Green Synthesized Composite AB-Polybenzimidazole/TiO₂ Membranes with Photocatalytic and Antibacterial Activity

Abstract

Novel AB-Polybenzimidazole (AB-PBI)/TiO₂ nanocomposite membranes have been prepared using a synthetic green chemistry approach. Modified Eaton’s reagent (methansulfonic acid/P₂O₅) was used as both reaction media for microwave-assisted synthesis of AB-PBI and as an efficient dispersant of partially agglomerated titanium dioxide powders. Composite membranes of 80 µm thickness have been prepared by a film casting approach involving subsequent anti-solvent inversion in order to obtain porous composite membranes possessing high sorption capacity. The maximal TiO₂ filler content achieved was 20 wt.% TiO₂ nanoparticles (NPs). Titania particles were green synthesized (using a different content of Mentha Spicata (MS) aqueous extract) by hydrothermal activation (150 °C), followed by thermal treatment at 400 °C. The various methods such as powder X-ray diffraction and Thermogravimetric analyses, X-ray photoelectron spectroscopy, Fourier transform infrared spectroscopy and Energy-dispersive X-ray spectroscopy, Electronic paramagnetic resonance, Scanning Electron Microscopy and Transmission Electron Microscopy have been used to study the phase and surface composition, structure, morphology, and thermal behavior of the synthesized nanocomposite membranes. The photocatalytic ability of the so-prepared AB-Polybenzimidazole/bio-TiO₂ membranes was studied for decolorization of Reactive Black 5 (RB5) as a model azo dye pollutant under UV light illumination. The polymer membrane in basic form, containing TiO₂ particles, was obtained with a 40 mL quantity of the MS extract, exhibiting the highest decolorization rate (96%) after 180 min of UV irradiation. The so-prepared AB-Polybenzimidazole/TiO₂ samples have a powerful antibacterial effect on E. coli when irradiated by UV light.

1. Introduction

Combining polymer materials with selected nano-scale size oxide particles enables the possibility of preparing materials with a wide spectrum of applications [1]. In the last decades, scientific efforts have been directed at seeking new types of polymer hybrid materials that have, at the same time, enhanced photocatalytic activity for the decolorization of organic pollutants in wastewater while manifesting efficient antimicrobial properties. By controlling their composition, the size of the nano-phase, and the chemical bonding between the organic and inorganic phases, one could obtain hybrid nanocomposites possessing the desired thermal, electric, optical, mechanical, and biological properties. Such types of composites have been studied intensively in lithium-ion batteries [2], in solar and fuel cells [3,4], and in photocatalysis [5]. In medicine, nanoparticles are used as antibacterial agents in drug delivery systems and in clinical diagnosis. The composites based on polymer hybrid materials with incorporated metal nanoparticles also have strong antibacterial properties, which could be applied in the production of packages for foodstuffs and in the treatment of textile materials.

A large number of metal oxides are used as photocatalytic/antimicrobial agents, such as zinc oxide (ZnO), titanium dioxide (TiO₂), CeO₂, etc. [6]. Among them, titanium dioxide is a non-toxic, insoluble, and stable metal oxide that is extensively used in various industries due to its excellent optical, photocatalytic, antimicrobial, and anticorrosive properties.

More and more publications are appearing in the current scientific research literature connected with the application of methods of green chemistry/synthesis, leading to obtaining oxide nanoparticles that have promising photocatalytic and antimicrobial properties [7,8,9]. The use of plant extracts is economically beneficial (one-step preparation) because they are cheap, bioresource-based, and environmentally friendly. Biosynthesis guarantees the production of materials with improved properties compared to traditional methods. Synthesis involving plant extracts can provide preparation of stable nanoparticles of controlled size and morphology and, last but not least, reduced cytotoxicity of nanoparticles thanks to their encapsulation in a biologically inert medium [10]. Recently, the plant-mediated synthesis was successfully applied to TiO₂ particle production [11]. S.S. Muniandy et al. have synthesized an active mesoporous powder of TiO₂ using a green chemistry approach that has a high photocatalytic efficiency for dye decolorization and good catalytic activity even after repeated use [12]. Other research groups have hydrothermally synthesized TiO₂ using Morindacitrifolia leaf extract, which shows promising activity against fungal pathogens and bacteria, both Gram-positive and Gram-negative strains [13].

The advancement of heterogeneous photocatalysis in various fields, such as decolorization of Methylene blue dye using Ni-doped ZnO nanostructures [14], decontamination of Cr(VI) and norfloxacin antibiotic using Au/MIL-101(Fe)/BiOBr photocatalyst [15], etc. has been discussed in the literature.

During the last few years, studies on the photocatalytic properties of hybrid nanocomposites have become the hot topic of the day on the basis of polymer/inorganic oxide composite materials in different forms for the purification of model pollutant dye-contaminated waters. The electron- and ion-conducting polymers, which most often have aromatic conjugated structures and metal oxides, possess good electron donor properties under the effect of light. The photocatalysts make use of the light energy, whereupon electron–hole pairs are formed, generating free radicals, which are involved in the reactions for demineralization of stable, persevering toxic compounds. In order to overcome the limitations, due to the fast recombination of the electron–hole pairs, the conjugated polymers are combined with semiconductor oxide nanoparticles of dimensions of about 2–10 nanometers. In such mixtures, the organic polymer acts as a donor of electrons, while the inorganic nanoparticles are electron acceptors [16].

TiO₂–polymer composites in membrane form are among the most often used in water purification and wastewater treatment [17]. The various polymers, such as polyacrylonitrile (PAN) [18], poly(vinylidene fluoride) (PVDF)/sulfonated polyethersulfone (SPES) [19], polyamide [20], polyvinylidenefluoride (PVDF) [21], polyethersulfone (PES) [22], polyurethane (PU) [23], polyethyleneterephthalate (PET) [24], polyester [25], poly(vinylidene fluoride) [26], and polytetrafluoroethylene (PTFE) [27], have been investigated as support membranes of photocatalysts.

Among them, the aromatic polybenzimidazoles (PBIs) group of heterocyclic high-performance polymers has been studied as an attractive candidate for the preparation of hollow fiber membranes used in liquid and gas separations. Some of these membranes have achieved industrial requirements under extremely harsh process environments (i.e., high temperatures, chlorine, organic solvents, and extreme pH values) due to their robust mechanical stability, structural rigidity, and outstanding chemical resistance [28]. Justo Lobato’s research group prepared several TiO₂/meta-PBI-based composite membranes for application in high-temperature, solid-state polymer electrolyte membrane fuel cells [29,30] and for green hydrogen production [31,32,33]. In a short communication, Penchev et al. revealed, for the first time, the preparation of some novel meta- and AB-PBI-based hybrid materials with in situ synthesized ZnO NPS in the form of powders and thin films and studied their photocatalytic activity towards Malachite green as a model dye contaminant [34].

The fight against infectious diseases is complicated due to the natural ability of microorganisms to adapt themselves and bypass the antimicrobial activity of a wide range of preparations. The application of antimicrobial polymers is a novel approach to overcome these limitations [35,36,37,38]. It has several advantages, as these products typically exhibit long-term activity and limited residual toxicity, they are chemically stable, non-volatile, and they do not penetrate the skin [39]. The antimicrobial features of TiO₂ particles have been studied and described in a large number of experimental and review studies [40,41,42,43,44]. It has been established that the formation of reactive oxygen species (ROS), similar to other metal oxides, leads to damage and destruction of microbial cells [45,46]. The data on the ability of titanium dioxide NPs to produce ROS in the dark and under irradiation are contradictory. Some publications report the presence of ROS only under UV light exposure, while others detect ROS both in the dark and upon irradiation. It is possible that these differences are due to the different methods used to prepare the nanoparticles, their different sizes, surface defects, and the conditions of the experiments [46,47]. The use of herbs, which have enhanced antibacterial and antioxidant properties, also has advantages, resulting in an improvement in the antimicrobial activity and a decrease in the cytotoxicity of the obtained TiO₂ particles.

According to our knowledge, the preparation of AB-PBI/TiO₂ membranes and the study of their photocatalytic/antimicrobial properties is still missing in the literature. The aim of this study is to evaluate the UV-initiated photocatalytic activity for Reactive Black 5 (RB5) azo dye decolorization, as well as the antibacterial activity towards E. coli (under dark and UV conditions), of the novel composite AB-PBI membranes containing TiO₂ particles, produced with different contents of Mentha Spicata plant extract.

2. Materials and Methods

2.1. Materials

3,4 diaminobenzoic acid (97%, Merck) (DABA), methanesulfonic acid (98%, Merck), and phosphorous (V) oxide (98%, Alfa Aesar) were used as received. Titanium (IV) isopropoxide (Alfa Aesar, Haverhill, MA, USA); acetic acid (Valerus, Sofia, Bulgaria); acetyl acetone (Sigma Aldrich, Saint Louis, MO, USA ReagentPlus, ≥99%); isopropanol (Sigma Aldrich, Saint Louis, MO, USA); Mentha Spicata leaves (Bulgarian ecological region); Reactive Black 5 (Sigma Aldrich, Saint Louis, MO, USA), dye content ≥50%; hydrochloric acid 37%, p.a. analytical reagent (Valerus, Sofia, Bulgaria).

2.2. Bio-Mediated TiO₂ Particles

Mentha Spicata (MS) is a plant that belongs to the family Lamiaceae. It is widely used in medicine, pharmacy, etc., due to its excellent antimicrobial and antioxidant properties. A Mentha Spicata-mediated green approach, combined with hydrothermal treatment, was used to obtain titanium dioxide nanoparticles. The MS leaves were washed and dried, and afterward, they were ground into powder and mixed with double-distilled water. In order to obtain the final extract, the solution was heated up, then it was centrifuged and filtered. The titanium precursor solution was prepared using Ti isopropoxide, acetic acid, and acetyl acetone (complexing agent) dissolved in isopropanol. The evaluated quantities of the as-prepared MS extract were added to the Ti solution. The resulting mixture was transferred into an autoclave at 150 °C. The as-obtained precipitates were washed and then dried in an oven, and thereafter, they were treated thermally at 400 °C for 3 h. Powders prepared from solutions having 40 mL MS, 70 mL MS, and 100 mL MS extracts are denoted as M0, M1, and M2, respectively.

2.3. Preparation of AB-PBI/Bio-TiO₂ Membranes

The incorporation of TiO₂ M0, M1, and M2 particle powders into the polybenzimidazole matrix was performed by in situ polycondensation of DABA monomer in the presence of anatase particles of 20 wt.% content, according to the theoretical total polymer concentration after polymerization, resulting in AB-PBI/TiO₂ composite stock solutions used further for the membrane preparation. In a typical procedure in a 50 cm³ three-neck flask supplied with a magnetic stirrer, 2.1 g of DABA was dissolved by vigorous stirring in 30 g Eaton’s reagent for 2 h at 70 °C under an argon atmosphere until full dissolution of the monomer. Next, 0.3 g of TiO₂ M0, M1, or TiO₂ M2 powders were mixed in the pre-polymerization solution of DABA with a short (5 min) pulsed ultrasonic treatment for best dispersion. In the convenient heating experiments, the TiO₂ M1 and M2 particles dispersion stock solutions were gradually heated up in a silicone oil bath and left for 48 h at 140 °C till viscous polymer composite dispersion was obtained, ready for film casting. In the microwave-assisted approach, the same composite pre-polymerized stock dispersions were heated in a ROTO SYNTH Rotative Solid Phase microwave reactor from Milestone, Bergamo, Italy, for 1 h at 100 Watts irradiation power, resulting in stable viscous AB-PBI/TiO₂ dispersions as well. Composite AB-PBI/TiO₂ membranes were prepared using the inverse precipitation technique. The heated stock dispersions of TiO₂ M0, M1, and TiO₂ M2 were first film cast using a doctor’s blade with a 200 µm knife gap onto a glass substrate. Immediately after the film casting, the glass substrate was poured into cold water, acting as a PBI precipitation bath. After several minutes of staying in the water bath, the inverse composite membranes self-peeled off from the glass and were poured for 12 h into a 2 L glass beaker filled with 5 wt.% ammonia solution under stirring in order to neutralize the remaining acids ionicly bonded in the AB-PBI matrix. The so neutralized AB-PBI/TiO₂ 20 wt.% composite gel membranes were finally washed overnight in deionized water in order to remove the remaining ammonia and its acid salts. The dry content of the composite membranes was determined to be 6.4 wt.% gravimetrically after vacuum drying a certain amount of the surface filter paper-soaked inverse membrane, and this value is close to the theoretically estimated value.

2.4. Physicochemical Characterization of Prepared AB-PBI/Bio-TiO₂ Membranes of Bio-Synthesized TiO₂ Particles

The phase composition and particle morphology of the samples were monitored using Transmission Electron Microscopy (TEM) investigations by means of a JEOL 2100 transmission electron microscope (JEOL, Tokyo, Japan) and a JEOL 2100 XEDS: Oxford Instruments, X-MAXN 80 T CCD Camera ORIUS 1000, 11 Mp, GATAN. The X-ray photoelectron spectroscopy (XPS) was carried out using an AXIS Supra electron-spectrometer (Kratos Analitycal Ltd., Manchester, England, AlKα radiation; photon energy of 1486.6 eV). The areas and binding energies (accuracy of ±0.1 eV) of C1s, O1s, and Ti2p photoelectron peaks were monitored. UV-vis diffuse-reflectance spectroscopy (DRS) was applied to determine the band gap energy, e.g., of TiO₂ green synthesis. The diffuse-reflectance spectra (DRS) were recorded on a Thermo Evolution 300 UV-VIS spectrometer (Thermo Fisher Scientific, Waltham, Massachusetts, USA) within the wavelength range of 190 nm up to 1100 nm. Electronic paramagnetic resonance spectra (EPR) were recorded at room temperature by a JEOL JES-FA 100 EPR spectrometer (JEOL Ltd., Tokyo, Japan) operating within the X-band using a standard TE011 cylindrical resonator at a modulation frequency of 100 kHz and microwave power of 3 mW.

2.5. Physicochemical Characterization of Prepared AB-PBI/Bio-TiO₂ Membranes

The powder X-ray diffraction (PXRD) analysis was performed on an X-ray powder diffractometer “Empyrean” (Malvern Panalytical, Almelo, The Netherlands) within the range of 2θ values between 3° and 100° using Cu Kα radiation (λ = 0.154060 nm) at 40 kV and 30 mA. The presence of phases was determined using High Score Plus, Version 4.9 (4.9.0.27512) software. The average crystallite size (D), lattice strain (ε), and unit cell parameter (a) of the titanium dioxide phase were determined using the Powder Cell software [48] and using the Williamson–Hall equation [49]:

β cosθ = 0.9 λ/D + 4εsinθ

(1)

where ε is the value of internal strain, β is the full-width half-maximum (FWHM) of diffraction, θ is the Bragg’s angle, λ is the wavelength of the X-ray beam used, and D is the mean crystallite size of the phase under study.

The X-ray photoelectron spectroscopy (XPS) measurements were performed using a VG ESCALAB II electron spectrometer (Birches Industrial Estate, Sussex, UK equipment) with AlKα radiation at an energy of 1486.6 eV. The binding energies were determined with an accuracy of ±0.1 eV utilizing the C1s line at 285.0 eV (from an adventitious carbon) as a reference. The chemical compositions of the films were investigated on the basis of areas and binding energies of C1s, O1s N1s, P2p, Na1s, and Ti2p photoelectron peaks (after linear subtraction of the background) and Scofield’s photoionization cross-sections.

Electronic paramagnetic resonance spectra (EPR) were performed at room temperature using a JEOL JES-FA 100 EPR spectrometer operating within the X-band using a standard TE011 cylindrical resonator at a modulation frequency of 100 kHz and microwave power of 3 mW.

Fourier transform infrared (FTIR) spectroscopic analysis was carried out on an IR Affinity-1 spectrophotometer (Shimadzu, Kyoto, Japan) equipped with a MIRacle ATR accessory (diamond crystal, the depth of penetration of the IR beam into the material is 2 μm).

The surface morphology of the membranes was studied by applying scanning electron microscopy (SEM) using a Zeiss Evo 10 microscope (Carl Zeiss Microscopy, Oberkochen, Germany). The images were taken in secondary electrons and backscattered electrons mode at an accelerating voltage of 25 keV, and no conductive coating was present on the samples. The chemical composition of the surface was studied using the electron dispersive spectroscopy (EDS) probe Oxford Ultim Max 40 (Oxford Instruments, Abingdon, UK). The results were compiled with AZtec software (version 6.1 HF4).

In order to perform the scanning electron microscopy analysis, a bacterial suspension of the strain Escherichia coli ATCC 25922 was prepared in the same manner as that for the antimicrobial activity tests. After an incubation period, the tested membranes were washed threefold with PBS, then fixed with 4% glutaraldehyde in 0.1 M Na cacodylate buffer (pH 7.2) during the incubation period for 2 h at 4 °C. The post-fixation procedure was carried out for 1 h in a solution of 1% OsO₄ again at 4 °C. The last steps were dehydration in a graded ethanol series during 15 min time intervals and sputtered with gold using a vacuum evaporator (Edwards, Irvine, CA, USA). Observations were made on a Lyra/Tescan scanning electron microscope (TESCAN GROUP a.s., Brno, Czech Republic) with an accelerating voltage of 20 kV.

The thermogravimetric (TG) analyses were performed in an argon atmosphere on a TGA-4000 PerkinElmer instrument (PerkinElmer, Waltham, MA, USA) (temperature range 40–800 °C, 10 °C/min, nitrogen atmosphere).

2.6. Photocatalytic Activity Tests

The photocatalytic decolorization of Reactive Black 5 dye as a model contaminant at an initial concentration of aqueous solution of the dye—5 ppm—has been studied. The synthesized AB-PBI/bio-TiO₂ membranes functioning as photocatalysts were tested. The UV-A illumination lamp, which had maximum emission at 365 nm, power of 18 W, and illumination intensity of 2.6 mW/cm², was used. The UV irradiation dosage for 180 min of exposure was 3.921 × 10⁻¹⁷ J/m². The distance of illumination between the photocatalytic reactor and the lamp was 10 cm. The photocatalytic activity tests were performed at pH = 3, adjusted by 1M HCl. The photocatalytic experiments were carried out in a semi-batch slurry photocatalytic reactor equipped with two frits blowing tiny bubbles of air (continuous flow) in order to saturate the solution in dissolved oxygen using 3 × 3 cm AB-PBI/bio-TiO₂ membranes as a photocatalyst and 75 mL (batch) of dye solution under a constant stirring rate (400 rpm). In the tested systems, the dye solution and photocatalyst were stirred in the dark for about 30 min before switching on the UV irradiation for 3 h in order to reach the adsorption–desorption equilibrium state. The investigations of the photocatalytic abilities of synthesized hybrid membranes were carried out by taking aliquot samples of the suspension out of the reaction vessel at regular time intervals. The reaction course was monitored by a UV–vis absorbance spectrophotometer UV-1600PC (VWR International, Radnor, PA, USA) within the wavelength range from 200 to 800 nm (λmax = 599 nm for RB5). The degree of decolorization was determined using dependence ((C₀ − C)/C₀) × 100, where C₀ and C are, respectively, the initial concentration before turning on the illumination and the residual concentration of the dye solution after illumination in the course of a given time interval.

2.7. Antibacterial Properties

• Media dia and test microorganisms

For the preparation of standardized bacterial cell suspensions and the cell number counting, Mueler–Hinton Broth (MHB) and Mueler–Hinton Agar (MHA) (HiMedia Laboratories, Mumbai, India) were used. Samples of the obtained materials, including the three types of TiO₂ nanopowders (M0, M1, M2), AB-PBI membrane, and AB-PBI/M0, AB-PBI/M1, and AB-PBI/M2 composites were autoclaved at 110 °C for 15 min. Escherichia coli ATCC 25922 and Bacillus subtilis ATCC 6633 were used for Minimum inhibitory concentration (MIC) testing. The photocatalytic activity was tested on Escherichia coli ATCC 25922.

MICs were tested by the Resazurin-based Microtitre Dilution Assay (RMDA) as previously applied by Sarker et al. (2007) [50] under static and dynamic conditions (on a rotary shaker, 120 rpm). The initial concentrations of the starting suspensions were 5 and 20 mg/mL for each nanopowder in distilled water. They were vortexed and sonicated before use.

The antimicrobial activity of the composite membranes was tested according to the ASTM method E 2149–10 [51], adapted for a 24-well sterile plate.

• Photocatalytic Antibacterial Activity of Bare and PBI-Stabilized TiO₂ Nanoparticles

For the photocatalytic antibacterial test, a standardized cell suspension of E. coli containing approximately 1.5 ÷ 3.0 × 10⁶ cells in sterile saline was used. The applied procedures were described earlier [52].

All antibacterial tests were performed in three replicates.

3. Results

3.1. TiO₂ Particles

TEM micrographs show that the powders have mostly spherical morphology. The particle sizes of MO samples are within the range of 15–20 nm (Figure 1). It must be noted that the M1 powders exhibit a more homogeneous distribution than that of M0 and M2. Most of the particle sizes vary between 8 and 20 nm. For sample M2, it is characteristic that 50% of the spherical particles have a size of 15 nm, but some single smaller and bigger particles also occur.

The chemical compositions of the nanoparticles, evaluated by means of XPS analyses, are shown in

Table 1. Chemical composition of green synthesized TiO₂ particles, evaluated by XPS analyses.

	C, at.%	O, at.%	Ti, at.%	OT/OL
MO	22.1	54.1	23.8	1.49
M1	20.4	56.8	22.8	1.56
M2	21.7	55.8	22.5	1.85

Note: O_T—total oxygen; O_L—lattice oxygen.

. The ratio O_T/O_L increases slightly upon increasing the plant extract quantity in the precursor solution. This fact indicates an increased number of oxygen vacancies inside the TiO₂ lattice.

The optical properties of the green synthesized TiO₂ were investigated by DRS analyses (Figure 2). The absorption edge of the MO sample is 380 nm, while edges for M1 and M2 possess slightly lower values (Figure 3). A Kubelka–Munk transformation was used to obtain the band gap value (Figure 3). Figure 4 represents Tauc’s types of plots and the estimated band gap values. From the extrapolation of the linear section of the graph (hνF(R∞))^1/n from hν, the value of the optical bandgap is determined, which for pure TiO₂ is 2.98 eV, and with the addition of mint, regardless of its amount, it shifts to a higher eV value.

In Figure 5, the EPR spectra of green synthesized TiO₂ particles are shown. The signal S1 at g = 2.000 could be attributed to the presence of oxygen radicals [53,54]. The signals at g < 2 can be due to the capture of an electron on titanium ions. The broad EPR line (S3) at g = 1.93 indicates the formation of Ti³⁺ species on surface sites, while the narrow EPR signal (S2) at g﬩ = 1.991 and g// = 1.976 are probably assigned to inner electron traps (Ti³⁺) of the particle [55]. The signal S4 at g = 2.054 can be assigned to adsorbed water or absorbed oxygen O²⁻ [54,55]. Chloride ions can react with the hole to form Cl• atoms with EPR signals at g = 2.063 (S5) and 1.997, which can originate from the plant extract (proved by EDX analysis). The signal S6 at g = 2.07 can be due to OH• radicals adsorbed on the surface [53]. When the MS concentration increases, the intensities of these EPR signals, which are attributed to the presence of Ti³⁺ and oxygen radicals, increase. Probably, the ability of plant compound constituents to reduce the metal cations is responsible for the presence of Ti³⁺ ions. According to the research of Liu et al., the use of acetic acid in the hydrothermal process can destroy the Ti-O bond and finally form surface oxygen vacancies and Ti³⁺ [56].

The physicochemical features of the obtained plant-synthesized titania particles, nanocrystalline anatase structure, oxygen radicals, and Ti³⁺ ions decreased the band gap value, making them suitable for possible applications in the fields of catalysis or photocatalysis.

3.2. Preparation of AB-PBI/Bio-TiO₂ Membranes

The synthetic approach for the synthesis of Poly(2,5-Benzimidazole), commonly known as AB-PBI, includes performing non-equilibrium polycondensation of single monomer 3,4 diaminobenzoic acid in modified Eaton’s reagent (10 wt.% P₂O₅ dissolved in methanesulfonic acid), acting both as a dehydrating agent as well as polymer/TiO₂ particles and a solvent/dispersant, compared with the use of conventional and microwave-assisted heating at 140 °C. AB-PBI is the structurally simplest member of the PBIs group of high-performance, chemically stable, and nonflammable polymers, and it possesses the highest imidazole ring density as a repeating unit in its macromolecular backbone as well as the highest degree of crystallinity [57,58]. A very important feature of AB-PBI is the fact that it is green synthesized based on the use of a single monomer–3,4 diaminobenzoic acid, which is a cheap and low-toxicity alternative to the very expensive and cancerogenic diaminobenzidine co-monomer used in the commercial meta- and para-PBI synthetic approaches, such as the Marvel–Vogel bulk melt polycondensation [59] and that originally developed by Benicewicz et al. using a PPA sol-gel method for the para-PBI [60,61]. We had a particular challenge in our effort to incorporate well-dispersed TiO₂ particles into the AB-PBI matrix. Titania particles in different forms are known for their high-agglomeration tendency when obtained ex situ in a dry powder form [62,63]. In that aspect, we found the performance of in situ DABA polycondensation in Eaton’s reagent containing well-dispersed TiO₂ particles to be a particularly useful approach. The strong polarity and in situ dehydrating power of Eaton’s reagent helped to achieve good de-agglomeration of the pristine TiO₂ powders, as well as the step-growth polymerization of DABA and the formation of a shear viscosity AB-PBI solution acting as a surface stabilizing agent. A particular reason for the use of Eaton’s reagent as the polymerization medium is the solubility limitation of high-molecular AB-PBI, which is known to be poorly soluble in organic solvents traditionally used for other biphenyl-based PBIs such as dimethylacetamide/LiCl mixture [64]. We found the other reaction solvent alternative, e.g., polyphosphoric acid (PPA), particularly inappropriate for its inherent high shear viscosity and the requirement of high-power stirring. Also, the synthesis of AB-PBI in PPA solution, from our experience, results in solutions that are too viscous, which creates difficulties in film casting and obtaining membranes with poor mechanical properties.

The general synthetic procedure for AB-PBI/TiO₂ composite membrane preparation is represented in Figure 6. In recent years, the use of the microwave irradiation (MW) approach for the synthesis of high-performance polymers, including different PBIs, has gradually emerged [65,66]. It is known from the practice that the use of the MW technique permits a significant reduction of the reaction time interval, and it even results in higher molecular weight meta-PBI type polymers with the use of Eaton’s reagent as a solvent and reaction medium, compared with the conventional heating technique [67]. In the present study, we employed (based on our knowledge) for the first time one pot MW and conventional heat synthesis of composite AB-PBI/TiO₂ stock dispersions using three different plant extract-synthesized anatase particles of 20 wt.% loading, as well as blank experiments based on the synthesis of pure AB-PBI with the use of Eaton’s reagent. A significant reduction of reaction time interval was observed in both cases using MW irradiation, achieving high viscosity polymer solutions useful for further film casting within a 30–60 min (AB-PBI/TiO₂–pure AB-PBI) reaction time period of polycondensation being performed. The conventional heating approach, as a comparison, required a minimum 24 h reaction time period for achieving proper viscosity solutions and a high degree of imidazole cyclization for the preparation of high-molecular AB-PBI membranes possessing good physicomechanical properties. We observed a slight reduction of the apparent viscosity of the final composite AB-PBI/TiO₂ dispersions compared with the pristine PBI solutions in both cases. It could be attributed to the inherent viscosity modulation properties of the anatase nanoparticles/agglomerates phase on one side and to the kinetic polycondensation influence and side reactions on the other side. A detailed study of this observation will be the subject of further detailed investigations. An interesting UV illumination apparent color response of the pristine composite stock solution before and after aqueous membrane inversion was observed as well, which could be attributed to the complex TiO₂/AB-PBI interphase and solvent environment interactions. During the membrane water inversion process, different physicochemical changes occur both in the titania phase surface and in the polymer matrix, such as degree of protonation, hydrolysis of remaining P₂O₅, concentrated acid hydration, etc. Three different AB-PBI/20 wt.% plant extract TiO₂ composites, as well as pristine AB-PBI membranes, were prepared by film casting/direct water inversion methodology, resulting in self-supporting acid-doped hydrogel membranes with an average thickness of 80 microns. In order to remove the remaining surface mineral acid residues ionically bonded to the PBI matrix and to the titania particles from the reaction media (methane sulfonic and ortho-phosphoric acids), the obtained inverse membranes were further neutralized with basic ammonia solution, and finally, they were washed with deionized water in order to remove any remaining salt impurities within. This was performed with the presumption that deprotonated Brønsted acids could act as free radical scavengers, and this could have a negative effect on photocatalytic properties investigations and could also interfere with the antimicrobial properties interpretation. The very high chemical stability of titania anatase particles in the highly acidic reaction media was also a very useful property and it makes this kind of composite material preparation feasible.

3.3. Characterization of Prepared AB-PBI/Bio-TiO₂ Membranes

3.3.1. XRD Study

The powder X-ray diffraction patterns of the green synthesized TiO₂ and AB-PBI/ bio-TiO₂ membranes are displayed in Figure 7. The existence of the anatase TiO₂ phase (Ref. Code: 01-083-5914) was registered in pristine titanium dioxide and AB-PBI/ bio-TiO₂ membranes. The crystallinity peaks of AB-PBI are also registered for all hybrid membranes [68]. The calculated average crystallite size, lattice microstrain parameter, and unit cell parameter of the TiO₂ phases for pure bio-synthesized TiO₂ samples (M0, M1, and M2) are 11.98 nm, 11.53, and 11.48 nm; 1.37 × 10⁻³ a.u., 2.06 × 10⁻³ a.u., and 2.13 × 10⁻³ a.u.; a: 3.783 Å, 3.786, and 3.784 Å, respectively. The higher content of the plant extract leads to a slight decrease in the crystallite sizes of the obtained particles. Similar results about decreasing particle sizes of bio-synthesized ZnO using different amounts of neem leaf extract were established by [69]. This could be due to the phytochemicals in the leaf extract, which not only act as a stabilizing agent but also as a powerful reducing agent [69].

3.3.2. XPS Study

XPS was used to analyze the surface of the lyophilized membranes. Research shows that C1s, O1s, N1s, P2p, Na1s, and Ti2p peaks are registered on the surface. The atomic concentrations of the chemical elements on the surface of the membranes are represented in

Table 2. XPS results on the surface of the AB-PBI-TiO₂, M1 membranes (neutralized and acid-doped forms).

Samples	C, at. %	O, at. %	Ti, at. %	N, at. %	P, at. %	Na, at. %
AB-PBI-TiO2, M1 acid-doped form	57.0	26.8	0.3	9.6	5.3	1.0
AB-PBI-TiO2, M1 neutralized form	72.4	17.1	0.2	8.3	0.8	1.2

. The carbon spectrum is asymmetric, and shoulders toward higher binding energies are observed, which after decomposition are found to be due to C-C (285.0 eV), C-O/C=N (286.5 eV), C=O (288.3 eV), and COOH (only in the acidic form of the membrane) bonds [70]. These carbon bonds are also evidenced by the decomposed broad spectrum of oxygen, which also contains oxygen bound to phosphorus and oxygen in the water molecule. The nitrogen spectrum is associated with an imine -N=C and an amine -NH- group, with binding energies at 398.2 eV and 400 eV, respectively (Figure 8) [71]. The concentrations of up to 100% of the different C and N species are represented in

Table 3. The percentages of C and N in various species.

Binding Energy, eV	Chemical Bonding	Concentration, %
		AB-PBI-TiO2, M1 Acidic Form	AB-PBI-TiO2, M1 Neutral Form
285.0	C-C	61.4	76.1
286.5	C-O/C=N	26.0	14.6
288.3	C=O	9.5	9.3
289.4	COOH	3.1	-
398.2	Imine, -N=C	28.7	71.3
399.9	Amine, -NH-	27.9	72.1

. The phosphorus peak is attributed to metal phosphate with a binding energy of 133.6 eV [72], which is also proved by the Na1s peak with a binding energy of 1071.2 eV. A very small amount of titanium 0.2–0.3 at. % is also registered on the surface.

3.3.3. EPR Spectroscopy

EPR spectroscopy was used to characterize paramagnetic centers. In the pure polymer (polybenzimidazole PBI), an intense central signal with g₂ = 2.0025 and ∆H = 0.6 mT (peak to peak line width) is registered, along with two weakly intense satellite lines with g₁ = 1.984 and g₃ = 2.027 (Figure 9, spectrum 1). This triplet signal was previously observed in N-TiO₂ and is due to the interaction of an unpaired electron with a nuclear spin I = 1, attributed to N_b • centers formed by an unpaired 2p electron belonging to an N atom [73]. Some authors attribute this triplet to paramagnetic nitrogen oxides (such as NO, NO₂, NO₂⁻, NO₂²⁻) [74]. Additionally, a signal at g = 2.056 is registered, which could be due to adsorbed water, OH•, or absorbed oxygen O₂⁻.

The signal S1 at PBI/TiO₂ samples was less intensive than in the pure polymer. A slight shift of g-factors was observed: g₁ = 1.98, g₂ = 2.002, and g₃ = 2.015. On the other hand, a significant increase in the EPR intensity of the central peak of the triplet was observed after UV irradiation (for 40 min irradiation time, its intensity increases two times) (Figure 9, spectrum 2 and 3). As the irradiation time increases, the S1 signal intensity continues to increase. Some authors have found that bulk N_b• species are photo-dependent (photoactive), unlike NO. In our experiment, we observed an increase in signal (S1) intensity with g = 2.0025; therefore, the signal is probably due to N_b• centers particles [75]. In a study of N-doped TiO₂, some authors attributed this signal to an electron trapped in an oxygen vacancy O⁻ [74].

A broad asymmetric signal S2 was also recorded in the PBI/TiO₂ samples, possibly due to the overlap of two signals with close g-factors. The superimposed line recorded in the frame of 315–327 mT magnetic field may be due to adsorbed water or absorbed oxygen O₂⁻ [54,55] and OH• radicals adsorbed on the TiO₂ surface [53].

Ti³⁺ ions are characterized by very short relaxation time, so their EPR spectrum is better observed at low temperatures (Figure 9, spectrum 4). A broad, weakly intense line S3 appears at g = 1.93. Signals at g < 2 may be due to electrons captured by the titanium ion. The broad EPR line indicates the formation of Ti³⁺ particles at surface sites.

3.3.4. SEM and EDS Mapping/Spectra Analysis of the AB-PBI/Bio-TiO₂ Membranes

Figure 10 and Figure 11, and Figure S1 show in-plane SEM images, and Figure S2 presents EDS spectra of the synthesized AB-PBI/bio-TiO₂, M1 (neutralized and acid-doped forms) membranes. The presence of individual and clustered micron- and submicron titania particles on the membrane surface is evidenced in both ex situ acid-doped forms and in the ammonia-neutralized membranes. The EDS spectra show mainly the presence of C, O, and Ti for both AB-PBI/bio-TiO₂, M1 (basic and acid-doped forms) membranes. The P and S contents in the AB-PBI/bio-TiO₂, M1 (acid-doped form) were due to the Eaton’s reagent used (methansulfonic acid/P₂O₅) during the polycondensation process and the plant extract. The observed peaks of the other elements could be attributed to the Mentha Spicata plant extract used that adhered to the titania phase surface during the synthesis. The EDS results of synthesized AB-PBI/bio-TiO₂, M1 (basic and acid-doped forms) membranes, and Mentha Spicata plant are represented in

Table 4. EDS results of prepared AB-PBI/bio-TiO₂, M1 (basic form) membrane.

Element	wt. %	Atomic %
C	75.28	83.37
O	16.96	14.10
Na	0.86	0.50
Cl	1.30	0.49
K	0.80	0.27
Ti	4.04	1.12
Zn	0.76	0.15
Total	100.00	100.00

, Table S1 and Table S2. The EDS mapping analysis (Figure 11) of the prepared AB-PBI/bio-TiO₂, M1 (basic and acid-doped forms) membranes is in agreement with the conventional SEM observations, as seen from the images.

3.3.5. FTIR Investigations

The FTIR spectra of the green synthesized TiO₂ and AB-PBI/bio-TiO₂ membranes are shown in Figure 12. In the FTIR spectra of green synthesized TiO₂, one can observe a broad band around 3389 cm⁻¹ corresponding to the stretching vibration of the O–H bond. The peak at 1633 cm⁻¹ is attributed to the O–H bending vibration of adsorbed water molecules on the surface of synthesized TiO₂. The peaks within the region 443 cm⁻¹–453 cm⁻¹ are associated with the Ti–O bending vibration, characteristic of the formation of metal–oxygen bonding [76,77]. The pristine AB-PBI represents some characteristic absorption bands at 1602, 1547, and 1286 cm⁻¹ arising from the C–N stretching, –N–H stretching, and from the bending mode of the imidazole ring, respectively [78].

The appearance of new bands at 2920 and 2894 cm⁻¹ in the FTIR spectrum of AB-PBI/TiO₂, M2 due to the C–H stretching mode confirms the presence of organic guest moieties. Some common bands that were also present in the pure AB-PBI and AB-PBI/TiO₂, M1 membrane showed the insertion of TiO₂ in the polymer matrix without changing its chemical structure.

With the inclusion of TiO₂ in the composite membranes, the peak attributed to the N-H stretching broadens to a wider range of 3400–2920 cm⁻¹. This is due to the presence of an H-bond in the PBI repeating unit and the TiO₂ particles. This is also an indication of a good distribution of TiO₂ inside the polymer matrix.

3.3.6. Thermogravimetric Analysis (TG) of the Prepared AB-PBI/Bio-TiO₂ Membranes

TG curves of lyophilized pristine AB-PBI and different AB-PBI/bio-TiO₂ composite membranes (Figure 13) revealed the characteristic high chemical decolorization temperature of AB-PBI starting above 650 °C. Initial loss of up to ∼300 °C is attributed to the free moisture and the loss of the imidazole H-bonded water molecules, as observed with other aromatic polybenzimidazoles [79]. The closely packed macromolecular structure of AB-PBI compared with the much loosely-packed meta- and para-PBIs is responsible for the sluggish release of water due to stronger high-density imidazole H-bonding—NH group density, defined as the percent of molar mass of N–H group per repeat unit = 12.9%, compared to m-PBI (NH group density = 9.7%) [80]. Similar moisture loss was noted for both analyzed composite samples with incorporated plant extract synthesized titania particles with a higher degree of crystallinity (samples M1 and M2). The char yields at 800 °C of both M1 and M2 titania composite samples (30 and 35% resp.) were found to be much lower than those of the pristine AB-PBI (45%). This is an indication of the efficient incorporation of nanocrystalline TiO₂ particles within the AB-PBI matrix, as the titania anatase phase is thermooxidation stable even at this relatively high polymer carbonization temperature.

3.4. Photocatalytic Study of Prepared AB-PBI/Bio-TiO₂ Membranes

The photocatalytic activity of the prepared AB-PBI/bio-TiO₂ membranes has been studied in the reaction of the photocatalytic decolorization of Reactive Black 5 dye as a model pollutant in aqueous solutions (Figure 14, Figure 15 and Figure 16). In the presence of AB-PBI/bio-TiO₂, M1 and M2 (neutralized and acid-doped forms), the photocatalytic decolorization of RB5 dye proceeds in two stages (0–60 min and 90–180 min), while in the case of AB-PBI/bio-TiO₂, M0 (basic and acid-doped forms), it is in one stage (0–180 min). The values of the calculated apparent rate constants are AB-PBI/bio-TiO₂, M1 (basic form) > AB-PBI/bio-TiO₂, M1(acid-doped form); AB-PBI/bio-TiO₂, M2 (basic form); AB-PBI/bio-TiO₂, M2 (acid-doped form); AB-PBI/bio-TiO₂, M0 (basic form); AB-PBI/bio-TiO₂, M0 (acid-doped form) for 180 min under UV irradiation. The basic forms of all investigated AB-PBI/bio-TiO₂ membranes demonstrated higher photocatalytic ability compared to the acidic forms. The highest degree of decolorization of the model dye was achieved in the presence of photocatalyst membranes AB-PBI/bio-TiO₂, M1 (96% and 89% for the basic and acid forms, respectively). Composites M2 show slightly lower decolorization values, 88% and 56%, for the neutralized and acid-doped forms, respectively. Membranes containing MO particles have the lowest value of photocatalytic decolorization, 57% and 58% for the basic and acid-doped forms, respectively.

Table 5. The removal of organic pollutants using hybrid photocatalysts by literature data.

Materials	Processes	Type of Pollutant Removed	Lamp Source	Removal Efficiency	References
PVDF@TiO2	blending, dip coating, and grafting	Methylene blue dye	UV light	95.4%	[81]
PVA /TiO2	electrospinning	Rhodamine B dye	Vis light	80%	[82]
PANI-TiO2	template free method	Reactive Black 5 dye	Vis light	96%	[83]
PPy-TiO2	chemical oxidative polymerization method	Methylene blue dye	Solar light	93%	[84]
PVDF/TiO2	phase inversion Method	brilliant green dye	UV light	~45%	[85]
TiO2/PFT	surface modification	phenol	Vis light	74.3%	[86]

gives the results presented in the literature about the removal of organic pollutants using hybrid photocatalysts.

It is well known that the presence of point defects and oxygen vacancies in the TiO₂ lattice influences many surface reactions. Among them, Ti³⁺ ions are significant reactive agents for many adsorption processes. They can induce the formation of oxygen vacancies [87]. It was proved that both Ti³⁺ and oxygen vacancies have a synergistic effect on the rate of the photocatalytic reactions due to their possibility to separate the light-induced electron–hole pairs (thanks to the formation of sub-levels in the conduction band of TiO₂) and increase its redox capacity [88]. The samples AB-PBI/bio-TiO₂, M1 exhibit the highest photocatalytic reaction rate towards the azo dye as a result of the optimal combination of the presence of Ti³⁺ defects, oxygen vacancies, crystallites sizes, degree of crystallinity, decreased band gap value, and homogeneous distribution of the nanocrystallites (proved by TEM, XRD, XPS, and EPR analyses). The presence of aggregates composed of particles having different sizes in M2 samples negatively influences their photocatalytic activity by the following physicochemical effects: (i) decreased light intensity reaching the surface, thus resulting in a reduced quantum yield; (ii) less active sites, available for the dye adsorption on the surface of the membrane. These above-mentioned effects could be responsible for the slower decolorization rate of the dye in the presence of AB-PBI/bio-TiO₂, M2 composites.

The following mechanism of photocatalytic reaction could be proposed: There are probably three mechanisms occurring simultaneously. (1) Direct oxidation of pollutant molecules on the surface of the TiO₂ particles–electron transfer to positively charges holes in the valence band-heterogeneous catalysis. (2) In homogeneous catalysis, there are two mechanisms of indirect oxidation: the interaction of pollutant molecules with hydroxyl radicals (a radical-chain mechanism) and their interaction with oxygen radical anions. Figure 15 shows the degree of decolorization of RB5 dye during the UV irradiation time period using AB-PBI/bio-TiO₂ membranes (neutralized and acid-doped forms) as photocatalysts. Figure 16 represents UV–Vis absorption spectra of RB5 dye during the irradiation time period using AB-PBI/bio-TiO₂, M1 (basic form) as the photocatalyst.

Figure 17 presents the adsorption capacities of AB-PBI/bio-TiO₂, M0, M1, and M2 (neutralized and acid-doped forms) membranes after a 30 min dark period. The highest adsorption capacity is demonstrated by AB-PBI/bio-TiO₂, M2 (acid-doped form). The lowest adsorption capacity is demonstrated by AB-PBI/bio-TiO₂, M0 (neutralized form). The adsorption capacities of AB-PBI/bio-TiO₂, M1 and M2 (neutralized forms) are equal. In the available current literature, the data about the influence of the adsorption ability on the photocatalytic efficiency are quite contradictory. In our case, the higher adsorption abilities of all the types of membrane contact do not lead to any enhancement of the photocatalytic characteristics. It can be concluded that the photocatalytic reaction rate does not depend on the adsorption capacity of the AB-PBI/bio-TiO₂ membranes.

The investigated AB-PBI/bio-TiO₂ photocatalysts preserved their photocatalytic ability relatively well after three photocatalytic runs, which, in fact, demonstrates their suitable long-term stability (Figure 18). As can be seen, the degree of RB 5 dye decolorization decreased very slightly, starting from the first until the third cycle.

3.5. Antibacterial Activity of Prepared AB-PBI/Bio-TiO₂ Membranes

The quantitative RMDA method, applied by us, showed that the TiO₂ nanopowders had no antimicrobial activity at the tested concentrations (5, 20 mg/mL), both under static and dynamic conditions. In comparison, Chakansin et al. (2022) [89] report resazurin bleaching at approximately > 50 mg/mL of TiO₂. The thorough review of Serov et al. (2024) [44], which uses the data from tens of publications, points out TiO₂ NPs MICs range from 0 to approx 2500 µg/mL.

The data on TiO₂ NP antimicrobial activity not only vary widely in MIC values, but they are contradictory in view of the requirement for visible or UV light irradiation. Some authors report the antimicrobial activity of TiO₂ in the dark [90,91,92]. According to other authors, the antimicrobial application of TiO₂ requires illumination in the visible or the UV spectrum [40,41,93,94]. Our experiments also show definite antimicrobial activity of titanium dioxide nanoparticles only under UV light irradiation. This has been demonstrated by the parallel testing of a bacterial suspension containing 0.5 mg/mL of each TiO₂ nanopowder and the same suspension irradiated by UV light. Both suspensions were under continuous stirring to avoid particle sedimentation. The results for M0, M1, and M2 showed a significant reduction in CFU as early as the 10th minute of exposure by three exponents and a further reduction of up to four exponents at the 20th minute. Another interesting result was observed for M1 and M2: at the 30th minute of irradiation, the number of registered living cells was greater than that at the 10th and 20th minutes but still remained significantly lower (by two to three exponents) than that of the non-irradiated suspension (Figure 19). The same phenomenon was described by Sirelkhatim et al. (2015) [45]. A slight enhancement of the antibacterial effect was observed in the sequence M0 < M1 < M2, which coincides with the increase of the O_total/O_lattice ratio of the nanoparticles.

In order to determine whether the composite materials have antimicrobial activity in the dark, we applied the ASTM method, which is aimed at testing the antimicrobial properties of polymer materials in dynamic contact with bacterial suspension. Similar to the results from the RMDA procedure, this method showed no antimicrobial activity of the composite membranes in the dark. A slight stimulatory effect of the materials towards bacterial cells was even observed, which is probably due to the properties of the PBI carrier membrane (Figure 20).

The antimicrobial properties of the composite membranes were clearly manifested after irradiation by UV light (Figure 21). All three composites showed very good antimicrobial activity, with CFU counts decreasing by two to three logarithms after 10 min of irradiation. The presence of TiO₂ NPs in the membranes enhances the effect of UV light, and it leads to a stronger reduction in CFU compared to the control sample, which contains only bacterial cells and it also shows some reduction in CFU due to the UV light. The AB-PBI/bio-TiO₂, M1 material showed the best antimicrobial effect.

Remarkably, the test samples of pristine AB-PBI membrane in neutral form did not show a decrease in cell number upon UV-light irradiation (Figure 21). As it was expected, the control samples containing only bacterial suspension showed a decrease in the number of cells as a result of the UV-irradiation by approx. one logarithm. It is likely that pure AB-PBI membrane possesses distinguished UV-photosensitization properties, which partially neutralize the bactericidal effects of UV radiation. Taking into account the well-known common sunscreen low-molecular constituent, 2-phenylbenzimidazole-5-sulfonic acid (PBSA), this hypothesis is highly likely. In a previous experiment, which aimed to test the effect of the combination of cellulose membrane surface impregnated with meta-PBI, we observed an increase in the number of viable E. coli and B. subtilis cells over 120 h [52]. To the best of our knowledge, such an experiment using AB-PBI membrane is conducted for the first time, and the data obtained from it, are pristine for further investigation. In the scientific literature available to us, we could not find any studies of the effect of any PBIs on microbial culture growth or inhibition. Further experiments are needed to investigate these interesting features.

• Bacterial morphology assessment by SEM

After UV irradiation, during cultivation on composite membranes AB-PBI/bio-TiO₂, M0, swellings on the cell surface were observed (indicated by white arrows). We observed similar structural changes in our previous studies on Pseudomonas aeruginosa biofilms treated using cationic polymer micelles loaded with silver nanoparticles [36]. Moreover, in the represented micrograph of this double-treated sample, a certain destructive effect on the bacterial cells is evident. Localized damage appears on the bacterial cell wall, likely affecting the underlying membranes, resulting in rupture and separation of cell contents (indicated by white triangles) (Figure 22a).

It is important to note that for normal cellular function, the permeability of the bacterial cell wall and membranes is vital due to their protective role in oxidative phosphorylation mechanisms. In contrast, in the untreated sample, the bacterial cell morphology was relatively preserved, except for the appearance of swellings in localized areas of the cell wall. A residual amorphous substance was also prevalent around or on the bacterial cells (indicated by white stars) (Figure 22b). In all the double-treated samples, following UV activation of TiO₂ nanoparticles, shortened cells were observed, a finding not visible in the case of unirradiated samples (Figure 22a,c,e). This heterogeneity in the bacterial population is probably due to the combined effects of UV irradiation and the concentration of added TiO₂ nanoparticles. These factors likely delayed cell growth and they altered the physiological activity of the cells, as it is confirmed by the antibacterial effect analysis (Figure 20 and Figure 21). Additionally, a reduction in the number of adhered bacterial cells was reported in all three treated samples (Figure 22b,d,f) compared to the untreated ones. In the non-irradiated sample AB-PBI/bio-TiO₂, M2, the localization of the cells resembled biofilm consortia with a ubiquitous, multilayered architecture (Figure 22). Suppression of bacterial adhesiveness is a key mechanism in preventing the spreading and colonization of new niches by pathogenic bacteria, which are associated with infectious and biofilm diseases [95]. The accumulated amorphous substance likely contains TiO₂ nanoparticles, which, in samples AB-PBI/bio-TiO₂, M0 and AB-PBI/bio-TiO₂, M1, surround some of the bacterial cells (Figure 22a,b,d). In sample M2, this substance is primarily concentrated in the underlying substrate, forming small island-like protrusions (Figure 22e). Both in AB-PBI/bio-TiO₂, M1 and in AB-PBI/bio-TiO₂, M2 samples, bacterial cells with atypical shapes and disorganization in the cell wall manifested as indentations (indicated by white arrows) were observed, suggesting a destructive effect resulting from the treatment (Figure 22c,e). We assume that the damage to the cell structure is probably due to the presence of reactive oxygen species (ROS) generated by the irradiation of TiO₂ nanoparticles embedded in the membranes, as confirmed by other researchers [46,96,97].

4. Conclusions

A series of novel AB-Polybenzimidazole/green synthesized TiO₂ (20 wt.%) membranes (in basic and acidic form) were successfully prepared. It established the optimal MS extract quantity for the synthesis of TiO₂ particles for incorporation in AB-PBI membranes having high photocatalytic and antimicrobial activity. It has to be noted the key role of the plant extract for the decrease of the band gap value, defective lattice structure, and distribution of the obtained titania particles. Both EPR and XPS analyses proved that both the Ti³⁺ and oxygen vacancies in the lattice increase with the increase of the extract quantity. By means of TEM, studies have revealed that the MS leaf extract plays a stabilizing and capping role, preventing the aggregation of the particles. We evaluated the optimal plant extract quantity, leading to a homogeneous distribution of the particles.

The AB-PBI/bio-TiO₂ membranes possess high photocatalytic activity towards the decolorization of RB 5 dye under UV light irradiation (96% and 89% for neutralized and acid-doped forms, respectively)). All basic membranes exhibit long-term stability under three consecutive photocatalytic cycles. The newly synthesized AB-PBI/bio-TiO₂ membranes have a powerful antibacterial effect on E. coli when irradiated by UV light. The presence of TiO₂ NPs in the membranes enhances the effect of UV light and it leads to a stronger reduction in CFU compared to the control sample. The combined effects of UV irradiation and the TiO₂ particle concentration likely delayed cell growth and altered the physiological activity of the cells.