Enhanced Photo-Catalytic Performance of Activated Carbon Fibers for Water Treatment

: The synthesis, characterization, and performance of composite photocatalytic adsorbents are investigated in this work using the dip-coating and the electrophoretic coating methods for the deposition of titanium dioxide (TiO 2 ) on porous activated carbon ﬁber (ACF) substrates. The adsorption and photocatalytic e ﬃ ciency of the synthesized catalytic adsorbents were compared using phenol as the model pollutant. Both immobilization techniques resulted in composite ACF / TiO 2 adsorbents characterized by large surface area (844.67 ± 45.58 m 2 g − 1 ), uniform distribution of TiO 2 nanoparticles on the activated carbon ﬁbers, and high phenol adsorption. The method and the treatment time a ﬀ ected the phenol adsorption, while the highest sorption was determined in the case of the ACF / TiO 2 sample prepared by the electrophoretic coating method (at 20 V) for an electrolysis time of 120 s (7.93 mg phenol g − 1ACF / TiO2 ). The UV-A irradiation of most ACF / TiO 2 samples led to a faster removal of phenol from water as a result of the combined sorption and heterogeneous photocatalysis. The stability and the e ﬀ ective regeneration of the most promising composite photocatalytic adsorbent was proved by multiple ﬁltration and UV-A irradiation cycles.


Introduction
In the last two decades, intensive research activity has been ongoing toward the development of so-called advanced oxidation processes (AOPs) as effective and environmentally friendly methods for disinfection, chemical detoxification, and removal of taste and odor compounds in water and wastewater [1][2][3]. The remarkable efficiency of AOPs is based on the generation of highly reactive hydroxyl radicals ( • OH) that are powerful, non-selective, oxidizing species, which can indiscriminately degrade organic pollutants until their mineralization leads to CO 2 , H 2 O, and, eventually, inorganic ions as final products [4]. AOPs have the advantage of being environmentally-friendly, as they neither transfer pollutants from one phase to the other (as in chemical precipitation and volatilization), nor produce massive amounts of hazardous side-streams [5]. In comparison to chemical and biological processes, AOPs operate at near ambient conditions, are less energy-demanding, and do not require special installations [6]. The growing interest of academic and industrial communities in AOPs is reflected in the increasing number of publications in several peer-reviewed journals, patents, and international conferences dedicated to the environmental applications of AOPs. For example, over 7000 articles have been published in journals of the Science Citation Index in the last five years dealing with processes such as photocatalysis, ozonation, Fenton reactions, electrochemical processes, wet air oxidation, and ultrasonic cavitation; about 25% of them are devoted to photocatalytic processes. AOPs can provide effective technological solutions for water treatment, thus supporting and enhancing the competitiveness of different industrial sectors, including the water technology sector in the global market.
The most popular technique among AOPs is the heterogeneous photocatalysis, where a semiconductor photocatalyst interacts with light of sufficient energy (or of a certain wavelength) to produce reactive oxidizing species, which can lead to the photocatalytic degradation of pollutants [7]. Large scale application of heterogeneous photocatalysis may be constrained, however, by technical challenges mainly related to the effective recovery and extended use of the catalyst particles in a continuous process [8,9]. An alternative promising approach is the immobilization of photocatalysts on porous adsorbents, such as carbon, clays, zeolite, and others, which offer the synergistic combination of adsorption and catalytic degradation of the adsorbed pollutants by UV/visible light irradiation. Such an approach allows continuous reuse of the photocatalyst and improved removal of pollutants from water [10].
Activated carbon fibers (ACFs) are microporous carbon materials that present uniform pore size distribution, larger surface area, and faster adsorption rates than other activated carbons, such as granular and powdered [11]. Moreover, ACFs can be made in various forms, such as yarn, thread, fabric, and felt, making them suitable for various applications [12]. These features justify the selection of ACFs as a potential substrate for the successful immobilization of catalytic particles, semiconducting metal oxides in particular, for photocatalytic and photoelectrochemical applications [13,14]. The high adsorption capacity of ACFs, and of carbon-based supporting materials in general, helps form a high concentration environment of adsorbates around the catalytic particles, resulting in an increase in the photo-degradation rate [15]. Among the many different photocatalysts proposed in literature, titanium dioxide (TiO 2 ) has been the most widely studied and used in many applications because of its greater photocatalytic activity compared to other semiconductors, its long-term stability and durability under extreme conditions (pH, temperature, ultraviolet radiation, high shear stresses), and its high resistance to chemical and photodegradation, that all favor its recycling and reuse. Moreover, TiO 2 is superhydrophilic, biologically inactive, environmentally friendly, inexpensive, and commercially available [16].
The immobilization of TiO 2 on porous substrates has been investigated by different methods, such as sol-gel, dip-coating from suspension, electrophoretic coating, hydrothermal, etc. [17]. Table 1 lists representative works of composite TiO 2 adsorbents employing as substrates, specifically carbon-based materials in the form of fibers. Most research groups applied the dip-coating method for TiO 2 immobilization onto carbon fabrics, usually followed by calcination at elevated temperatures (varying from 300 to 750 • C) in argon or nitrogen atmosphere to obtain highly crystalline TiO 2 . A TiO 2 sol was applied in most studies using tetrabutyl orthotitanate (Ti(OBu) 4 ) as the titanium precursor and ethanol as the solvent. The gradual evaporation of ethanol resulted in Ti(OBu) 4 hydrolysis into anatase TiO 2 with the water vapor in the moist environment. Only a few studies used mixed-phase TiO 2 catalysts for this purpose, such as the commercially available Degussa P25 TiO 2 nanoparticles (25% rutile and 75% anatase according to the manufacturer) [18], although this has been proven as the leading semiconductor photocatalyst for the degradation of a multitude of refractory organic contaminants [19,20]. This trend is probably due to the distinct features of anatase, such as the lower rates of recombination [21] and higher surface adsorptive capacity [22] that renders it the most photochemically active phase of titania. Photocatalytic studies with composite catalytic adsorbents show, in general, high degradation rates of contaminants, which greatly vary depending on the immobilization method, the properties of the porous adsorbent (i.e., specific surface area, pore volume, pore size distribution), and the treatment conditions (initial solution pH, radiation intensity, TiO 2 loading).
A promising alternative to the dip-coating/sol-gel method is the electrophoretic coating method, which has been proven to be a simple and effective method for obtaining reproducible and highly active photocatalytic materials [23]. In a recent work by Pereira et al. [24], TiO 2 electrodeposition on both carbon fiber and boron doped diamond/carbon fiber substrates using TiO 2 sol as the titanium precursor led to controlled anatase formation and to the production of electrodes with different photocatalytic activities that were largely affected by the TiO 2 deposition time (increased thickness and homogeneity with time) and the nature of the substrate. Motivated by the promising attributes of electrophoresis (i.e., easy method employing simple potentiostatic electrolysis and post-heating at low temperatures, synthesis of reproducible, recyclable, and high photocatalytic activity adsorbents), the present work investigates the adsorptive and photocatalytic activity of ACF/TiO 2 composites prepared by the electrophoretic coating method using phenol as the model organic pollutant. For comparison, ACF/TiO 2 composites are also prepared according to a dip-coating method using titanium (IV) isopropoxide and Degussa P25 as the titanium source. The effect of coating time and the source of titanium dioxide on the surface and structural characteristics of the composite fiber materials are systematically investigated. In-situ deposition Commercial carbon fiber Fabricated three-dimensional electrode composed of dendritic Ag@Pt core-shell catalyst, reduced graphene oxide, TiO 2 spheres, and carbon fiber exhibited improved photo-electrocatalytic performance for methanol oxidation compared to other comparative electrodes [28] Mixing with Ti(OH) 4 and H 2 O 2 Commercial carbon fiber The photocatalytic degradation of Methyl Orange found to be phase composition-dependent and pH dependent [29] Dip-coating and annealing under superheated steam Polyacrylonitrile carbon fiber TiO 2 /CF composites achieved up to 98.7% degradation rate of Acid Orange II after 2.5 h of irradiation [30] Hydrothermal Commercial activated carbon fiber The composite ACF/TiO 2 presented good uniformity, high crystallinity, and large benzene photo-oxidation and sorption affinity [14] Ultra-sonication induced adsorption or electrospinning Polyacrylonitrile nanofiber TiO 2 nanoparticles decorated carbon nanofibers, prepared by ultra-sonication, presented higher Methylene Blue adsorption capacity and photo-catalytic efficiency than those obtained by electrospinning [18]

Materials
Activated carbon fibers (ACFs) in felt form, supplied by SO-EN Co. Ltd. (Takasaki, Japan) (code number ACFNW-EM3), were used in this study as substrate for the immobilization of TiO 2 . ACF specimens were cut into disks of~59 mm diameter and pretreated before use ( Table 2). All ACF specimens were weighted before and after the immobilization of TiO 2 for assessing the loading of catalyst on the adsorbent surface. Titanium (IV) isopropoxide (TTIP), purchased from Aldrich (≥97%, Steinheim, Gemrnay), and Aeroxide ® P-25 (Degussa-Evonik, Frankfurt, Germany) were used as titanium precursors. High purity Milli-Q water (resistivity 18.2 MΩ, Millipore, Milford, MA, USA), concentrated solutions of nitric acid (65% w/w, Chemlab NV, Steinheim, Germany), and sulfuric acid (95%-97% w/v, Sigma-Aldrich, Steinheim, Germany) were used for the preparation of the composite ACF/TiO 2 adsorbents. Photocatalytic degradation studies were performed with aqueous solutions of phenol (≥99%, Sigma-Aldrich, Steinheim, Germany) at 10 mg L −1 in deionized water.   Table 2 summarizes the immobilization protocol and the respective conditions applied for each of the two TiO 2 coating methods investigated in this work. The pretreatment and the electrophoretic deposition were carried out at room temperature using a VersaSTAT3 potentiostat/galvanostat (Princeton Applied Research, AMETEK, Oak Ridge, TN, USA), connected to a PC for continuous monitoring of cell current (VersaStudio software, V3-200, Oak Ridge, TN, USA). These operations were performed in a transparent three-electrode Plexiglas cell using a stainless steel rod as a counter electrode and an Ag/AgCl electrode (E = 0.207 V vs SHE) from Metrohm as the reference electrode. Electrolysis was applied for the pretreatment of all ACF specimens, aiming to oxidize them and impart some hydrophilicity as ACFs are highly hydrophobic.

Characterization
The surface area and pore volume of the samples were measured by nitrogen adsorptiondesorption with a TriStar 3000 porosimeter (Micromeritics, Norcross, GA, USA). The specific surface area (A) of the electrodes was measured at liquid nitrogen temperature (77 K), using the Brunauer−Emmett−Teller (BET) method in the relative pressure (P/P0) range of 0.005-0.50. The morphologies of the composite ACF/TiO 2 adsorbents and the original ACF were examined using a JSM6300 microscope (JEOL, Peabody, MA, USA) equipped with an X-ray Microanalysis-Energy Dispersive Spectroscope (EDS), operating at 20 kV. The samples were gold sputtered to avoid charging effects on the images.

Adsorption and Photocatalytic Degradation Studies
Adsorption and photocatalytic degradation tests were performed in a magnetically stirred closed cylindrical vessel made of transparent Plexiglas (>90% transmittance at wavelengths >320 nm) of 600 mL volume. ACF and ACF/TiO 2 specimens were placed in a special Teflon plate (outer/inner diameter 60/51 mm) and immersed in the vessel at a close distance to the wall. For the scope of the electrophoretic deposition, a stainless steel ring was also placed in the Teflon plate in contact with the ACF and used as a charge carrier (connection to the potentiostat by means of a Cr/Ni wire). Photocatalytic experiments were conducted using a low pressure ultraviolet mercury lamp (Blackight, 365 nm, 5 W) placed outside the Plexiglas vessel at a distance of 2-3 cm from the Teflon plate (and the ACF/TiO 2 adsorbent). Samples were collected from the vessel at certain time intervals for determining residual phenol concentration using a UV-vis spectrophotometer (Spectroquant ® Pharo 300, Merck, Darmstadt, Germany) at 270 nm. Measurements of the UV-Vis spectra of samples were also performed, in the range 190-390 nm, for the qualitative assessment of the oxidative reactions taking place in the phenol solution. It is noted that all experiments were conducted in duplicate, and the averaged measured values are reported in this study. The variation was estimated to be within ±8% and ±14% for the adsorption and photocatalysis experiments, respectively. Figure 1 shows SEM photomicrographs of the ACF specimens before and after the immobilization of TiO 2 nanostructures on their surface. According to Figure 1a, the original ACF consists of unidirectional bundles of pure carbon fibers of about 12 µm in diameter. Figure 1b-d indicate that all ACF fibers were covered by TiO 2 nanostructures of different size and morphology depending on the applied coating method and the deposition time. The presence of Ti on the outer surface of the fibers is also confirmed by the corresponding EDS analyses, according to which the characteristic Ti peaks at 4.6 and 4.95 keV appear only in the modified fibers.

Characterization of ACF/TiO 2 Composite Adsorbents
As expected, the dip-coating method resulted in compact and large TiO 2 agglomerates on the fiber surface (Figure 1b,c). This is particularly evident in the case of TTIP rather than in Degussa P-25. Specifically, the use of TTIP as a titanium precursor led to the formation of uniform aggregates with increased thickness at longer deposition times. In contrast, Degussa P-25 displayed irregular structures of TiO 2 crystals, which were unevenly distributed on the outer surfaces of the fibers. The electrophoretic method promoted the uniform distribution of small spherical TiO 2 nanoparticles on the fiber surface (Figure 1d), which can contribute to a better and more stable photocatalytic performance, as discussed next.
The porous structure of the ACF was not significantly altered by the deposited TiO 2 nanoparticles, regardless of the immobilization method and the source of the used titania. As shown in Table 3, specific surface area and average pore volume were reduced by ∼29.4%, 24.4 και 1.2% in the case of the ACF-TTIP, ACF-P25, and ACF-EDP samples, indicating that the pores of the ACF were moderately blocked by the loaded TiO 2 . In the case of electrophoresis, the deposition of TiO 2 had negligible effect on the surface characteristics of ACF fibers, with the highest electrophoretic coating time (120 s) leading to improved specific surface area and mean pore volume (Table 3). Obviously, some TiO 2 nanoparticles are squeezed to the fiber surface, thereby increasing the surface roughness and the specific surface area.
isotherm seems to fit better the results of phenol adsorption on ACF (R 2 = 0.958), implying that a monolayer adsorption takes place on the energetically homogeneous ACF surface, with no transmigration of adsorbate molecules in the plane of the surface. The driving forces for adsorption of phenol probably originate from physisorption based on hydrophobic π-π interactions between the carbonaceous surface and the aromatic ring of the phenol molecules. Chemisorption of phenol molecules due to the interaction with the edges of the basal planes in the carbon fibers is also possible.  (Table 1).
The adsorption rates of phenol molecules on the nine prepared ACF/TiO2 composite adsorbents are shown in Figure 3. The respective pseudo-second-order reaction kinetics data, also shown in the inset of Figure 3, indicate that adsorption performance is in the following order: ACF-TTIP-10 > ACF-EDP-30 > ACF-TTIP-30 > ACF-EDP-60 > ACF-EDP-120 > ACF-TTIP-60 = ACF-P25-30 = ACF-P25-60 > ACF-P25-10. In comparison to the original ACF (Figure 2a), the kinetic constants significantly decreased from 47% to 93%. However, the respective amounts of phenol adsorbed per unit mass of  (Table 1).  Figure 2 shows the results of preliminary adsorption tests performed with the unmodified ACF. A rather rapid adsorption is observed, and the experimental data are best correlated by pseudo-second-order reaction kinetics (Figure 2a, inset). This implies that the occupation rate of adsorption sites is proportional to the square of the number of unoccupied sites. The values of adsorption constants and correlation coefficients obtained for two well-known adsorption models (Freundlich and Langmuir) are presented in the inset of Figure 2b. The adsorption isotherms were obtained by fitting the adsorbed amounts of phenol on the ACF surface (in units mg Phenol g ACF −1 ) and the equilibrium phenol concentration ([Phenol] e ) in aqueous solutions (Figure 2b). The Langmuir isotherm seems to fit better the results of phenol adsorption on ACF (R 2 = 0.958), implying that a monolayer adsorption takes place on the energetically homogeneous ACF surface, with no transmigration of adsorbate molecules in the plane of the surface. The driving forces for adsorption of phenol probably originate from physisorption based on hydrophobic π-π interactions between the carbonaceous surface and the aromatic ring of the phenol molecules. Chemisorption of phenol molecules due to the interaction with the edges of the basal planes in the carbon fibers is also possible.

Phenol Adsorption on Original ACF and ACF/TiO 2 Composites
ACF/TiO2 adsorbent (at the pseudo-equilibrium time of 120 min) were lower, varying from 21.9% to 65.6% (Table 3). This difference may be attributed to the mass loss of the composite fibers (smaller denominator) due to their electrolytic and thermal pre-oxidation. Based on the experimental data, the adsorption capacity differed between the composite adsorbents, depending on the method and the coating time applied, being greater in the case of adsorbents prepared by the electrophoretic method. According to Table 3, such an adsorption performance is justified by the high specific surface area of the ACF-EDP composites, especially of the one prepared for 120 s of electrodeposition.

Phenol Photocatalytic Degradation by ACF/TiO2 Composites
The results of the photocatalytic degradation tests with the ACF/TiO2 composites are summarized in Figure 4. In the same Figure, the results are compared with those corresponding to phenol adsorption on unmodified ACF and ACF/TiO2 composites. A reference photolysis test in the absence of carbon fibers was also performed, showing the limited effect of UV-A irradiation on phenol degradation (14.3% removal). This was expected, since the blacklight lamp used emits at 365 nm, which is far from the maximum UV absorption of phenol (270 nm). In comparison to the simple adsorption tests, the respective photocatalytic experiments resulted in improved removal efficiencies for most of the composite ACF/TiO2 adsorbents tested due to the combined effects of adsorption and The adsorption rates of phenol molecules on the nine prepared ACF/TiO 2 composite adsorbents are shown in Figure 3. The respective pseudo-second-order reaction kinetics data, also shown in the inset of Figure 3, indicate that adsorption performance is in the following order: ACF-TTIP-10 > ACF-EDP-30 > ACF-TTIP-30 > ACF-EDP-60 > ACF-EDP-120 > ACF-TTIP-60 = ACF-P25-30 = ACF-P25-60 > ACF-P25-10. In comparison to the original ACF (Figure 2a), the kinetic constants significantly decreased from 47% to 93%. However, the respective amounts of phenol adsorbed per unit mass of ACF/TiO 2 adsorbent (at the pseudo-equilibrium time of 120 min) were lower, varying from 21.9% to 65.6% (Table 3). This difference may be attributed to the mass loss of the composite fibers (smaller denominator) due to their electrolytic and thermal pre-oxidation. Based on the experimental data, the adsorption capacity differed between the composite adsorbents, depending on the method and the coating time applied, being greater in the case of adsorbents prepared by the electrophoretic method. According to Table 3, such an adsorption performance is justified by the high specific surface area of the ACF-EDP composites, especially of the one prepared for 120 s of electrodeposition.
percentage removal of phenol (~93%) being close to that measured for the adsorption on the unmodified ACF (~96%). These results clearly demonstrate the significant effect of the TiO2 immobilization method and the coating/deposition time on process performance. The electrophoretic method proves to be the most effective, followed by the dip-coating method with TTIP as the titanium precursor. Regarding the coating time, different effects were observed between the two methods; i.e., an increase in contact time was beneficial only for the synthesis of ACF-TTIP composites. The improved performance of the composite ACF/TiO2 adsorbents is linked with their threedimensional structure, which favors both the binding of phenol molecules on the surface of activated carbon fibers and their degradation by the strong hydroxyl radicals ( • OH) generated by the diffused ultraviolet radiation on the immobilized photocatalyst (TiO2). The radicals can interact both with the adsorbed phenol molecules and those dissolved molecules close to the surface of the photocatalyst. Other potential enhancement mechanisms may be involved in the separation of photo-generated electron-hole pairs and in reducing the recombination by the formation of interfacial heterojunction

Phenol Photocatalytic Degradation by ACF/TiO 2 Composites
The results of the photocatalytic degradation tests with the ACF/TiO 2 composites are summarized in Figure 4. In the same Figure, the results are compared with those corresponding to phenol adsorption on unmodified ACF and ACF/TiO 2 composites. A reference photolysis test in the absence of carbon fibers was also performed, showing the limited effect of UV-A irradiation on phenol degradation (14.3% removal). This was expected, since the blacklight lamp used emits at 365 nm, which is far from the maximum UV absorption of phenol (270 nm). In comparison to the simple adsorption tests, the respective photocatalytic experiments resulted in improved removal efficiencies for most of the composite ACF/TiO 2 adsorbents tested due to the combined effects of adsorption and photocatalysis. Specifically, seven out of the nine composites exhibited significantly improved performance, which varied from approximately 85% to 93%. A decline in phenol removal efficiency was observed only in the case of the ACF-TTIP and ACF-P25 samples, prepared at low deposition times (10 s), for which phenol removal decreased by 18.9% and 42.2%, respectively. In contrast, the lower deposition time (30 s) in the case of the ACF-EDP composite led to the best results, with the percentage removal of phenol (~93%) being close to that measured for the adsorption on the unmodified ACF (~96%). These results clearly demonstrate the significant effect of the TiO 2 immobilization method and the coating/deposition time on process performance. The electrophoretic method proves to be the most effective, followed by the dip-coating method with TTIP as the titanium precursor. Regarding the Water 2019, 11, 1794 9 of 13 coating time, different effects were observed between the two methods; i.e., an increase in contact time was beneficial only for the synthesis of ACF-TTIP composites.
Water 2019, 11, x FOR PEER REVIEW 9 of 13 [31]. Such mechanisms seem to be more potent in the case of the ACF-EDP composites due to their larger active surface area (Table 3) and the uniform distribution of small spherical TiO2 nanoparticles on the fiber surface (Figure 1d). These attributes likely favor the adsorption of phenol molecules and the activation of photocatalysis to a greater extent in comparison to the ACF-TTIP and ACF-P25 composites. The activation of TiO2 photocatalysts can also be explained through the UV spectra (190 to 390 nm) of samples collected during the experiments. According to Figure 5, significant variations in UV absorbance were observed throughout the wavelength range with treatment time. Specifically, a smooth sharp decay and shifting of the first characteristic peak at 205 nm to lower wavelengths are observed. Moreover, with increasing photocatalysis time, the main peak at 270 nm tends to decay, exhibiting several small peaks of gradually reduced absorbance. These trends are more pronounced in the case of the ACF-TTIP and ACF-EDP composites, thus providing additional support to the above findings.
The combined effect of adsorption and photocatalysis on phenol removal is also reflected in the kinetic analysis of the experimental data (Table 4). In particular, the rate of phenol removal may follow multiple-order kinetics. In the case of ACF-TTIP samples, phenol removal can be satisfactorily described by first-order kinetic equations, while the results obtained with the ACF-P25 and ACF-EDP samples are best fitted by second-order kinetic equations. Among the nine ACF/TiO2 composite adsorbents, ACF-P25-30 and ACF-EDP-30 exhibit the greatest k1 and k2 kinetic constants, respectively. Variations in phenol removal rate are apparently associated with the differences in the adsorption and catalytic activity of each sample. Specifically, the exponential reduction of phenol (second-order kinetics) can be attributed mainly to the hydrophobic interactions of ACF fibers with phenol molecules, whereas the direct dependence of phenol removal on the amount of TiO2 immobilized on the carbon fiber (ACF-TTIP) is linked with first-order kinetics. These findings lead to the conclusion that the mechanism of phenol likely consists of two steps that are dynamic and dependent on the load and the morphology of the TiO2 nanostructures on the fiber surface: (1) adsorption of phenol molecules on the surface of ACF/TiO2; (2) surface photodegradation of phenol. The desorption of degradation products from the surface of the composite adsorbent is also possible; however, this is unclear in the present data, as the UV spectra ( Figure 5) show an almost total disappearance of aromaticity from the treated solution (zero absorbance at wavelengths 250-280 nm).  The improved performance of the composite ACF/TiO 2 adsorbents is linked with their three-dimensional structure, which favors both the binding of phenol molecules on the surface of activated carbon fibers and their degradation by the strong hydroxyl radicals ( • OH) generated by the diffused ultraviolet radiation on the immobilized photocatalyst (TiO 2 ). The radicals can interact both with the adsorbed phenol molecules and those dissolved molecules close to the surface of the photocatalyst. Other potential enhancement mechanisms may be involved in the separation of photo-generated electron-hole pairs and in reducing the recombination by the formation of interfacial heterojunction [31]. Such mechanisms seem to be more potent in the case of the ACF-EDP composites due to their larger active surface area (Table 3) and the uniform distribution of small spherical TiO 2 nanoparticles on the fiber surface ( Figure 1d). These attributes likely favor the adsorption of phenol molecules and the activation of photocatalysis to a greater extent in comparison to the ACF-TTIP and ACF-P25 composites.
The activation of TiO 2 photocatalysts can also be explained through the UV spectra (190 to 390 nm) of samples collected during the experiments. According to Figure 5, significant variations in UV absorbance were observed throughout the wavelength range with treatment time. Specifically, a smooth sharp decay and shifting of the first characteristic peak at 205 nm to lower wavelengths are observed. Moreover, with increasing photocatalysis time, the main peak at 270 nm tends to decay, exhibiting several small peaks of gradually reduced absorbance. These trends are more pronounced in the case of the ACF-TTIP and ACF-EDP composites, thus providing additional support to the above findings.
The combined effect of adsorption and photocatalysis on phenol removal is also reflected in the kinetic analysis of the experimental data (Table 4). In particular, the rate of phenol removal may follow multiple-order kinetics. In the case of ACF-TTIP samples, phenol removal can be satisfactorily described by first-order kinetic equations, while the results obtained with the ACF-P25 and ACF-EDP samples are best fitted by second-order kinetic equations. Among the nine ACF/TiO 2 composite adsorbents, ACF-P25-30 and ACF-EDP-30 exhibit the greatest k 1 and k 2 kinetic constants, respectively. Variations in phenol removal rate are apparently associated with the differences in the adsorption and catalytic activity of each sample. Specifically, the exponential reduction of phenol (second-order kinetics) can be attributed mainly to the hydrophobic interactions of ACF fibers with phenol molecules, whereas the direct dependence of phenol removal on the amount of TiO 2 immobilized on the carbon fiber (ACF-TTIP) is linked with first-order kinetics. These findings lead to the conclusion that the mechanism of phenol likely consists of two steps that are dynamic and dependent on the load and the morphology of the TiO 2 nanostructures on the fiber surface: (1) adsorption of phenol molecules on the surface of ACF/TiO 2 ; (2) surface photodegradation of phenol. The desorption of degradation products from the surface of the composite adsorbent is also possible; however, this is unclear in the present data, as the UV spectra ( Figure 5) show an almost total disappearance of aromaticity from the treated solution (zero absorbance at wavelengths 250-280 nm).  The stability of the photocatalytic performance was examined in the case of the composite adsorbent that presented the most promising results (ACF-EDP-30). Table 5 shows the results of three cycles of phenol treatment by the same ACF-EDP-30 specimen. Each cycle consisted of two steps: (1)  The stability of the photocatalytic performance was examined in the case of the composite adsorbent that presented the most promising results (ACF-EDP-30). Table 5 shows the results of three cycles of phenol treatment by the same ACF-EDP-30 specimen. Each cycle consisted of two steps: (1) photocatalytic treatment of a 10 mg L −1 phenol solution for 120 min; (2) regeneration of the composite adsorbent by applying UV irradiation for 120 min in a Milli-Q water solution. Results confirmed the successful regeneration of the composite photocatalyst and its steady performance, thus justifying the expectation of a prolonged service life of the synthesized ACF/TiO 2 composite by the electrophoretic method, which is a key issue for practical applications.

Conclusions
This study demonstrates the attributes of ACF/TiO 2 composite adsorbents, prepared by an electrophoretic coating method, for the removal of organic pollutants from water. SEM and BET analyses showed that TiO 2 coating on the ACF/TiO 2 composite adsorbents may negatively affect the adsorption capacity of the original ACFs, but it can still play an important role in the treatment of water by combined adsorption and photocatalysis. Experiments with phenol as a model organic pollutant shows that the removal efficiency is inversely proportional to the time of TiO 2 electrodeposition. The application of short electrophoresis times seems to favor the uniform distribution of TiO 2 nanostructures on the surface of the activated carbon fibers, which facilitates the rapid removal of phenol by a combined adsorption-photocatalysis mechanism. This behavior, along with the stable performance of the composite adsorbent prepared with electrophoresis in multiple cycles, provides evidence of the great potential of such materials for practical photocatalytic process applications. Ongoing research and development is aimed at material optimization and testing of the optimum photocatalytic adsorbent as anodic electrode in photo-electro-catalytic oxidation studies.
Funding: This research has been co-financed by the European Regional Development Fund of the European Union and Greek national funds through the Operational Program Competitiveness, Entrepreneurship and Innovation, under the call RESEARCH-CREATE-INNOVATE (project code: T1EDK-04456).