THE INFLUENCE OF PEG400 AND ACETONE ADDITIVES ON SELECTIVITY AND PERMEABILITY IMPROVEMENT OF POLYSULFONE BASED ULTRAFILTRATION MEMBRANE FOR HUMIC REMOVAL

• Main Author: Teta Prihartini Aryanti, Putu
• Format: Dissertations
• Language: Indonesia
• Subjects: Teknik kimia
• Online Access: https://digilib.itb.ac.id/gdl/view/33463
• Institution: Institut Teknologi Bandung

ABSTRACT Polysulfone ultrafiltration (UF) membranes have gained an important role in water treatment industry due to their excellent capabilities in removing particles, colloids, natural organic compounds, and microorganisms at a low operating pressure. In water treatment field, UF membrane performance is limited by fouling of natural organic matter (NOM), which contributed to the deterioration of membrane performances and flux loss. As a major of the fraction of NOM, humic substances are considered as the most foulant. Recently, the humic substances removal in water gains increased attention due to their reactivity with water clarifying antiseptics (such as chlorine) to form carcinogenic substances. Due to its small molecular size, the UF membranes may not be effective for the humic substances removal. Therefore, modification of UF membrane became inevitable to be conducted to obtain a tight membrane skin structure to improve its selectivity and water flux. In this research, the polysulfone (PSf) based UF membrane was modified by adding polyethylene glycol (PEG400) and acetone as additives. PEG 400 is a hydrophilic polymer which has a strong affinity with water. Consequently, high concentration of entrapped PEG400 in the membrane matrix increases the rate of water through the membrane. Furthermore, PEG400 also acts as a pore-forming agent to enhance membrane porosity, which contribute to the increase of water flux of the UF membrane. On the other hand, rapid loss of acetone leads to higher polymer concentration in the membrane skin layer and a tight skin during viii membrane structure formation. The formation of a tight membrane skin structure improves the UF membrane selectivity towards humic substance. The PSf UF membrane was prepared by phase inversion method, which involves a phase change of initially homogenous membrane solution into solid membrane matrix. Before solidified, the homogeneous membrane solution (composed of PSF, PEG400, acetone, and DMAc) will be separated into two phases, namely rich polymer phase, which give rise to the membrane matrix, and lean polymer phase, which form the membrane pores. In this research, phase separation of the membrane solution occurs when the membrane is immersed in a non-solvent bath (water). A concentration of water diffuses into the membrane, and then the phase separation is occured. Membrane formation by phase inversion method is strongly influenced by the composition of membrane solution, covering PSf, DMAc, PEG400, and acetone. Therefore, effects of each concentration of components on the membrane characteristic were investigated comprehensively, for both morphology and membrane performances during peat water filtration. The initial step of the experiment was the selection of the PSf concentration for preparation of ultrafiltration membranes before investigating the effect of additive concentration on the UF membrane characteristics, covering flux and rejection of humic substances. At this step, PSf was dissolved in DMAC at various concentrations, ie from 14% to 24% by weight. Then, PEG400 and acetone were added with a fixed concentration, ie 25% and 4% by weight. The next step was investigating the influence of each additive on the membrane performance at a fixed PSf concentration. PEG400 concentration was varied from 0 to 35% by weight, whereas the concentration of acetone was varied from 0-10% by weight. The resulting membrane performance was tested against fouling during two-hour of peat water filtration. Fouling on the membrane was tested by measuring water flux during peat water filtration, flux recovery ratio (FRR) after membrane cleaning, and fouling resistance formed on the fouled membrane. The expected result is an UF membrane that has water flux of above 100 Lm-2h-1 and humic substances rejection above 80%. The final stage of this research was studying ix thermodynamic properties of membrane solution through an equilibrium curve in ternary diagram. Experimental results showed that UF membranes prepared at a 14% wt. concentration of PSf had an open pore structure in the membrane skin layer, which resulted in low rejection of humic substance, ie by 72%. In addition, UF membranes that prepared from 14% wt. of PSf had low mechanical strength, which was damaged when operated at 30 psig. The resulted membrane structure became tight (dense) with the increase of PSf concentration during UF membrane preparation. Tight membrane structure contributed to higher humic substances rejection but lower in membrane flux. Furthermore, tight structure of the membrane enhanced intrinsic membrane resistance thus no water passed through membrane pore when operated at 10 psig. The resulted UF membrane can be operated at 10 up to 30 psig when 18 – 20% of PSf was blended into membrane solution. At 18% wt. of PSf, the resulted membrane showed a high flux reduction during two hours of peat water filtration. It is suspected that irreversible fouling of small particles has occurred due to open pore formation in the membrane skin layer. Above 100 Lm-2h-1 of water flux and above 80% of humic substance rejection were found when blending 20% of PSf to membrane solution. Additionally, a stable flux was resulted when the said membrane was operated at various different transmembrane pressures. The influence of PEG400 concentration of membrane characteristics and performances were investigated at the next research step of this research, which PEG400 concentration was varied from 0% wt. to 35% wt. While concentration of PSf was fixed at 20% wt. with the absence of acetone. The experimental data showed that the formation of larger pores with the addition of high concentration PEG400 improved membrane hydrophilicity, which was indicated by decreasing water contact angle on UF membrane surface. However, higher PEG400 concentration led to formation of larger pores in the membrane surface structure, which contributed to irreversible fouling and reduced the flux recovery ratio x (FRR) of the membrane. Furthermore, larger pores formed in the membrane skin layer resulted in a lower rejection of humic substances. Fouling and low rejection problem can be minimized by adding a small amount of acetone into polysulfone membrane solution. The experimental results indicated that high rejection of humic substances was achieved due to the formation of tight skin membrane structure. However, that structure gave lower flux as consequences. Over 80% of humic substance rejection and 126 Lm-2h-1 of water flux was achieved by blending 4%wt of acetone in 20%wt of PSf and 25%wt of PEG400. Based on solute rejection (dextran) characterization, the approximate pore size of the obtained UF membrane was 20 kDa. Tight structure of membrane surface resulted in higher FRR (98%) compared with unmodified membrane (35%) after peat water filtration and flushing method. Most of fouling formed on the membrane surface during the peat water filtration are reversible and can be swept easily by cross-flow in membrane module during cleaning process with flushing method. It showed that acetone modified membrane had a lower tendency to fouling by organic compounds resulting stable membrane flux. The last step of this research was a thermodynamic study of the UF membrane system consisting Air / DMAC / PSF / PEG400 to reveal the influence of the membrane composition on the membrane solution solubility. The resulting analysis can be used as a basis for further research to predict suitable membrane composition and produce desired membrane structure. The thermodynamic model Flory-Huggins (F-H) was used to predict equilibrium condition of membrane solution, which was influenced by interaction of membrane solution components. In this research, interaction parameters of membrane solution (water/DMAc/PSf/PEG400) were calculated based on equilibrium composition obtained from cloud point experiment. The cloud point test was conducted at various membrane solution, which prepared by blending 20%wt of PSf with various PEG400 concentrations (0-20%wt). The computing calculation showed that interaction parameters for each component in the membrane solution are: X13 (water-PSf) = 6.76, g14 (water-PEG400) = 0.51, g23 (DMAc-PSf) = 0.46, g24 xi (DMAc-PEG400) = 0.46, and g34 (PSf-PEG400) = -0.53. These parameters were tested at higher concentration of PEG400 (25-35%wt), and followed on different PSf concentration (12%wt). Experimental results showed that parameters used in thermodynamic model resulted in a binodal curve, which matched to the cloud point experimental result. The thermodynamic model with the defined parameters then used to predict the thermodynamic behaviour of the UF membrane solution at certain concentrations. Based on the thermodynamic analysis, UF membrane made from 20%wt of polysulfone with high concentration of PEG400 (20-35%) has a low solubility, which can be easily separated by the presence of a low water concentration, thus the formation of membrane structure become faster. Based on analysis of membrane characteristics, which supported by thermodynamic analysis of the membrane solution, good membrane performances were achieved when the UF membrane was prepared by blending 20% wt. of PSf, 25% wt. of PEG400, and 4% wt. of acetone into DMAc. Above 100 Lm-2h-1 of water flux and above 80% of humic substance rejection was achieved. Keywords: polysulfone membrane, interaction parameter, liquid-liquid demixing, polymer additive, humic compound.