Journal of Membrane Science & Technology

Water, this essential element for life, is abundant on earth (estimated volume of about 1.4 × 109 km³). However, 97.5% is salt water. Of the remaining 2.5% that is fresh water, 70% is frozen in the polar icecaps; the rest is mainly present as soil moisture or in inaccessible subterranean aquifers. Only less than 1% of the world’s fresh water resources are readily available for human use; but this resource is very unevenly distributed [1]. Besides, the available fresh water is not always clean as it may contain natural toxic pollutants (heavy metals metal-complex dyes, bacteria). Excessive release of heavy metals into the environment due to industrialization and urbanization poses great problem worldwide. Metals such as cadmium, cooper, lead, are contained in wastewater derived from electroplating, mining, batteries, plastic and paint. Unlike organic pollutants, heavy metals cations do not degrade into harmless end products. Due to their high toxicity, they can cause many health disorders (to the central nervous system, kidneys and reproductive system) [2].

Besides, in many countries the significant numbers of textile industries are the main sources of fresh water pollution. The wastewaters from these industries are loaded with a considerable amount of dyes and surfactants. A variety of dyes are highly toxic for animals and humans, and affect water transparency reducing light penetration and gas solubility in water [3,4], thus disturbing the structure of the ecosystem. These dyes are also highly soluble, and resistant to degradation by organisms. Consequently, their removal from wastewater remains a difficult, but necessary, task. Treatments such as chemical oxidation, precipitation and coagulation of the pollutants, reverse osmosis are less effective at commercial scale due to the high cost and complexity [4,5] than adsorption process. Advanced Oxidation Processes (AOP) is modern chemical methods and effective for the treatment of water containing non-biodegradable/toxic substances and for the decontamination of drinking water [6-10].

Among the advanced oxidation processes (AOPs), photocatalysis is recognized for the ability to mineralize a wide range of organic compounds as it involves the generation of highly reactive radical species, predominantly the hydroxyl radical (HO•) which is a powerful oxidative agent, active for degrading recalcitrant high molecules of dyes. In this purpose, many photocatalysts such as semiconductors: TiO₂, SnO₂, WO₃ [11], coupled semiconductors: TiO₂/SnO₂, TiO₂/ZnO, TiO₂/WO₃ [12] were reported.

Heterogeneous photocatalysis based on reactions onto the surface of wide band gap semiconductor (TiO₂) irradiated with solar or artificial light, are of interest because of their ability to mineralize organic pollutants. For simultaneous removal of heavy metals and dyes from the waste waters, several researchers have coated photocatalysts (TiO₂) onto a variety of surfaces like, modified fly ash [13], diatomite, bentonite and clay [14]. The TiO₂ particles immobilized on adsorbent support can be more easily filtered, this is necessary for industrial applications. Thus, a new material, of zeolite-type, was developed in environmentally friendly conditions, as substrate in adsorption and as heterogeneous photocatalyst.

This paper presents the obtained results obtained the new synthetized composite (clay- TiO₂) via mild hydrothermal synthesis, from Degussa P25 and clay of north of Ivory Coast in alkaline media, aimed to be used as substrate in the advanced treatment of wastewater loaded with dye (methylene blue) and heavy metal (cadmium).

Raw material

The clay materials used in this work were collected from two different regions of north of Ivory Coast, namely: Katiola and Fronan. The clay was named according to their color in French (Blanc (B), Vert (V), and Rouge (R)).

Reagents

All the reagents were used as received without further treatment. The TiO₂ used was from Degussa (Degussa P25, 80% anatase and 20% rutile; specific surface area 50 m²g^-1 and a mean particle size of 30 nm), Cadmium chloride hemi(pentahydrate) (CdCl₂•2.5H₂O, <98% purity) from ScharlauChemie S.A, methylene blue from Fluka (C₁₆H₁₈N₃S), and sodium hydroxide (NaOH, ≥ 99%) from Fulka.

Material substrate preparation

Before being used, the clay materials (B, V and R) were washed with ultra-pure water under mechanical stirring (100 rpm, Nahita GJ-1 stirrer) at room temperature (22 ± 1°C) for 24 h, in order to remove the unwanted soluble compounds. The ratio between clay materials and ultra-pure water ratio was 1:10 (g: mL). Afterwards the suspension was filtered and dried at 105-115°C. The washed and dried clay materials were mechanically sieved (Analysette 3 Spartan) and the 40 μm fraction was selected for experiments. During the hydrothermal process the washed clay materials was treated with NaOH 2N solution. The modified clay materials were obtained under stirring in autoclave at 100°C and 5 atm. After the reactions were completed, the suspended matter was washed with ultra-pure water until constant pH (pH=9.9); afterwards it is filtered and dried at 105-115°C overnight. The modified clay materials obtained were denoted BW-NaOH_2N, VW-NaOH_2N and RW-NaOH_2N.

The composite substrate of clay and TiO₂ (B-TiO₂, V-TiO₂ and R-TiO₂) was obtained mixing 30 g BW-NaOH_2N (or VW-NaOH_2N or RW-NaOH_2N) with 30 g Degussa P25 with 36 g NaOH (mass ratio 1:1:1.2). Hydrothermal synthesis parameters were: T=100 °C, P=5 atm. during 24h.

Characterization of the material substrate

The crystalline structure of the composite substrate was investigated by XRD (Bruker D8 Discover Diffractometer), over the 2θ range 10- 70°. Morphology studies (roughness and macro pore size dis-tribution) were done using AFM (Ntegra Spectra, NT-MDT modelBL222RNTE); images were taken in semi-contact mode with golden silicon cantilever (NCSG10), with constant force (0.15 N/m), having the tip radius of 10 nm. Scanning was conducted on three or more different places with a certain area of 5 × 5 μm for each position, randomly chosen at a scanning grate of 1 Hz. Further surface investigations were done using scanning electron microscopy (SEM, S-3400N-Hitachi) at an accelerating voltage of 20 KV. Surface composition was evaluated using energy dispersive X-ray spectroscopy (EDS, Thermo Scientific Ultra Dry). Surface characterization was completed by microporosity analysis (AFM) and BET surface measurements (Autosorb-IQ-MP, Quantachrome Instruments). The information related to the functional groups on the surface was provided by FTIR (PerkinElmer BX II 75548).

Adsorption and photocatalytic experiments

Batch adsorption tests were done in open cylindrical flasks, at room temperature (22 ± 1°C) by mixing 0.1g of substrate with 50 mL solution (MB; MB + Cd²⁺) under mechanical stirring. The adsorption duration of MB, Cd²⁺ cations on the composite substrate (B-TiO₂, V-TiO₂ and R-TiO₂): was varied up to 240 min, followed by filtration and filtrate analysis. The initial concentrations of the pollutant systems were 0.01N CdCl2 prepared in bi-distilled water using CdCl₂•2.5H₂O (ScharlauChemie S.A., <98%purity) and 0.03125 mM, methylene blue (C₁₆H₁₈N₃S; Fluka AG, reagent grade). The adsorption mechanism and the kinetic data were evaluated. These experiments are denoted with (A).

- Photodegradation investigations were done on B-TiO₂, V-TiO₂ and R-TiO₂ suspensions with the same composition as in the adsorption studies, with and without Fenton reactive and hydrogen peroxide (30%), under UV irradiation, and the results are denoted with (F). The photocatalytic reactor equipped with three F18W/T8 black tubes (Philips), emitting UV-A light in the region of 340–400 nm and λ_max(emission) = 365 nm.

The mean value of the radiation flux intensity, reaching the middle of the reacting suspension was 3 Lx (Mavolux5032B/USM) and the irradiance is 846 W m⁻². During the adsorption studies the radiation flux intensity was in the range of 0.7–1.6 Lx, with an average irradiation of 215 W m⁻².

During the kinetic studies, aliquots were taken at fixed moments (up to 360 min) when stirring was briefly interrupted and, after filtration on 0.45 mm filter, the supernatant was analyses. Preliminary experiments proved that dyes losses due to adsorption on the beaker walls or on the filtering paper were negligible.

The initial and residual metal concentration in the aqueous solution was analyzed by AAS (Analytic Jena, ZEEnit 700, at λ_Cd = 228.8 nm) and the MB was analyzed by UV–vis spectrometry (Perkin Elmer Lambda 25), on the calibration curve registered at λ= 664 nm, respectively.

The adsorption/photodegradation efficiency for the cadmium cations and dyes was evaluated using Equation (1):

(1)

Where: c₀ represents the initial concentration of the pollutants and c_t represents the concentration of the pollutants at time t.

The characterization of composites material

The XRD analyses (Figure 1) show more complex composition for the new composites B-TiO₂, V-TiO₂ and R-TiO₂, the structures are composed of the anatase phase mostly (Tables 1-3).

Figure 1: XRD graphs: (A) BW; (B) B-TiO₂; (C) VW; (D)V-TiO₂, (E) RW and (F ) R-TiO₂.

Table 1: The composition of the crystalline phases, B-W and B-TiO₂.

Table 2: The composition of the crystalline phases, V-W and V-TiO₂.

Table 3: The composition of the crystalline phases, R-W and R-TiO₂.

The crystalline structures of washed clay (B-W, V-W and R-W) and of the composites (B-TiO₂, V-TiO₂ and R-TiO₂) were comparatively investigated. The crystallite sizes were calculated using the Scherrer formula, Equation (2) [15].

(2)

Where: τ is the size of crystallites, K is the shape factor with a typical value 0.94, λ is the X-ray wavelength (1.5406 Å), β is the line broadening at half the maximum intensity (of a peak), and θ is diffraction angle.

The XRD spectra (Figure 1) display overall crystalline percentages of 73.9 % in B-W, 82.1 % in V-W and 63.9 % in R-W. The major crystalline components of washed clay are: SiO₂ (as αSiO₂ quartz, quartz syn, quartz low, cubic and orthorhombic SiO₂). The XRD data show that the new substrates, B-TiO₂ (for example), have well embedded the anatase phase (anatase syn with 237.7 Å crystallite size) and the rutile - TiO₂ (with 120.7 Å crystallite size), one may conclude that the hydrothermal treatment supported a TiO₂ recrystallization process on the microsized B-W, extending the crystallite dimensions. B- TiO₂ c ontains TiO₂ polymorphs. New compounds are identified on the XRD spectra of B-TiO₂ substrate (kyanite (Al₂SiO₅)), confirming that chemical restructuring occurs within B-W when hydrothermally processed. The crystalline modifications are accompanied by a significant increase in the BET surface, from 19.35 m²/g in B-W to 43.37 m²/g in B–TiO₂; from 25.38 m2/g in V-W to 44.29 m2/g in V-TiO₂ and 15.26 m2/g R-W to 32.88 m2/g in R–TiO₂.

Information on the new substrates (morphology/topography) was obtained from AFM and SEM micrographs (Figures 2-4). The AFM images (only for BW and B–TiO₂.) and pore size distributions show different granular shapes. The highest roughness value (120.712 nm) corresponds to the new composites which has more aggregates with different, almost round and stable shapes (Figure 4). The composite has a large specific surface (larger than Degussa P25, 50.3 nm), with small micro-pores (large enough to accommodate the dyes). The re-organizing process is confirmed by the AFM pictures, outlining significant differences between the randomly structured surfaces of clay washed (Figure 2) and the regular aggregates with droplet shape (Figure 3) assembled in rough structure that leave open macro-pores on the clay: TiO₂ surface.

Figure 2: AFM topography of BW; Average roughness: 46.905 nm.

Figure 3: AFM topography of B-TiO₂; Average roughness: 120.712 nm.

Figure 4: Scanning electron microscopic images of (A) B-TiO₂; (B) V-TiO₂ and (C) R-TiO₂.

The predominant polar/ionic surface energy corresponds, in an oxide material in alkaline pH (larger than the point of zero charge), to a negatively charged surface and shows that the mild hydrothermal process increased the surface polarity/ionic degree. This combination of increased specific surface, homogeneity and roughness, and negative charges supports the use of the clay TiO₂ composite as substrate in adsorption processes of neutral or cationic species [16]. Additional surface investigations were done and the SEM images are presented in Figure 4. The SEM images confirm that the clay grains are cracked and micro-restructuring occur with significant modification of the surface aspect, as results of dissolution/re-precipitation reactions and TiO₂ particles are embedded on surface of the clay.

The FT-IR spectra analyses of the composites B-TiO₂, V-TiO₂ and R-TiO₂ synthesized are displayed in Figure 5. The spectra analysis conducted to investigate the vibration frequency changes of the functional groups in the adsorbents materials (functional groups), indicating the complex nature of the adsorbent as shown in Figure 5 and in Table 4.

Figure 5: FT-IR spectra analyses of B-TiO₂, V-TiO₂ and R-TiO₂.

Table 4: The vibration frequency of the functional groups in the adsorbents materials.

Adsorption and photocatalytic processes on composites

The active species generated by photo-irradiation can attack the pollutant, if this is in the very close vicinity of the substrate. Thus, adsorbed pollutant species will be removed efficiently by photocatalytic processes. Studies are already reported linking the pollutants’ structures, adsorption and the photocatalytic efficiency [17,18]. The most active photocatalytic component of the composite is TiO₂ because the TiO₂ polymorphs have band gaps of 3.0 eV (rutile), 3.2 eV (anatase) thus are active under UV radiation with wavelengths lower than 413 nm for rutile, 387 nm for anatase. As the UV wavelength used in our experiments was 365 nm, we may conclude that only anatase and rutile are actually activated under irradiation and can exhibit the well-known coupling effect [19].

The solutions containing only MB (0.03125 mM) are well treated using all materials (Figure 5). But, the solutions containing MB and Cd²⁺, efficiencies of MB decrease because of an affinity orders of the species, towards the substrate: Cd²⁺ > MB [20]. Cd²⁺ is firstly adsorbed on the substrates and then MB. The chemical structure of MB is shown in Figure 6. The photocatalysis process efficiency after 360 min on materials are presented in Table 5.

Figure 6: MB and Cd²⁺ photodegradation efficiency vs. time on B-TiO₂ (A), V-TiO₂ (B) and R-TiO₂ (C).

Table 5: Photocatalysis process efficiency [%] on substrates with TiO₂ embedded.

The adsorption efficiency after 30 min, for MB and Cd²⁺ removal on all composites are presented on Table 6. Better results were obtained using R-TiO₂ material. The XRD graph of R-TiO₂ shows the presence of new compounds (montmorillonite) that have good adsorption proprieties [14].

Table 6: Adsorption process efficiency [%] on substrates.

The adsorption mechanism

In system containing two or more pollutants and the substrate, several adsorption processes develop/occur [13-16]:

a) The Cd²⁺ can be absorbed by Ti–OH of the layer, but with lower efficiency (Equation 3):

(3)

b) In the multicomponent solutions, Cd²⁺ + MB interaction can be developed, further influencing the adsorption rate and its mechanisms. Many organic substances like aromatic compounds are attached to the modified clay (B-TiO₂, V-TiO₂ and R-TiO₂) surface by hydrogen bonding, but stronger interactions with formation of new bonds can be observed for other molecules.

The possible reactions are proposed in Figure 7 involving the lone pair of electrons from the pyridine nitrogen atom or chlorine in MB molecules. Stereochemistry complexes with cadmium are determined only by the ionic volume strength of the strong electrostatic and covalent bonds. The volume effect makes the Cd²⁺ to be more apt to forms tetraor hexa-coordinated complexes with distorted octahedral structure Figure 8. These interactions/bonds can be correlated with FTIR spectra Figure 9. The peak at 605.64cm^-1 of MB disappears, there. The Si-O vibration band is at 605.64 cm^-1and disappears with Cd²⁺ in solution.

Figure 7: The chemical structure of Methylene Blue.

Figure 8: The interaction of Cd2+ with Methylene Blue molecule.

Figure 9: FT-IR spectra of the MB and MB+Cd²⁺.

The photo-fenton mechanism

Under UV radiation the electron–hole pair is formed. The most common mechanisms involve the holes for hydroxyl radical production. In alkaline media several other reactions involving the O₂/HO−, O₂/ HOO− and the O₂/H₂O₂ couples are possible [21,22]. Another undesired process is the electron–hole recombination. The photocatalytic efficiency strongly depends on the system’s ability to limit this process. The hydrogen peroxide addition is expected to support electron trapping. In this work, in presence of 0.1 g composite + 10 μL H₂O₂ + 50 mL pollutant solution + 20 μL Fe²⁺ (Fenton’s reagent (Fe²⁺/H₂O₂), there is an are increase in the efficiency of organic pollutants degradation (Figure 10). As exhibited in this Figure, a mixture of H₂O₂ and Fe²⁺ in medium with the range of pH 4-5 has good oxidizing properties of the organic pollutants developing the hydroxyl radicals [23]. The classical mechanism is a simple redox reaction in which Fe²⁺ ions are oxidized to Fe³⁺ ions and the H₂O₂ is reduced to hydroxide ion and hydroxide radical (Equation 4).

(4)

Figure 10: The influence of H₂O₂ + Fe²⁺ in photocatalytic process on: (a) B-TiO₂, (b) V-TiO₂, (c) R-TiO₂.

The Fe³⁺ ion produced in reaction (4) can be reduced back to Fe²⁺ ion by a second molecule of hydrogen peroxide (equation 5).

(5)

The stoichiometry of Fenton degradation reactions is complex. In addition to Fe²⁺ / Fe³⁺ and hydrogen peroxide, can involve the participation of the hydroperoxyl radical, HOO• iron (IV) or ferryl, FeO²⁺, dissolved molecular oxygen, organic hydroperoxides, and other intermediates formed during the degradation. Under irradiation of Fenton systems with UV light strongly accelerated the rate of degradation of a variety of organic pollutants for example MB dye. This behavior upon irradiation is due principally to the photochemical reduction of Fe³⁺→ Fe²⁺, for which the net reaction can be written as (Equation 6): [24]

(6)

The Cd²⁺ can be adsorbed and is activated by H₂O₂–corroborated with a partial dehydration (thus are higher mobility); the significant increase in the removal efficiency of MB in this condition (Equation 7):

(7)

The photo-Fenton process efficiency, after 360 min, on all materials is presented in Table 7.

Table 7: Photo-Fenton process efficiency [%] on materials.

The kinetic studies: kinetic parameters of pollutants for adsorption and photodegradation

The kinetic parameters are best fitted by the pseudo-second order kinetic model [25], with the linear form given by Equation (8):

(8)

Where:

k₂ is the pseudo second-order rate constant (g mg⁻¹min⁻¹) and can be evaluated from the slope of the plot. Based on Equation (8) the kinetic parameters were calculated and are presented in Table 8 (A, B and C).

Table 8: Kinetic parameters of pollutants ((A): Cd²⁺ (MB+Cd²⁺); (B): MB; (C): MB (MB+Cd²⁺)) removal in photocatalysis; photo Fenton processes and adsorption.

These data show that in the system containing multi pollutant here is the high cadmium mobility comparative to methylene blue, in adsorption and photodegradation process. This is the confirmation that cadmium is firstly adsorbed on the substrate and then methylene blue. The adsorption process of multi pollutant in the experimental condition is significantly controlled by cadmium mobility.

In photo-Fenton process, the Cd²⁺ cations removal rate is higher than photocatalysis process; the same applied to other pollutants, namely MB. The HO- ions (chemo) sorbs and negatively charge TiO₂, resulting in an activated surface with increased affinity for Cd²⁺ [26,27].

A new substrate was obtained in a hydrothermal process, starting with clay from regions of north of Ivory Coast, namely: Katiola and Fronan and TiO₂ Degussa P25 and was tested for simultaneous removal of heavy metal cations and dyes, in a single step process, involving adsorption and photocatalysis.

The clay–TiO₂ structural and morphology analysis showed that the substrate has a high crystallinity degree and surface, homogeneity and good roughness for adsorption of MB and of cadmium cations.

The adsorption efficiency of MB (0.03125 mM) and Cd²⁺ (0.01 N) on R-TiO₂ is 91.19 % and 27.05 %, respectively. For photocatalysis, efficiency increase for the same pollutants on R-TiO₂ from 91.19 % to 97.48 % and from 27.05 % to 44.37 %. For photo-Fenton process, there are increases the efficiency of organic pollutants degradation and heavy metal removal than photocatalysis process. The High photo-Fenton efficiency resulted of hydroxides radicals produced in the medium by mixture of H₂O₂ and Fe²⁺.

The kinetic studies show that the substrate has a good adsorption capacity and fast adsorption processes which is, mainly based on the electrostatic attractions between the substrate and the pollutant species. The new substrate has the grains in the micrometric range, representing thus a promising alternative to Degussa P25 slurries. This is a simpler and cost-effective method to recuperate/recycle the substrate in industrial wastewater treatment processes. High adsorption efficiencies are registered for MB, and simultaneous removal of the MB and Cd2+ is possible, on all materials B-TiO₂, V-TiO₂, R-TiO₂.

This work was financed by Agence Universitaire de la Francophonie (Eugen Ionescu scholarship).