Porous Layered Double Hydroxide / TiO 2 Photocatalysts for the Photocatalytic Degradation of Orange II

: Layered Double Hydroxide (LDH) / TiO 2 nanocomposites with photocatalytic properties were synthesized by both impregnation and the direct coprecipitation of LDH matrices using a colloidal suspension of TiO 2 nanoparticles. While the two methods led to an e ﬃ cient TiO 2 nanoparticle immobilization, the direct coprecipitation allowed us to tune the amount of immobilized TiO 2 within the materials. The LDH / TiO 2 nanocomposites obtained were deeply characterized by chemical analysis (ICP-AES), Powder X-ray di ﬀ raction (XRD), Fourier Transformed Infra-Red (FTIR), Thermogravimetric analysis (TGA), and High-Resolution Transmission Electron Microscopy (HRTEM). Clearly, the immobilization of TiO 2 by direct coprecipitation promoted a modiﬁcation of the textural properties and a net increase in the surface area. The crystallized TiO 2 nanoparticles can be distinctly visualized by HRTEM at the surface of the layered material. Several parameters, such as the nature of the chemical composition of LDH (ZnAl and MgAl), the method of immobilization and the amount of TiO 2 , were shown to play a crucial role in the physicochemical and photocatalytic properties of the nanocomposites. The photocatalytic e ﬃ ciency of the di ﬀ erent LDH / TiO 2 nanocomposites was investigated using the photodegradation of a model pollutant, the Orange II (OII), and was compared to a pure TiO 2 colloidal solution. The degradation tests revealed that the nanocomposite obtained from MgAl LDH at a low MgAl LDH / TiO 2 ratio was the most e ﬃ cient for the photodegradation of OII leading to complete mineralization in 48 h.


Introduction
Dyes, as all other organic compounds, are introduced into water bodies through waste from industrial processes such as pharmaceutical, dyes and textiles, petrochemicals, and pesticide factories [1,2]. The presence of these compounds in water bodies, such as rivers, lakes, and oceans, results in huge environmental pollution by industrial waste. Water bodies polluted by toxic and carcinogenic organic compounds pose a threat to aquatic organisms, plants, animals, and humans [3][4][5][6]. Besides, the accumulation of these dyes in aquatic species can lead to animal and human toxicity through the food chain [7,8]. In some cases, with these non-biodegradable and often highly oxidizable compounds, the pH of the medium may decrease and thus cause a considerable increase in toxicity for the ecosystems. For the organisms living in an aquatic medium, animals are more sensitive to toxicity than plants. Often, even at very low concentrations, the specific odors of these chemical pollutants are emitted into the water and even contaminate the flesh of fish [9,10]. These polluted waters can, in turn, contaminate the groundwater reserve of natural drinking In this work, different approaches were investigated to synthesize porous TiO 2 @LDH photocatalytic nanocomposites, in which the immobilized TiO 2 photocatalytic properties are combined with the LDH adsorption, with the aim of establishing contact between the reactive sites (h VB + ; e CB − ) and pollutant required in addition to • OH and other reactive species. The effect of the TiO 2 immobilization technique, the LDH chemical composition, and TiO 2 aging on the nanocomposite properties were studied in detail with the aim of shedding light on the relationship between the material properties and the photocatalytic efficiency recorded.

Materials Preparation
Orange II ≥ 85% (OII); HClO 4 70%; KNO 3 99%, Na 2 CO 3 99%, and NaOH 97% were purchased from Merck. KOH; Zn(NO 3 3 , and TiCl 4 99% were purchased from Acros Organics or Sigma-Aldrich. These products have been used without further purification. Deionized water obtained from a water purification system showing resistivity of 18.0 MΩ cm was systematically used with no additional degassing step. N 2 gas (99%) was used without further drying or deoxygenation as a protective atmosphere.
To prepare colloidal TiO 2 nanoparticles, 3.5 mL of pure TiCl 4 was added dropwise into 900 mL of deionized water, stirred vigorously and cooled to 1 • C as previously reported [52]. The solution was then further stirred for 30 min and finally dialyzed with Spectrapor membrane up to pH = 2.50. When necessary, deionized water was added to obtain a total volume of the solution of 960 mL, which corresponds to a concentration of Q-TiO 2 equal to 33.3 mM (2.66 g L −1 ). The granulometric analysis by dynamic light scattering reveals that synthesized Q-TiO2 particles were of 5 nm in diameter, as shown in Figure 1B. Aged TiO 2 was obtained by letting the colloidal solution for three months at 4 • C. Prior to the immobilization step, a basic colloidal dispersion of TiO 2 was prepared by quickly adding a 2 M KOH solution to the acidic colloidal TiO 2 solution under ultrasound. The clear dispersion thus obtained has a pH = 9.30.
ChemEngineering 2020, 4, x FOR PEER REVIEW 3 of 16 photocatalyst's recovery by filtration; (3) limit the TiO2 nanoparticle aggregation through a homogeneous distribution on the two-dimensional materials. In this work, different approaches were investigated to synthesize porous TiO2@LDH photocatalytic nanocomposites, in which the immobilized TiO2 photocatalytic properties are combined with the LDH adsorption, with the aim of establishing contact between the reactive sites (hVB + ; eCB − ) and pollutant required in addition to • OH and other reactive species. The effect of the TiO2 immobilization technique, the LDH chemical composition, and TiO2 aging on the nanocomposite properties were studied in detail with the aim of shedding light on the relationship between the material properties and the photocatalytic efficiency recorded.
To prepare colloidal TiO2 nanoparticles, 3.5 mL of pure TiCl4 was added dropwise into 900 mL of deionized water, stirred vigorously and cooled to 1 °C as previously reported [52]. The solution was then further stirred for 30 min and finally dialyzed with Spectrapor membrane up to pH = 2.50. When necessary, deionized water was added to obtain a total volume of the solution of 960 mL, which corresponds to a concentration of Q-TiO2 equal to 33.3 mM (2.66 g L −1 ). The granulometric analysis by dynamic light scattering reveals that synthesized Q-TiO2 particles were of 5 nm in diameter, as shown in Figure 1B. Aged TiO2 was obtained by letting the colloidal solution for three months at 4 °C. Prior to the immobilization step, a basic colloidal dispersion of TiO2 was prepared by quickly adding a 2 M KOH solution to the acidic colloidal TiO2 solution under ultrasound. The clear dispersion thus obtained has a pH = 9.30. [Zn2Al] and [Mg2Al] phases were synthesized by coprecipitation at constant pH = 8.00 and pH = 10.00, respectively [34]. Typically, a 0.2 M metal salt aqueous solution (M II /M III = 2) was slowly added in a reactor containing either potassium nitrate (1 M) or sodium carbonate (0.5 M), according to the [Zn 2 Al] and [Mg 2 Al] phases were synthesized by coprecipitation at constant pH = 8.00 and pH = 10.00, respectively [34]. Typically, a 0.2 M metal salt aqueous solution (M II /M III = 2) was slowly added in a reactor containing either potassium nitrate (1 M) or sodium carbonate (0.5 M), according to the anion to be intercalated. For the nitrate intercalated LDH, a protective N 2 atmosphere was used. The pH of the reaction medium was kept constant by the controlled addition of a 2 M sodium hydroxide solution. After addition, the reaction medium was stirred for 24 h. The solid was then collected by centrifugation and subjected to three wash cycles. For the [Zn 2 Al-NO 3 ] sample used for TiO 2 immobilization by impregnation, one part was dried overnight at room temperature to get LDH powder (noted dry), while another part was kept dispersed in an aqueous solution to avoid the irreversible aggregation of the particles during the drying step (noted wet).
TiO 2 colloidal particles were immobilized by impregnation on [Zn 2 AlNO 3 ] by adding a suspension of 100 mL of colloidal TiO 2 (0.5 g L −1 ) in a reactor containing 0.1 g of LDH, which corresponds to a mass ratio of LDH/TiO 2 = 2. The solution was subjected to mechanical stirring for approximately 20 h under a protective N 2 atmosphere. The LDH precursors used were either in the form of dried powder or wet suspension and the resulting samples are hereafter noted [Zn 2 Al/TiO 2 ] dry and [Zn 2 Al/TiO 2 ] wet .
In an alternative strategy, TiO 2 colloidal particles were also directly immobilized during LDH coprecipitation. A similar procedure as previously described for the LDH synthesis was performed except that a basic colloidal solution of TiO 2 was initially introduced into the reactor. Typically, a fixed volume of 0.2 M metal salt solution was added simultaneously with 2 M NaOH/0.5 M Na 2 CO 3 to the reactor containing TiO 2 , to maintain the pH at 9.30 and 10.00 for [Zn 2 Al/TiO 2 ] and [Mg 2 Al/TiO 2 ], respectively. The theoretical LDH/TiO 2 ratios were fixed by controlling the volume of basic colloidal TiO 2 added ([TiO 2 ] = 2.66 g L −1 ), the metal salt volume being maintained constant. The [Zn 2 Al/TiO 2 ] coprecipitated sample was prepared for an LDH/TiO 2 ratio of 2 while four [Mg 2 Al/TiO 2 ] samples were prepared for ratios of 8, 4, 2, and 0.66. After the addition, the reaction medium was stirred for 24 h. The solid was then collected by centrifugation and subjected to three washing cycles and dried at room temperature.

Characterization
The sizes of the TiO 2 particles were measured by dynamic light scattering (DLS). These measurements were carried out on a Zetasizer Nanoseries (Malvern Instrument) on very dilute dispersion. Zeta potential (ξ) was also determined using the same equipment. The errors were estimated based on the standard deviation of three measurements and a reproducibility test. All X-ray diffraction measurements were made on a X-PERT PRO PHILIPS powder diffractometer equipped with X-Celerator detector. The diffractograms were carried out under the following operating conditions: angular domain 2θ (5 • -70 • ), angular increment 2θ (0.0836 • ), and counting integration time (130.17 s). The infrared spectroscopy measurements were obtained using a NICOLET 5700 Fourier Transform spectrometer. The measurements were carried out in attenuated total reflection (ATR). The spectra were acquired in absorbance for wavenumbers between 4000 and 400 cm −1 , a resolution of 8 cm −1 , and a number of 32 scans. The chemical compositions of the samples were determined by inductively coupled plasma atomic emission spectroscopy (ICP-AES) on a HORIBA-Jobin-Yvon ULTIMA C spectrometer. Analyses by high-resolution transmission electron microscopy (HRTEM) were carried out on a JEOL JEM 3010 microscope (200 kV, LaB6cathode) equipped with a CCD camera. The samples were dispersed in ethanol and further treated in ultrasound for 20 min. In order to carry out the analyses, a drop of the dilute suspension was placed on a holey-carbon-coated copper grid and was left to dry at room temperature. The Energy Dispersive X-ray (EDX) spectra were collected from several locations. Nitrogen adsorption-desorption experiments were performed at −196 • C with a Micromeritics ASAP 2020. The samples were pretreated at 80 • C under vacuum for 12 h. The surface areas were calculated using the Brunauer-Emmett-Teller method. Thermogravimetric analyses (TGA) were recorded on a Setaram TG-DTA 92 instrument in the temperature range of 25-1000 • C, with a heating rate of 5 • C min −1 . The measurement of the total organic carbon concentration to follow the mineralization of OII was monitored using a TOC-5050 Analyzer (Shimadzu).

Photocatalytic Activity
The irradiation was performed in a setup with three medium-pressure mercury vapor lamps (MAZDA HPW 125W) whose emission, filtered by a black globe, was mainly at 365 nm (93% light).
A water-jacketed Pyrex tube centered into the setup was used as a reactor. The amount of photocatalyst involved in the photodegradation experiment was systematically calculated to correspond to [TiO 2 ] = 0.5 g L −1 . The concentration of OII used as a model pollutant was fixed at 5.10 −5 M and the solution was continuously stirred during the experiment. The amount of OII in the solution was determined at different intervals of time by UV-vis spectroscopy using a Varian Cary 100 Scanning UV-vis spectrophotometer, by monitoring the absorbance at λ max = 484 nm and by using the Beer-Lambert law. Being an anionic dye, OII strongly adsorbs on LDH, as shown in Supplementary Materials, Figure S1. To discriminate between adsorption and photodegradation, the LDH matrices were dissolved before OII dosage by adding an acidic solution (HCl 1 M) until pH = 3.0, to induce the release of the OII adsorbed on the LDH in solution. The calibration curve and the "apparent" absorption coefficients were established in these conditions. Photodegradation experiments, involving [Zn 2 Al-NO 3 ] and [Mg 2 Al-CO 3 ] LDH precursors without TiO 2 , were performed to confirm that, in these conditions, the amount of OII photodegraded was negligible.

TiO 2 Colloidal Dispersion
Granulometric analysis of the colloidal dispersion of TiO 2 at pH 2.50 revealed that the particle size was 5 nm after synthesis, as shown in Figure 1B. A limited size increase was observed after three months of aging at 4 • C, with a particle size centered at 11 nm and a ξ of +39 mV, as previously reported [52]. However, such aging was accompanied by the enhancement of the crystallinity of the particles, evidenced on the PXRD displaying reflection lines typical of the anatase phase, as shown in Figure 1A. The width of the reflections was in good agreement with very small coherent domain sizes. To promote the use of colloidal TiO 2 for environmental applications in neutral or basic pH and to make the surface charge of the nanoparticles compatible with the LDH, a titration was carried out using concentrated KOH solution. Titanium oxides are known to be amphoteric in aqueous suspensions. The equilibrium constants determined for Degussa P25 TiO 2 are pK a1 = 4.50 and pK a2 = 8.00 [53]. These results lead to the value of the zero-charge point pH ZPC = 6.25. In good agreement, the studied TiO 2 colloidal suspension coagulates for intermediate pH values in the range 6-8 due to the neutralization of the surface charge, leading to aggregation, as shown in Figure 1B. At pH higher than 9, the charge inversion proceeded and the change in particle size was limited. A clear basic colloidal dispersion of TiO 2 was obtained at 9.30 with a particle size centered at 15 nm (ξ = −41 mV) in good agreement with TEM image, as shown in Supplementary Materials, Figure S2. As a result, in the following parts, the TiO 2 dispersion at pH 9.30 was used for the synthesis of LDH/TiO 2 assemblies.

LDH/TiO 2 Nanocomposite Preparation and Characterization
Various [LDH/TiO 2 ] composites were prepared to evaluate the influence of parameters such as the immobilization method, the LDH chemical composition, and the TiO 2 physico-chemical properties in the photocatalyst efficiency of OII degradation.
Two different strategies were tested to prepare the LDH/TiO 2 nanocomposites: the impregnation and the direct coprecipitation. The impregnation process is based on a simple contact in solution under stirring between an LDH precursor phase and colloidal TiO 2 nanoparticles. At basic pH, the TiO 2 immobilization is promoted by electrostatic interaction between LDH layers and TiO 2 nanoparticles. In a preliminary step, a [Zn 2 Al-NO 3 ] LDH phase was prepared by classical coprecipitation. The PXRD pattern displays the main characteristic reflections expected for a pure LDH phase in a rhombohedral R-3m structure, especially with the 003, 006, 120, 110 and 113, as shwon in Figure 2A. The position of the 00l reflection lines indicated an interlamellar distance equal to 0.83 nm, confirming the formation of nitrate intercalated LDH. The presence of nitrate as interlayer anion is further confirmed by FTIR and the presence of an intense band at 1403 cm −1 , as shown in Figure 2B. The band observed at 1357 cm −1 could indicate partial contamination by carbonate anions adsorbed at the LDH particle surface. The other bands observed were characteristic of the LDH matrices with the vibrations of the OH groups at 3400-3600 cm −1 , the vibrations of the H 2 O molecules in the interlayer at 3100-3200 cm −1 for the OH valencies, and 1650 cm −1 for the deformation vibrations. The bands below 850 cm −1 were attributed to the vibrations of the M-O and the deformation of the M-O-M bonds.
interlayer anion was more pronounced when an undried [Zn2Al-NO3] suspension was used, which seems to indicate that, in these conditions, anion exchange reaction is easier to carry out. When the dried powder of the precursors was involved in the process, only a partial anion exchange occurred, evidenced by the presence of two overlapped 003 diffraction lines around 11°, as shown in Figure  2A, and the nitrate bands at 1403 cm −1 which can still be observed on the FTIR spectrum, as shown in Figure 2B and in the supporting materials, Figure S3. The presence of a band at 1357 cm −1 , characteristic of carbonate anions, was attributed to the contamination of the matrix. However, the global decrease in the band at 1403-1357 cm −1 in the nanocomposites seems to indicate the replacement of nitrate and carbonate anions at the LDH particle surface. The positive surface charge compensation of the LDH layer could be mostly ensured by the negatively charged TiO2 nanoparticles. On the FTIR spectra, the bands related to the M-O and O-M-O bonds were unmodified after TiO2 impregnation, which further confirmed that the LDH layer was not affected by the immobilization process. As fresh TiO2 was used in the impregnation approach, the very small size of the colloidal For the [Zn 2 Al-NO 3 ] precursor, a zeta potential (ξ) of + 25 mV was measured in water, evidencing the positively charged LDH particle surface. The charge inversion of the TiO 2 surface by controlled basic titration led to a ξ value of -41 mV. These opposite ξ values are in favor of electrostatic interaction between the two components and the obtention of [ZnAl/TiO 2 ]. After a contact period of 24 h, a broadening and a decrease in the intensity of the diffraction peaks were observed, highlighting a decrease in the crystallinity of the materials. While the position of the reflection lines linked to the intralayer order was not modified, a shift in the 00l lines to 11.6 • (0.76 nm), which corresponds to the conversion of the initial nitrate phase into a carbonate intercalated LDH, was found, despite the nitrogen atmosphere used. We assumed that the carbonate anions could come from the TiO 2 basic dispersion, since no atmosphere control was applied during the titration. Such modification of the interlayer anion was more pronounced when an undried [Zn 2 Al-NO 3 ] suspension was used, which seems to indicate that, in these conditions, anion exchange reaction is easier to carry out. When the dried powder of the precursors was involved in the process, only a partial anion exchange occurred, evidenced by the presence of two overlapped 003 diffraction lines around 11 • , as shown in Figure 2A, and the nitrate bands at 1403 cm −1 which can still be observed on the FTIR spectrum, as shown in Figure 2B and in the supporting materials, Figure S3. The presence of a band at 1357 cm −1 , characteristic of carbonate anions, was attributed to the contamination of the matrix. However, the global decrease in the band at 1403-1357 cm −1 in the nanocomposites seems to indicate the replacement of nitrate and carbonate anions at the LDH particle surface. The positive surface charge compensation of the LDH layer could be mostly ensured by the negatively charged TiO 2 nanoparticles.
On the FTIR spectra, the bands related to the M-O and O-M-O bonds were unmodified after TiO 2 impregnation, which further confirmed that the LDH layer was not affected by the immobilization process. As fresh TiO 2 was used in the impregnation approach, the very small size of the colloidal TiO 2 particles, and therefore of the crystalline domains, as shown in Figure 1, induced the absence of TiO 2 characteristic reflections on the PXRD patterns, as shown in Figure 2A.
The TiO 2 immobilization on the [ZnAl] and [MgAl] matrices was next carried out by direct coprecipitation at fixed pH in the presence of TiO 2 dispersion. First, the volume of TiO 2 used was calculated to correspond to an LDH/TiO 2 ratio of 2, as previously used for the impregnation strategy. Figure 3A shows the PXRD patterns recorded for the [ZnAl] and [MgAl] LDH in using freshly prepared TiO 2 . Systematically, the formation of crystallized LDH in the presence of TiO 2 was evidenced by the presence of the main diffraction lines characteristic of LDH. The [ZnAl] phase showed a particularly high crystallinity, linked to the presence of the carbonate anions in the interlayered domain, which was confirmed in particular by the interlamellar distance measured at 0.76 nm on the PXRD patterns. As for the impregnation strategy, while all of the reflection lines of the LDH matrix were present, no reflection line characteristic of TiO 2 was observed on the pattern. However, when aged TiO 2 was involved in the process for the [MgAl] matrix, in addition to the LDH reflection lines, the reflections belonging to the anatase TiO 2 were also distinguished, especially the 101 line at 25.5 • , which confirmed an efficient immobilization of the TiO 2 nanoparticles within the material. Secondly, the volume of aged TiO 2 dispersion involved during the [MgAl] matrix coprecipitation was varied to prepare nanocomposites with various LDH/TiO 2 ratios, ranging from 8 to 0.66. As described above, the presence of both LDH and TiO 2 was clearly demonstrated by the observation of their characteristic diffraction lines on the patterns, as shown in Figure 3B. The relative intensities between the corresponding reflection lines of each component were strongly affected by the LDH/TiO 2 ratio. The lower ratios, which correspond to the higher TiO 2 loading, displayed, as expected, higher intensities for the TiO 2 reflection lines compared to the LDH reflection lines. On the FTIR spectra, as shown in Figure S4   TiO2 particles, and therefore of the crystalline domains, as shown in Figure 1, induced the absence of TiO2 characteristic reflections on the PXRD patterns, as shown in Figure 2A.
The TiO2 immobilization on the [ZnAl] and [MgAl] matrices was next carried out by direct coprecipitation at fixed pH in the presence of TiO2 dispersion. First, the volume of TiO2 used was calculated to correspond to an LDH/TiO2 ratio of 2, as previously used for the impregnation strategy. Figure 3A shows the PXRD patterns recorded for the [ZnAl] and [MgAl] LDH in using freshly prepared TiO2. Systematically, the formation of crystallized LDH in the presence of TiO2 was evidenced by the presence of the main diffraction lines characteristic of LDH. The [ZnAl] phase showed a particularly high crystallinity, linked to the presence of the carbonate anions in the interlayered domain, which was confirmed in particular by the interlamellar distance measured at 0.76 nm on the PXRD patterns. As for the impregnation strategy, while all of the reflection lines of the LDH matrix were present, no reflection line characteristic of TiO2 was observed on the pattern. However, when aged TiO2 was involved in the process for the [MgAl] matrix, in addition to the LDH reflection lines, the reflections belonging to the anatase TiO2 were also distinguished, especially the 101 line at 25.5°, which confirmed an efficient immobilization of the TiO2 nanoparticles within the material. Secondly, the volume of aged TiO2 dispersion involved during the [MgAl] matrix coprecipitation was varied to prepare nanocomposites with various LDH/TiO2 ratios, ranging from 8 to 0.66. As described above, the presence of both LDH and TiO2 was clearly demonstrated by the observation of their characteristic diffraction lines on the patterns, as shown in Figure 3B. The relative intensities between the corresponding reflection lines of each component were strongly affected by the LDH/TiO2 ratio. The lower ratios, which correspond to the higher TiO2 loading, displayed, as expected, higher intensities for the TiO2 reflection lines compared to the LDH reflection lines. On the FTIR spectra, as shown in Figure S4 and described above, all of the nanocomposites showed the LDH characteristic bands, including the ν CO3 2− at 1357 cm −1 , which confirmed the presence of carbonate anions in the interlayer domain. The decrease in the LDH/TiO2 ratio induced enlargement of the bands below 1000 cm −1 attributed to the greater contribution of the νO-Ti-O of TiO2. To get a better insight into the efficiency of the direct coprecipitation for immobilizing TiO2 nanoparticles, the elementary analysis of the nanocomposites was performed, as shown in Table 1. To get a better insight into the efficiency of the direct coprecipitation for immobilizing TiO 2 nanoparticles, the elementary analysis of the nanocomposites was performed, as shown in Table 1. Systematically, the LDH matrices precipitated with an Mg/Al ratio close to that theoretically fixed at 2, whatever the amount of TiO 2 involved in the synthesis. Moreover, the higher the TiO 2 content involved in the coprecipitation, the higher the amount of TiO 2 in the nanocomposite. Compared to the TiO 2 amount involved, in all cases the percentage of TiO 2 immobilized by [MgAl] LDH was greater than 80%, even for the highest amount of TiO 2 , highlighting the efficiency of the process. This value can, for instance, be advantageously compared with the one obtained in using the delamination/restacking process to immobilized TiO 2 nanoparticles, leading to an immobilization of only 53% [54]. In parallel, the mole number of H 2 O in each nanocomposite was determined by TGA, according to the mass loss measured between 25 • C and 125 • C, as shown in Figure 4A. The number of water molecules progressively increased in the nanocomposites when the amount of TiO 2 was higher. Oppositely, the total mass loss decreased for the lower LDH/TiO 2 ratio from 41.6% to 27.0% for a [MgAl]/TiO 2 ratio of 8 and 0.66, respectively, as shown in Table 1.  The morphology of the [MgAl/TiO2]2 nanocomposite obtained by coprecipitation was investigated by TEM and HRTEM as a representative sample. From the TEM image of the nanocomposite displayed in Figure 5a, it was possible to distinguish the presence of slightly darker TiO2 nanoparticles, well dispersed on the surface of the ill-defined LDH platelets. Such a decrease is in good agreement with the increased amount of TiO 2 into the hydroxylated matrices, while the difference in the hydration rate into the nanocomposite may be explained by a modification of the textural properties. The surface area (Brunauer-Emmett-Teller BET) and the pore characteristics were investigated by N 2 adsorption-desorption measurements, as shown in Figure 4B and Figure S5, which clearly showed important modifications. While according to the IUPAC classification, the pure displayed a Type II adsorption isotherm with a H3 hysteresis loop, the TiO 2 immobilization into the materials progressively led to a change of the isotherm shape to a Type 4 with a H4 hysteresis.
This change of the pore shape was accompanied by a significant increase in the surface area from 126 m 2 g −1 to 253 m 2 g −1 when the [MgAl]/TiO 2 ratio fell from 8 to 0.66, emphasizing that the TiO 2 immobilization at the [MgAl] surface generated a specific interarticular porosity which could positively affect their photocatalytic performances in facilitating the accessibility of the active sites.
The morphology of the [MgAl/TiO 2 ] 2 nanocomposite obtained by coprecipitation was investigated by TEM and HRTEM as a representative sample. From the TEM image of the nanocomposite displayed in Figure 5a, it was possible to distinguish the presence of slightly darker TiO 2 nanoparticles, well dispersed on the surface of the ill-defined LDH platelets. The morphology of the [MgAl/TiO2]2 nanocomposite obtained by coprecipitation was investigated by TEM and HRTEM as a representative sample. From the TEM image of the nanocomposite displayed in Figure 5a, it was possible to distinguish the presence of slightly darker TiO2 nanoparticles, well dispersed on the surface of the ill-defined LDH platelets. Note that the small size of TiO2 particles made their observation in TEM difficult in the presence of LDH particles. The HRTEM image, shown in Figure 5b, recorded at higher magnification, further evidencing that the TiO2 nanoparticles immobilized at the LDH surface were crystalline with a characteristic lattice spacing of 0.35 nm between the 101 planes of anatase TiO2. In parallel, the Note that the small size of TiO 2 particles made their observation in TEM difficult in the presence of LDH particles. The HRTEM image, shown in Figure 5b, recorded at higher magnification, further evidencing that the TiO 2 nanoparticles immobilized at the LDH surface were crystalline with a characteristic lattice spacing of 0.35 nm between the 101 planes of anatase TiO 2 . In parallel, the Selected Area Electron Diffraction (SAED) also helped to gain insight into the TiO 2 crystalline structure. The red curve in Figure 5c corresponds to the average intensity over circles in the function of the scattering vector and fits well with the one expected for an anatase structure of TiO 2 . EDX analyses at different selected locations, shown in Figure 5d and in the supporting materials Figure S6, confirm the results previously obtained by ICP-AES, shown in Table 1, and the local presence of the two components, LDH and TiO 2 .

Photodegradation of OII in the Presence of Colloidal TiO 2 and LDH Based Nanocomposites
The immobilization of TiO 2 nanoparticles on LDH particles should facilitate the recovery of the photocatalysts from a water solution and could enhance the photocatalytic efficiency in environmental treatment processes [48].
The photocatalytic activity of colloidal TiO 2 and nanocomposites, prepared by either impregnation or coprecipitation, was evaluated by the degradation of OII ([OII] = 5.10 −5 M) at a fixed TiO 2 loading ([TiO 2 ] = 0.5 g L −1 ) under UV-visible irradiation and compared to the efficiency of pure TiO 2 nanoparticles in basic suspensions. Blank experiments were systematically carried out in the absence of any inorganic particles (TiO 2 and LDH) and in the presence of only LDH particles (1.0 g L −1 ), which made it possible to verify that, under our conditions, the OII did not undergo direct hydrolysis and that the LDH had no photocatalytic effect. Figure 6A compares the degradation efficiency of pure colloidal TiO 2 in acid and basic conditions and for the fresh and aged dispersion. As previously reported [52,54], a net decrease was observed in the TiO 2 photocatalytic activity when the pH of the solution was increased. Note that acidic conditions are not compatible with environmental uses. Surprisingly, the aging of the TiO 2 dispersion, which led as described above to the enhancement of the nanoparticle crystallinity, did not strongly affect the photocatalytic behavior. Only a slight enhancement of OII degradation was noted after 2 h of irradiation.
photocatalysts from a water solution and could enhance the photocatalytic efficiency in environmental treatment processes [48]. The photocatalytic activity of colloidal TiO2 and nanocomposites, prepared by either impregnation or coprecipitation, was evaluated by the degradation of OII ([OII] = 5.10 −5 M) at a fixed TiO2 loading ([TiO2] = 0.5 g L −1 ) under UV-visible irradiation and compared to the efficiency of pure TiO2 nanoparticles in basic suspensions. Blank experiments were systematically carried out in the absence of any inorganic particles (TiO2 and LDH) and in the presence of only LDH particles (1.0 g L −1 ), which made it possible to verify that, under our conditions, the OII did not undergo direct hydrolysis and that the LDH had no photocatalytic effect. Figure 6A compares the degradation efficiency of pure colloidal TiO2 in acid and basic conditions and for the fresh and aged dispersion. As previously reported [52,54], a net decrease was observed in the TiO2 photocatalytic activity when the pH of the solution was increased. Note that acidic conditions are not compatible with environmental uses. Surprisingly, the aging of the TiO2 dispersion, which led as described above to the enhancement of the nanoparticle crystallinity, did not strongly affect the photocatalytic behavior. Only a slight enhancement of OII degradation was noted after 2 h of irradiation. The two nanocomposites obtained by the impregnation of ZnAl LDH matrices noted that [ZnAl/FreshTiO2]2dry and [ZnAl/FreshTiO2]2wet were tested, as shown in Figure 6B. It should be underlined that better degradation results were obtained when the impregnation process was carried out from an undried LDH suspension compared to LDH powder, the latter displaying almost no photoactivity. These results showed that the immobilization of TiO2 was probably more efficient using undried LDH, allowing for a greater surface interaction between the LDH and TiO2 nanoparticles, which led to a higher immobilized amount. However, both samples displayed a The two nanocomposites obtained by the impregnation of ZnAl LDH matrices noted that [ZnAl/FreshTiO 2 ] 2dry and [ZnAl/FreshTiO 2 ] 2wet were tested, as shown in Figure 6B. It should be underlined that better degradation results were obtained when the impregnation process was carried out from an undried LDH suspension compared to LDH powder, the latter displaying almost no photoactivity. These results showed that the immobilization of TiO 2 was probably more efficient using undried LDH, allowing for a greater surface interaction between the LDH and TiO 2 nanoparticles, which led to a higher immobilized amount. However, both samples displayed a reduced photocatalytic efficiency compared to pure TiO 2 basic dispersion, which could be explained by a lower quantity of photocatalyst and/or a partial aggregation and decrease in its photo efficiency.
Since impregnation did not appear as a suitable approach to immobilize TiO 2 on LDH and to produce efficient nanocomposites, we tested an alternative strategy; samples obtained by coprecipitation which corresponded to the direct formation of LDH in the presence of a dispersion of TiO 2 nanoparticles. In Figure 7A, the efficiency of the nanocomposites obtained from the chemical composition ZnAland MgAl-LDH and fresh TiO 2 were compared. [MgAl/FreshTiO 2 ] 2 led to the highest OII degradation. The results even further improved when aged TiO 2 was involved in the MgAl nanocomposite. In this latter case, the immobilized TiO 2 behaved at the earliest stages of the degradation in the same way as the pure colloidal aged TiO 2 , while after one hour a decrease in the rate of the reaction occurred, showing a kinetic of degradation in two steps.
nanocomposites towards OII degradation compared advantageously to those previously obtained by TiO2 immobilized on glass slides, TiO2/SiO2 composites, ZnCr LDH, and derived oxides, as shown in Table S1 [55][56][57]. Note that the reduced graphene oxide has also been used for the immobilization of TiO2 by a simple solvothermal method, leading to extremely efficient nanocomposites, which may be due to the ability of reduced graphene oxide to suppress the recombination of photo-induced electron-holes pairs [58]. The fate of OII in the solution and especially the complete mineralization by photodegradation using [MgAl/TiO2]0.66 was investigated by measuring the total organic carbon evolution in the solution, as shown in Figure 8A, B. Clearly, with increasing irradiation time, the concentration of To get a better insight on the photocatalytic properties of the [MgAl/TiO 2 ] nanocomposites obtained with aged TiO 2 , a systematic study based on the amount of TiO 2 immobilized, shown in Table 1, was conducted for the different LDH/TiO 2 ratios. Since the LDH/TiO 2 experimental mass ratios were systematically higher than the theoretical ones, shown in Table 1, the amount of used nanocomposites was precisely determined to involve 0.5 g L −1 of TiO 2 . The higher the TiO 2 amount in the nanocomposites, the greater the photodegradation of the OII, as shown in Figure 7B. Such a trend can be explained by the screening effect produced by the LDH, limiting the photon penetration in the whole suspension. Interestingly, for an experimental LDH/TiO 2 ratio of 1:1, which corresponded to the highest loading of TiO 2 within the nanocomposite tested, a faster degradation rate of OII was measured by comparison, with pure TiO 2 leading to more than 85% of degradation after 4 h of irradiation. We assumed that the homogeneous distribution of TiO 2 and the high surface area within the nanocomposite [MgAl/TiO 2 ] 0.66 were responsible for the enhancement of the photocatalytic properties. Under the conditions used, the photocatalytic efficiency of the optimized LDH-based nanocomposites towards OII degradation compared advantageously to those previously obtained by TiO 2 immobilized on glass slides, TiO 2 /SiO 2 composites, ZnCr LDH, and derived oxides, as shown in Table S1 [55][56][57]. Note that the reduced graphene oxide has also been used for the immobilization of TiO 2 by a simple solvothermal method, leading to extremely efficient nanocomposites, which may be due to the ability of reduced graphene oxide to suppress the recombination of photo-induced electron-holes pairs [58].
The fate of OII in the solution and especially the complete mineralization by photodegradation using [MgAl/TiO 2 ] 0.66 was investigated by measuring the total organic carbon evolution in the solution, as shown in Figure 8A, B. Clearly, with increasing irradiation time, the concentration of organic carbon into the solution decreased from 13.7 mg L −1 to 0 after 48 h, while simultaneously the inorganic carbon content increased until 16.2 mg L −1 . These evolutions highlighted that the [MgAl/TiO 2 ] 0.66 led to the effective total mineralization of OII during the photodegradation process. These results indicate that, during the photocatalytic process, the OII is completely degraded, producing harmless byproducts such as CO 2 and H 2 O. Based on the literature and our previous results [56,58], we may assume that, in the photodegradation mechanism, the HO• radical plays the main role, however, further experiments involving radical scavengers will be necessary to get a better insight into the OII degradation pathway.
[MgAl/TiO2]0.66 led to the effective total mineralization of OII during the photodegradation process. These results indicate that, during the photocatalytic process, the OII is completely degraded, producing harmless byproducts such as CO2 and H2O. Based on the literature and our previous results [56,58], we may assume that, in the photodegradation mechanism, the HO• radical plays the main role, however, further experiments involving radical scavengers will be necessary to get a better insight into the OII degradation pathway.

Conclusions
[LDH-TiO2] nanocomposites were synthesized by both impregnation and direct coprecipitation using fresh and aged TiO2 nanoparticles. The immobilization of TiO2 nanoparticles on the LDH surface was favored by electrostatic interactions at basic pH between negatively charged TiO2 and positively charged LDH layers. While the [ZnAl-TiO2] nanocomposite, obtained by the impregnation of a dried LDH powder, had no photocatalytic properties, the use of the non-dried ZnAl-LDH matrix favored the immobilization of TiO2 and led to OII photodegradation. We have shown that the coprecipitation is an efficient strategy for the immobilization of TiO2 with a high immobilization rate (>80%). The nature of the LDH chemical composition and the aging of TiO2 nanoparticles modified the photocatalytic efficiency of the materials, the best results being obtained by the coprecipitation of MgAl-LDH matrix and aged TiO2. In this coprecipitation approach, the LDH/TiO2 ratio was easily varied within the nanocomposites, which have the structural characteristics of each constituent. The presence of TiO2 nanoparticles strongly modified the textural properties of the materials. The higher the TiO2 loading, the greater the surface area. The best nanocomposite developed in this study was synthesized by the coprecipitation of the MgAl matrix in the presence of basic aged TiO2 dispersion. An experimental LDH/TiO2 ratio of 1:1 allowed us to develop a photocatalyst with more efficient than pure TiO2 nanoparticles and it was easy to recover, leading to total OII mineralization after 48 h of irradiation. This study highlighted the importance of the method of preparing nanocomposites and the conditions used on final photocatalytic performances.

Conclusions
[LDH-TiO 2 ] nanocomposites were synthesized by both impregnation and direct coprecipitation using fresh and aged TiO 2 nanoparticles. The immobilization of TiO 2 nanoparticles on the LDH surface was favored by electrostatic interactions at basic pH between negatively charged TiO 2 and positively charged LDH layers. While the [ZnAl-TiO 2 ] nanocomposite, obtained by the impregnation of a dried LDH powder, had no photocatalytic properties, the use of the non-dried ZnAl-LDH matrix favored the immobilization of TiO 2 and led to OII photodegradation. We have shown that the coprecipitation is an efficient strategy for the immobilization of TiO 2 with a high immobilization rate (>80%). The nature of the LDH chemical composition and the aging of TiO 2 nanoparticles modified the photocatalytic efficiency of the materials, the best results being obtained by the coprecipitation of MgAl-LDH matrix and aged TiO 2 . In this coprecipitation approach, the LDH/TiO 2 ratio was easily varied within the nanocomposites, which have the structural characteristics of each constituent. The presence of TiO 2 nanoparticles strongly modified the textural properties of the materials. The higher the TiO 2 loading, the greater the surface area. The best nanocomposite developed in this study was synthesized by the coprecipitation of the MgAl matrix in the presence of basic aged TiO 2 dispersion. An experimental LDH/TiO 2 ratio of 1:1 allowed us to develop a photocatalyst with more efficient than pure TiO 2 nanoparticles and it was easy to recover, leading to total OII mineralization after 48 h of irradiation. This study highlighted the importance of the method of preparing nanocomposites and the conditions used on final photocatalytic performances.