Abstract

Thin-film nanocomposite (TFN) membranes are gaining great interest in industrial wastewater treatment due to their superior performance, especially for higher water flux and pollutants’ rejection, as well as improved chemical, thermal and mechanical stabilities, compared to other types of membranes. The performance of TFN membranes can be enhanced using different types of nanofillers, among which ceramic nanomaterials, such as Al2O3, offer a promising potential for applications involving treatment of industrial wastewater such as textile effluents, and this is due to their stability, hydrophilicity, and relative low cost. The study of the TFN membrane properties and structure allows the optimization and improvement in the membrane performance. This study aims at the preparation and characterization of TFN membranes with Al2O3 nanoparticles for the treatment of textile industrial wastewater, more specifically for removing dyes and salt solutes as well as for separating dyes from salt solutes. This work prepared thin-film composite (TFC) membranes and TFN membranes on polyether sulfone (PES) support membranes prepared by phase inversion. A polyamide (PA) selective layer was then synthesized atop the PES support by interfacial polymerization using m-phenylenediamine (MPD) and trimesoyl chloride (TMC) as the PA precursors. TFN membranes incorporated different amounts of Al2O3 nanoparticles within the PA selective layer during its formation. The MPD:TMC ratio was varied in the preparation of TFC and TFN membranes and the Al2O3 nanoparticles amounts were also varied in the preparation of the TFN membranes. The prepared membranes were characterized using different techniques. Fourier Transfer Infrared (FT-IR) and X-ray Photoelectron Spectroscopy (XPS) were used to study the composition of TFC/TFN membranes and the effect of changing MPD:TMC ratio on the membrane structure and the degree of cross-linking within the PA layer. The porosity, pore structure and morphology of the membranes were investigated using Scanning Electron Microscopy (SEM) and nitrogen gas adsorption analyzed using the Brunner-Emmett-Teller (BET) theory. The hydrophilicity and surface roughness were analyzed by contact angle and Atomic Force Microscopy (AFM) measurements. The mechanical and thermal properties were assessed using Thermo Gravimetric Analysis (TGA) and tensile tests. The performance of TFC/TFN membranes were studied in terms of pure water flux, permeate flux, dye, and salt rejection rates, using single dye aqueous solutions, salt aqueous solutions, and dye/salt aqueous mixtures. The results showed that the PES support membranes exhibited drop-like macrovoids with smooth surface, confirmed by a low surface roughness of Ra 4.48 (±0.50) nm. The porosity of the support membrane indicated the predominance of mesopores of ca. 5 nm with fewer larger pores of more than 30 nm. The support membrane had a comparatively hydrophobic surface with a contact angle of 65.34o (±6.83). The effect of varying the Al2O3 nanoparticles content on the structure, properties, and performance of TFC/TFN membranes in separating dyes form salts, was investigated for an MPD:TMC ratio of 2 w/v% MPD and 0.4 w/v% TMC. Different loadings of Al2O3 nanoparticles were tested, and the results showed that the incorporation of Al2O3 changed the pore structure of the PES support membranes to wavy finger-like macrovoids. Additionally, TFN membranes exhibited larger porosity in the 3-10 nm range, higher hydrophilicity, 48.68 to 59.35o contact angles, smoother surfaces, 36.17 to 51.65 nm roughness, than the corresponding TFC membrane of ca. 5 nm mesopores, contact angle of 65.34o, and roughness of 61.32 nm. Furthermore, TFN membranes had thinner PA layers varying between 271.2 nm for the lowest Al2O3 loading (1x10-3 w/v%) to 145.1 nm for the highest Al2O3 loading (1x10-1 w/v%), as compared to 309.4 nm for the corresponding TFC membrane. Surface SEM images showed some Al2O3 aggregation in the high content TFN membrane (1x10-1 w/v%). Porosity analysis indicated larger mesopores in the range of 3-8 nm with higher number of pores of > 6 nm for TFN membranes as compared with the corresponding TFC membrane. TFN membrane performance in separating dye and salt solutes from a mixture exhibited improved performance for TFN membranes as compared to the TFC membrane. For brilliant green (BG) dye/NaCl mixtures, improvement in permeate flux values ranged between 4.26 and 12.56 L/m2h for the lowest and highest Al2O3 amounts, 1x10-3 and 1x10-1 w/v%, respectively, as compared to a flux of 3.34 L/m2h for the TFC membrane. The dye/salt separation performance, measured by the selectivity factor, demonstrated higher values ranging between S = 2.77 and 9.12 for TFN membranes, which corresponds to an increase of 18.37% and 289.74% compared to the TFC membrane. Other dyes, as bromothymol blue (BTB) and reactive red (RR) salt mixtures showed the same trend as BG/NaCl with lower values for the selectivity factor ranging between S = 2.53 and 7.17 for BTB/NaCl mixtures and between S = 2.07 and 6.01 for RR/NaCl mixtures. The incorporation of Al2O3 nanoparticles improved the thermal and mechanical properties for all TFN membranes. The best performing TFN membrane was found to have Al2O3 loading of 1x10-2 w/v% (M3 membrane). This membrane had relatively larger pore width of > 6 nm compared to other TFN membranes. The contact angle and surface roughness of M3 were 45.04o (±5.10) and 38.74 (±6.51) nm with a PA layer thickness of 160.4 nm. M3 hydrophilicity, larger pores, smooth surface, and thinner PA layer improved the pure water flux by ca. +150% relative to the TFC membrane. M3 exhibited the best performance for dye/salt separation with a high dye/salt selectivity factor (S) across different dye/salt mixtures, namely 9.12 (± 1.32), 7.17 (± 0.26), and 6.01 (± 0.22) for BG/NaCl, BTB/NaCl, and RR/NaCl, respectively. In determining the impact of cross-linking on membrane performance, a series of TFN membranes with a fixed amount of Al2O3 nanoparticles (1x10-2 w/v% Al2O3), and varying MPD:TMC ratios, 2:0.4 w/v% (M3 membrane), 2:0.2 w/v% (M5 membrane), and 2:0.1 w/v% (M8 membrane) were prepared. The results showed that increasing MPD:TMC ratio to 2:0.1 w/v% resulted in nearly doubling the degree of cross-linking to 0.73, from 0.39 for the MPD:TMC ratio of 2:0.4 w/v%. This was accompanied by an increase in N/O ratio to 0.83 from 0.66. The increase in the cross-linking degree resulted in less developed wavy finger-like pores in the PES support. The effect of cross-linking was more pronounced on the PA layer surface showing larger ridge and valleys features, more nanoparticle agglomeration, and higher surface roughness, increasing to 44.05 (±4.29) nm (M8 membrane), from 38.74 (±6.51) nm (M3 membrane). The higher cross-linking affected the PA layer thickness which decreased from 160.40 nm for M3 to 76.39 nm for M8, as well as the membrane porosity that showed smaller mesopores in the range of 10-20 nm for the M8 membrane with a larger number of pores at ca. 6 nm. M5 membrane with intermediate cross-linking showed a complex pore structure in the range of 3-8 nm, and no pores in the range of 10-20 nm. For membrane performance, the higher surface roughness, smaller pores, and decreased PA thickness, all associated with more cross-linking, led to a reduction in the permeate flux from 11.13 L/m2h for the M3 membrane to 3.93 L/m2h for the M8 membrane. The NaCl rejection increased from 10.96% for M3 to 70.02% for M8, which corresponds to +538.87% improvement. All TFN membranes exhibited high dye rejection rates using BG, as a representative dye in NaCl aqueous solution, as this showed the highest selectivity factor using the lowest MPD:TMC ratio of 2:0.4 w/v%. The selectivity factor followed the order of M3 (S=9.12) > M5 (S=3.15) > M8 (S=1.43), indicating that the M8 membrane exhibited high rejection rates for both the dye and the salt solutes. The effect of changing the content of Al2O3 nanoparticles using the higher MPD:TMC ratio of 2 w/v% MPD and 0.1 w/v% TMC, on the structure, properties, and membrane performance for dye and salt removal, was investigated. Different loadings of Al2O3 nanoparticles were tested, and the results showed that Al2O3 changed the pore structure of the support membranes to wavy finger-like macrovoids that are less developed compared to TFN membranes with lower MPD:TMC ratio (2:0.4 w/v%). In addition, TFN membranes exhibited larger porosity in the range of 3-10 nm, higher hydrophilicity with contact angle values ranging between 42.27 and 56.40o, smoother surfaces ranging between 37.91 and 68.22 nm roughness, than the corresponding TFC membrane of ca. 5 nm mesopores, a contact angle of 62.77o, and roughness of 82.23 nm. Moreover, TFN membranes had thinner PA layers varying between 120.3 nm for the lowest Al2O3 loading (1x10-3 w/v%) and 64.93 nm for the highest Al2O3 loading (1x10-1 w/v%), as compared to 152.8 nm for the corresponding TFC membrane. Surface SEM images showed some Al2O3 aggregation in the higher content TFN membrane (1x10-1 and 1x10-2 w/v%). Porosity analysis indicated larger mesopores in the range of 3-8 nm with higher number of pores of > 6 nm for TFN membranes as compared to the corresponding TFC membrane. M7 membrane, with the lowest Al2O3 loading (1x10-3 w/v%), showed a complex pore structure in the range of 3-10 nm, and no pores in the range of 10-20 nm. The performance of TFN membrane in removing both dye and salt solutes from a mixture of BG, used as a representative dye in NaCl aqueous solution, showed improved performance for TFN membranes as compared to the TFC membrane. BG was selected as a representative dye as it exhibited high selectivity factor using MPD:TMC ratio of 2:0.4 w/v%. The permeate flux increased to 2.05 and 5.90 L/m2h for the lowest and highest Al2O3 amounts, 1x10-3 and 1x10-1 w/v%, respectively, compared to 1.52 L/m2h flux of TFC membrane. The efficiency of dye/salt removal, indicated by the selectivity factor, demonstrated low values ranging between S = 1.21 to 2.31 for TFN membranes, relative to S = 1.30 for TFC membrane. The best performing TFN membrane for the removal of the dye and salt solutes was found to have an Al2O3 loading of 1x10-2 w/v% (M8). This membrane had comparable porosity relative to other TFN membranes. The contact angle and surface roughness of M8 were 42.27o (±6.62) and 44.05 (±4.29) nm with a PA layer thickness of 76.39 nm. M8 hydrophilicity, larger pores, smooth surface, and thinner PA layer improved the pure water flux by ca. +50% relative to the TFC membrane. M8 membrane exhibited the best performance for efficient dye/salt removal with high dye rejection (99.96% for BG dye), and high NaCl salt rejection (70.02% from the BG/NaCl mixture), resulting in dye/salt selectivity factor (S) of 1.43 for BG/NaCl mixture.

School

School of Sciences and Engineering

Department

Chemistry Department

Degree Name

PhD in Applied Sciences

Graduation Date

Winter 1-31-2024

Submission Date

1-22-2024

First Advisor

Adham Ramadan

Committee Member 1

Amal Esawi

Committee Member 2

Ahmed El Gendy

Committee Member 3

Magdi Abadir

Extent

175 p.

Document Type

Doctoral Dissertation

Institutional Review Board (IRB) Approval

Not necessary for this item

Download

Available for download on Wednesday, January 21, 2026

Share

COinS