The effect of the number of hydrophobic tails of cationic ammonium surfactants on mesoporous TiO2 synthesized

Mesoporous titanium dioxide (TiO2) is one of the most studied mesoporous materials considering its special character and various applications. In the present work, mesoporous TiO2 was synthesized by a sol–gel method employing different hydrophobic tails of ammonium cationic surfactants templates. The prepared samples were characterized by various techniques. The XRD profiles confirmed that all samples crystallized into the TiO2 anatase phase. The crystallite size of all samples was found to vary in the range of 8.60 nm to 13.61 nm. The transition temperature of the anatase phase was increased to several Celsius degrees since TiO2 was fabricated with a template assistant. The surface area of the mesoporous TiO2 was increased in the range of 93 m2.g−1 (CTAB) − 116.8 m2.g−1 (MTAB). These values were larger than the TiO2 synthesized without a template (72 m2.g−1). The total pore volume was also increased between 0.1704 cm3.g−1 (CTAB) and 0.300 cm3.g−1 (MTAB), while the TiO2 synthesized without a template was only 0.161 cm3.g−1. Using CTAB and DDAB yield a uniform mesopore size distribution. MTAB tends to produce non-uniform pore of the mesoporous system. The soft-templating method opens up new possibilities for synthesizing mesoporous metal oxides.


Introduction
Mesoporous TiO 2 has a unique structure and special characteristics which directly correspond to its physical properties and functions. It is considered for its outstanding properties and promising applications in various fields, including photocatalysis, energy, and biosensing [1,2]. It proves that mesoporous TiO 2 is capable of improving the photocatalytic activity of visible light radiation [3] and UV light radiation [4], minimizing the recombination of the electron-hole pairs and optimizing the mass and charge [5], enhancing the conversion efficiency of dye-sensitized cells (DSSC) utilities [6], increasing the performance of lithium-ion battery [7] and stabilizing the storage and thermal of the immobilized enzyme [8].
In the last few years, mesoporous TiO 2 with different morphology, surface areas, and pore volumes have been synthesized with different methods, including sol-gel [9], hydrothermal [10], incipient wetness [11], microwave-assisted hydrothermal [12], solvothermal [13], and template assistant [14]. Using a template during synthesis processes can benefit controlling particle size, morphology, and structure of the material [15]. Generally, the template used can be divided into hard-templating, semi-hard-templating, and soft-templating. The hard-templating usually employs nano-casting with a specific shape and rigid material as a template, Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. including metal oxide [16], block copolymer [17], one-dimensional nanocrystals (carbon nanotubes) [18], and two-dimensional colloidal crystals [19]. The hard templating can precisely control production target dimensions and specifications with high reproducibility and stability. For example, the template of mesoporous silica has succeeded in producing porous and crystalline frameworks crystalline mesopores of TiO 2 [20]. However, the mainly problem of this method arises on separating the template step, which allows extreme processes such as under high hydrolysable templates and high-temperature environments. It eventually may cause damage to the TiO 2 structure [21]. Semi-hard-template usually uses microbiology and micellar arrangements as the template. For example, the bacteriophage M13 like a rod-shaped virus, has highly stable without genetic or chemical manipulation and room temperature. It was reported that the template assistance produced a cylindrical shape of mesoporous TiO 2 structure with high stability [22]. The drawback of this method originated from the complexity of the synthesis process since this method requires high maintenance and a difficult separation process.
As the last method, the soft-templating commonly employs surfactants (such as ammonium salt, heterocyclic, sulfonate salt, and non-ionic molecule), polymer, and biopolymer as templates [1]. In mesoporous materials synthesis processes, aggregates are formed through intermolecular or intramolecular interaction forces (hydrogen bonding, chemical bonding, and static electricity). Inorganic species are deposited on the surface or inside templates through electrochemical, precipitation, and other processes, forming specific shapes and sizes of particles [23]. Mesostructured phases can be synthesized using different charged surfactant molecules and inorganic solution species. Different amphiphilic molecule groups form various arrangements of surfactant in solution, such as liquid crystals, vesicles, micelles, and microemulsions, and self-assembled to reflect the final mesostructured production under different reaction conditions [24].
Numerous types of soft templates have been developed to synthesize mesoporous TiO 2 to facilitate the synthesis of mesoporous TiO 2 . Highly crystalline mesoporous anatase TiO 2 nanospheres were obtained by the sol-gel method and the amino group of the heaxadecylamine (HAD) as a template. This material showed more reactive oxygen species (ROS) than that Degussa P25 commercial under UV-light irradiation due to the mesoporous spherical TiO 2 mainly produces .OH and Degussa P25 TiO 2 commercial generates both 1O2 and . OH [25]. Muniz, et al [26] prepared mesoporous TiO 2 nanostructured films using Triton X-100 non-ionic surfactant as a template, yielding an average energy conversion efficiency of 5.2% for dye-sensitized solar cells application. Nguyen, et al [27] synthesized mesoporous TiO 2 with Pluronic non-ionic surfactant and peroxo titanic acid (PTA) sol as precursor. It allows fabricating of the high surface area and bi-crystallinity phases of meso-TiO 2 which play important roles in enhancing its photocatalytic properties. Meanwhile, the very high surface areas and narrow pore size distributions of mesoporous TiO 2 can be fabricated from the self-assembly of anionic surfactant sodium dodecyl sulfate (SDS) and cosurfactant benzyl alcohol (BA) mixtures under neutral pH [28]. Selvaraj et al [29] have successfully synthesized highly crystalline mesoporous TiO 2 nanoparticles by using cationic cetyltrimethylammonium bromide (CTAB) dodecyl trimethylammonium bromide (DTAB) surfactants as a template. Their solar cell efficiency conversion value is higher than commercial TiO 2 powder (P25). Abdel-Azim et al [30] also reported that the anatase phase was completely produced in an acidic medium. Anatase and brookite phases were produced in a basic medium using the cationic surfactant. It is also reported that the synthesized catalyst has shown good photocatalytic performance activity in the visible irradiation region. It is noted that the morphology characteristic also plays an important role in application titania. The increasing pore size has an effect of increasing the photocatalytic activity [31]. Increasing porous and pores size able to improve the TiO 2 properties as an electron transport layer and photon absorption enhancer [32] and a better long-term stability [33] in solar cell application. Lavasani et al [34] reported that the well-ordered mesoporous nanoparticles of TiO 2 showed higher photodegradation activity compared to slit-shape mesoporous nanoparticles of TiO 2 .
The hydrophobic effect plays an important role in the aggregate absorption of surfactants. It can affect the interaction of surfactants on solid surfaces and contribute to the interaction of surfactants with water. The interaction of water with surfactant head groups and various types of hydrocarbon tails will promote the formation of various surfactant micellization in water [35]. Hydrolysis and condensation of titanium species associated with hybrid micelles results in the formation of a new liquid condensed phase producing titanium oligomers, polymerizing and condensing into a denser with time, resulting in the formation of a mesostructured inorganic framework [25]. Surfactants are amphiphilic molecules which normally contain both hydrophilic group (water-soluble) and hydrophobic group (water-soluble). The present of surfactant in water can reduce the surface tensile of dissolved substance [36]. It is known that the greater number of alkyl groups have the stronger hydrophobic interaction. Nong et al [37] reported that critical micelle concentration (CMC) and surface tensile values decreased with increasing of number of alkyl surfactant. The more the number of hydrophobic chains create the aggregates in the bulk, thus leading to a reduction in surfactant solubility and, consequently, an increase in the KT value.
Relevant works discussing the effect of various cationic surfactants in synthesizing TiO 2 are limited to different hydrophobic surfactant tails. The aim of this research is the effect of the number of hydrophobic tails of cationic ammonium surfactants on the crystal structure, morphology, thermal, and porous characteristic of TiO 2 synthesized. There are three kinds of cationic ammonium surfactant used, including mono-alkyl chain, dialkyl chain, and triple-alkyl chain of cationic surfactants. The various analyzed techniques, including phase formation, structural analysis, thermal analysis, morphology, and N 2 adsorption-desorption isotherm will be used to investigate the material synthesized. These studies were related to the structure of surfactants used as the main subject of this work.

Materials
The chemical compounds used for synthesize process are tetra-n-butyl orthotitanate (TBOT) from Sigma Aldrich, N-cetyl-N,N,N from Merck, trimethyl-ammonium bromide (CTAB) from Merck, Didodecyldimethyl-ammonium bromide 98% (DDAB) from Sigma Aldrich, Methyltrioctylammonium bromide 97% (MTAB) from Sigma Aldrich. Ammonium hydroxide 25% from Merck, ethanol absolute from Merck, nitric acid (HNO 3 ) from Merck, and deionized water were used to prepare the solution. All chemicals were commercial-grade quality without any further treatment.

Synthesis of mesoporous TiO 2
An amount of 2.829 mmol of either CTAB, DDAB, or MTAB surfactant was mixed in 50 ml of ethanol absolute under magnetic stirring. The pH of the solution was adjusted to 3 by adding an HNO 3 solution. Further, a 10 ml stock solution of TBOT was added under vigorous stirring and diluted by water to become 150 ml of the total volume solution. A 100 ml NH 4 OH 2 M solution was added dropwise to the mixture solution under magnetic stirring. The mixture undergoes hydrolysis and a polymerization process; finally forms a transparent, homogenous, and stable TiO 2 sol. After overnight aging, the gel in the liquid was separated using a centrifuge with an angular speed of under 9000 rpm. The final product of TiO 2 sol was washed with deionized water until the wash water became neutral and finally dried overnight at 100°C in an oven. The obtained solid was calcined at 500°C for 4 h to remove the remaining surfactant and in conjunction with the fabrication of the desired TiO 2 . In this research, we created four different samples, which are distinguished by the type of template used during the synthesizing process, the denotation of the samples are as follows; TiO 2 for the sample made without surfactant, TiO 2 -CTAB for the sample made with the assistance of CTAB surfactant, TiO 2 -DDAB for the sample made with the assistance of DDAB surfactant, and lastly TiO 2 -MTAB for the sample made with the assistance of MTAB surfactant.

Samples characterization
The obtained samples were subsequently characterized by using an x-ray diffractometer (XRD, Empyrean Panalytical with Cu-Kα radiation, λ = 1.154 Å) with step size scan of 0.0263, in the 2θ range of 10-80°. The samples' morphology was investigated using scanning electron microscopy (SEM) with 10,000 times magnification. The thermal analysis was investigated by Jade differential scanning calorimeter (DSC) Perkin Elmer with a scan rate of 10°C/min for a non-calcination sample. The N 2 adsorption-desorption isotherm was measured using Quantasorb SI-4-Kr/MP Quantachrome instruments under the liquid nitrogen temperature (77 K). Before measurement, all samples were degassed at 300°C for 3 h under the pressure of 10 −5 torr. The surface area, pore volume, and pore size distribution values were determined using Quantachrome ® QuandraWin tm software installed in the equipment.

Result and discussion
The structure of cationic surfactants employed in this research is shown in figure 1. All three surfactants used have the ammonium bromide functional group. However, the number of alkyl chains is different. CTAB, DDAB, and MTAB surfactants have mono-alkyl chains, di-alkyl chains, and triple-alkyl chains, respectively. The structure of CTAB and DDAB have straight chains with the functional group position on the edge and centre, respectively. MTAB has a three branches chain structure like an isosceles triangle with an ammonium bromide functional group on the centre top.
The phase formation of all samples was confirmed by matching the obtained x-ray diffraction profiles in the range of 2θ = 10-80°with open crystallography database (COD) of titanium dioxide (TiO 2 ). The results of x-ray diffraction profiles for all samples are collected in figure 2. To have an easier interpretation of the obtained data, the reference peaks belong to the TiO 2 with the COD No. 9009086 is also depicted at the bottom of figure 2. The prominent peak at around the position of 2θ ≈ 25°which correspond to the (011) crystal plane of TiO 2 phase is found in all sample. Furthermore, from the comparison between the experimental diffraction peaks with the reference database, it can be noticed that all samples are only crystallized to form the single-phase formation of TiO 2 . The crystallite size of each sample is measured by using the Scherrer equation. The detail of the equation is explained by using the equation below.
where D sch is crystallite size, K is the shape factor which equals 0.89, λ is the wavelength of Cu-Kα radiation (λ = 1.5406 Å), b hkl is the full-width at half-maximum (FWHM) of the prominent peak where the value was obtained from Gaussian fitting function, and θ is the diffraction angle [38,39]. The calculated crystallite size of all samples is summarized in table 1. The crystallite size of the obtained TiO 2 with the soft-templating method was smaller than that of the TiO 2 without the template sample. The TiO 2 without template has the crystallite size value of 13.61 nm. After TiO 2 were prepared by template, the crystallite size decreases in range of 8.60-9.83 nm. The TiO 2 -DDAB is the lowest crystallite size value. The aim of the presence of surfactants in solution is to minimize surface energy or to decrease the surface tensile forces, thereby suppressing the growth of agglomeration and sedimentation so that smaller crystalline size is obtained by using surfactants in the TiO 2 manufacturing process [37]. As reported by Hasab M.G et al [40] that longer hydrophobic tail length is smaller crystallite size [41]. The higher concentration of surfactant in solution can also decreases the size an orientation of crystalline.
To deepen the information from the x-ray diffraction measurement of the obtained TiO 2 , Rietveld refinement analysis was performed by using General Structure Analysis System (GSAS) software. Details of crystallography information, such as lattice parameters, the volume of the unit cell, phase compositions, etc, can be determined by using this method. The quality of the refinement process depends on several factors, including the goodness of fit (χ 2 ) and background factors. For a reliable and appropriate refinement process, the χ 2 value should be 1.00 to 1.30. In addition, the difference between the weighted diffraction profile factor (R wp ) and the expected diffraction profile factor (R p ) should be less than 2% [42]. The results of the Rietveld-refinement process for all samples are depicted in figure 3. The extracted crystallography information and its reliability factors are summarized in table 1. It can be noticed that all samples exhibited a relatively flat (blue line), resulting from the difference between the experimental data (black circle) and reference diffraction pattern (red line). It is summarized that the experimental data refer to the reference database of the TiO 2 anatase phase (tetragonal structure, s.g. I41/amd (141)), and the claim of single-phase formation in all samples is valid. This claim is also strengthened by the obtained χ 2 values, which were found to be in the range of 1.18-1.24, and the difference  Table 1. Refinement analysis of the prepared sample by using the Rietveld method. between the background factors is less than 2% in all samples. The lattice parameters of a = b remained relatively constant in the 3.779-3.782 Å, while the lattice parameter of 'c' fluctuated in the range of 9.481-9.490 Å. As a result of variations in the 'c' lattice parameter, the obtained TiO 2 anatase phase can be oriented more easily along the c-axis than in other directions. Thermal analysis is the study of material properties toward a controlled temperature treatment. The various physical and chemical properties phenomena of material can be observed as a function of temperature, including phase transitions, dehydration, reduction, oxidation, decomposition, and crystallization [43]. The thermal analysis was carried out using differential scanning calorimetry (DSC) measurement at 50°C-450°C with a heating rate of 10°C min −1 . The result of the DSC curves as the function of the temperature of all samples is displayed in figure 4. Two-characteristic peaks appeared in the DSC curve of the TiO 2 sample; the first peak (T 1 ) and the second peak (T 2 ) respectively correspond to the exothermic peak at about 110°C and endothermic peak at about 425°C. The T 1 appears due to hemihydrate's dehydration process and the water evaporation that was adsorbed into the TiO 2 matrix [44]. The T 1 position of the cationic surfactant-assisted samples was slightly lower (T 1 < 100°C) compared to the TiO 2 without template sample. The T 2 formation in the TiO 2 sample indicates the formation and crystallization process of the TiO 2 anatase phase. In comparison with the TiO 2 without template sample, the cationic surfactant-assisted samples exhibited a higher position of the T 2 temperature where the T 2 positions were found at around 450°C. It may be due to the incorporation the cationic surfactant during the synthesizing process. The results are relevant to the work by Hanaor, et al [45], who mentioned that the transition temperature of the anatase phase above 400°C might vary depending on the raw materials and processing method. The effect of cationic surfactants can also be seen in the DSC curves with the formation of the third peak (T 3 ) at around 240°C. The T 3 formation indicates that the template compounds evaporated during the heating process. All cationic surfactant-assisted samples exhibited a similar transformation during the TiO 2 crystallization process.

Lattice parameters R factors
In figure 5(a) the results of SEM observations show that TiO 2 synthesized without using surfactant tends to form a spherical morphology with a relatively large crystalline size of about 13.61 nm according to the results of identification using diffraction techniques. The effect of CTAB on the formation of TiO 2 is seen in figure 5(b) where the TiO 2 crystalline size is obtained smaller around 9.83 nm, so that the total surface area becomes increasingly enlarged which will certainly increase the electrostatic attraction force between particles also enlarged so that it tends to form a kind of agglomeration. Similarly, with the use of surfactant DDAB, TiO 2 particles are formed finer with crystallite size of 8 nm, so as shown in figure 5(c) looks to have formed a more compact agglomeration compared to the agglomeration shown in figure 5(c) or tend to form a sedimentation. Slightly different from the effect of adding MTAB, the crystallite size obtained is about 9.39 nm. From the SEM data in figure 5(d), there is agglomeration of particles with nanoparticles gathered, this is almost like that found in the sample resulting from the addition of CTAB (see figure 5(b)).
The results of the N 2 adsorption-desorption measurement of all samples are depicted in figure 6. In general, all the obtained samples exhibited the adsorption-desorption mechanism, which is related to the characteristic of porous materials where the broad hysteresis loop around the relative pressure (P/P 0 ) of 0.4-0.9 implies the capillary condensation within the mesoporous network. This broad hysteresis loop curve is identified as the IV types of IUPAC recommendation physisorption isotherms [46]. It looks deeper, these curves have different  patterns which attribute different mesoporous pore structure characteristics due to the different types of templates used in synthesizing TiO 2 . The hysteresis loop of TiO 2 without a template (figure 6(a)) is identical to the type H2(a) loop, which indicates either pore-blocking/percolation in a narrow range of pore necks or to cavitation-induced evaporation. The curve also shows a drastic increase near the P/P 0 ≈ 1, which attributes to the capillary condensation in an open-ended cylindrical channel with uniform size and shape. The hysteresis curve of TiO 2 with the CTAB template (figure 6(b)) similarly corresponds to the type H2(b) loop, which is also associated with pore blocking, but the size distribution of neck widths is now much larger. TiO 2 prepared by DDAB (figure 6(c)) and MTAB (figure 6(d)) template consists of slit shape pores with non-uniform size and shape [47,48].
The surface area of porous materials is determined using the Braeuer -Emmett-Teller (BET) method. The N 2 adsorption-desorption curve in the P/P 0 = 0.05-0.30 was used to calculate the specific surface area since this range directly corresponded to the monolayer completion and the beginning of multilayer adsorption. The specific surface area results are shown in figure 7. The TiO 2 without the template has a BET-specific surface area of 75.0 m 2 g −1 . The presence of cationic surfactants as soft-templating media in the preparation process has effectively increased the BET-specific surface area of the samples. The obtained values varied depending on the template used. The sample prepared using the di-alkyl chain surfactant (DDAB) template exhibited the highest surface area of TiO 2 , followed by triple-alkyl chain (MTAB) and mono-alkyl chain (CTAB) surfactants. Compared to the TiO 2 without the template, the BET-specific surface area increased to 116.8 m 2 g −1 for TiO 2 -DDAB, 103.6 m 2 g −1 for TiO 2 -MTAB, and 92.6 m 2 g −1 for TiO 2 -CTAB. The prepared TiO 2 has higher specific surface area value than that Degussa P25 commercial (54 m 2 .g −1 ) and Aeroxide TiO 2 P25 commercial (35-65 m 2 .g −1 ), but lower specific surface area value than that Hombikat UV 100 commercial (>250 m 2 .g −1 ).
Further, the pore volume value of all samples is analysed from the N 2 adsorption-desorption curve in the range of 0.95 < P/P 0 < 1. We observe two types of pore volume: total pore volume and Barrett-Joyner-Halenda (BJH) pore volume. The total pore volume refers to the term of the liquid volume adsorbed at a specific pressure. Meanwhile, the BJH pore volume refers to the calculated volume, which is identical to the mesoporous characteristic [49]. The range of 0.95 < P/P 0 < 1 was selected by assuming that all the pores in the sample had been filled with condensed gas. The results of the total pore volume and BJH pore volume are displayed in figure 8. It can be noticed that the surfactant significantly influences the total pore volume of TiO 2, where all the samples have a larger total pore volume compared to the TiO 2 without the template. Furthermore, it is shown in the figure that the total pore volume value of TiO 2 increases with the increase of the number of surfactant alkyl. MTAB structure with a three-branches direction chain produces a higher value of total pore volume than other surfactants with a straight-structure chain. Referring to the calculation of BJH pore volume, all samples exhibited a small difference between the obtained BJH pore volume with the total pore volume. It implies that all samples are mainly mesoporous and have less microporous content.
Regarding the pore size distribution of the sample, all the mesoporous samples were characterized using the density functional theory (DFT) method. This method defines the characteristics of adsorbent-adsorbate interactions by considering the pores' morphological structure [50]. It uses statistical mechanics to describe the molecular configuration of the adsorbate (the adsorbed fluid). Thus, using the DFT method, an accurate pore size distribution can be obtained for mesopore materials [51]. Figure 9 shows the pore size distribution of all samples obtained from the DFT model calculation. TiO 2 nanoparticles prepared without a template exhibited a uniform pore size distribution domain at average diameter pore size of 4.5 nm. TiO 2 nanoparticles prepared  with one and two hydrophobic tail surfactants (CTAB and DDAB) showed a similar pore size distribution profile with the TiO 2 without a template and a slight shift to larger pore sizes with average diameter pore size value of 4.7 nm and 5 nm, respectively. Moreover, TiO 2 nanoparticle prepared CTAB, and DDAB assistance shows a wider size distribution profile than that of the TiO 2 without template (insert figure 9). As known, both surfactants have straight chain structures. Contrary to other samples, the TiO 2 nanoparticle prepared by three hydrophobic tail surfactants (MTAB) has a non-uniform profile, seen several peaks throughout the pore size distribution curve in the 5.0-8.0 nm. This result was probably caused by MTAB surfactant, which has a threebranches direction chain of the hydrophobic tail which subsequently created a larger structure with a nonuniform pore of the mesoporous system. The result is by previous work of Narayan et al [52] who reported that longer chains of surfactant lead to larger pores, while shorter chains of surfactant result in smaller pores of mesoporous materials.

Conclusions
In this work, mesoporous titanium dioxide (TiO 2 ) nanoparticles have been successfully synthesized using the soft-templating method with the different hydrophobic tails of surfactant. According to the Rietveld-refinement analysis of the diffraction pattern, all samples yield the crystallized single-phase of TiO 2 anatase, and the crystallite size (D) of TiO 2 with the soft-templating medium was smaller than that of the TiO 2 without the template. In the case of the lattice parameters, the obtained TiO 2 anatase phase is easier to orient along the c-axis than in other directions referring to the fluctuating 'c' lattice constant. The surface area and pore volume of the obtained TiO 2 tend to increase with the increasing amount of hydrophobic tail of surfactant. TiO 2 without the template and CTAB and DDAB assistant yield a uniform mesopore size distribution. However, MTAB cooperated TiO 2 synthesis produces non-uniform pores of mesoporous systems.

Data availability statement
All data that support the findings of this study are included within the article (and any supplementary files).