed on the surface due to water vapor until polymer solution enters into coagulation bath, and this gel layer slows down the phase inversion rate. In addition to the temperature of the coagulation bath, casting speed and the solution content also had an effect on the properties of the membrane. As the PVP concentration decreases, the viscosity of the solution decreases too and the phase inversion rate increases and so that smaller surface porous membranes are formed. As the casting speed slows down, the amount of solvent that evaporates from top of casted solution increases and resulting gel-like layer slows down the speed of the phase inversion, resulting in more sponge-like structures. In total, 11 membranes were produced and 9 of them were having regular porous structure. M11 coded membrane selected from membranes produced in this study is used as support layer in active layer development studies.Both literature and results of the membrane autopsies performed in Turkey were investigated and it was decided to carry out studies on the development of novel membranes with biofouling and chlorine resistance. In this context, copolymers containing zwitterionic segment have been added to the polyamide active layers. The copolymers containing zwitterionic segment have advantages such as high chlorine resistance and low organic tendency but have disability to be used as active layers alone. A new active layer was formed by integrating these two different polymeric network structures. In addition, polyamide matrix was formed by interfacial polymerization method, while zwitterionic segment containing copolymers formed by free radical polymerization. These two polymerization techniques were tried simultanously together for the first time in literature.When FTIR results, SEM images and EDS spectra are examined, it is understood that the active layers are formed successfully. The lowest contact angle was found in the EnTFC_4 membrane with 57.2º. In this membrane, 2.2% acrylic acid and 4.8% DMMSA were added to the polyamide active layer. In addition, the lowest surface zeta potential value was found in the EnTFC_4 membrane too. This is followed by membrane EnTFC_3. Addition of DMMSA contributed more to the negative surface formation rather than the addition of AA. However, the addition of these substances reduced both salt retention rates and water fluxes. It was understood that thickness of the active layer increased as the extra material was added to the PA layer when cross section SEM images were examined. It can be said that the increase in the thickness of the active layer increases the resistance against the passage of the water and thus decreases the water flux. In fact, it is expected to increase salt retention with increasing active layer thickness. however, the salt retention rate decreased in the opposite way. Salt may escape from gaps between polyamide and zwitterionic silane copolymers, or from defects that cannot be seen on the surface by SEM. When the flux change rates after chlorination were examined, the least change was observed in EnTFC_4 membrane with 22% and the most change was seen in the reference membrane with 64%. The lowest reduction in salt retention after chlorination was seen in EnTFC_6 with a ratio of 2,65%. There is no general correlation in salt retention rate reduction, but it can be said that the resistance to chlorination increased compared to the reference membrane (6,83% reduction).When confocal microscope images, biofouling flux reduction rates and SEM images after biofouling test are examined, it is seen that the membranes are given antibiofouling properties.In summary, interfacial polymerization and free radical polymerization were successfully applied simultanously for the first time. Antibiofouling properties of both DMMSA and AA have been added to membrane and chlorine resistance was increased. However, along with all these positive developments, decreases in salt retention and flux ratios were observed. This doctoral study opened the door to a new method. It is thought that this method can be improved with new materials and different conditions in the following studies. | en_US |