Introduction

Titanium dioxide (TiO2) is a semiconductor of relevant use due to its excellent photocatalytic, optical, and electrical properties (Dahl et al. 2014). Many studies have been carried out proving that TiO2 is adequate for several applications, such as photodegradation of organic contaminants in water and air (Humayun et al. 2018; Katal et al. 2020; Bianchi et al. 2014), dye-sensitized solar cells (Wu et al. 2015; Ahmad et al. 2017), self-cleaning windows and walls (Shen et al. 2015), hydrogen products, and solar cells (Chen et al. 2017). Among several applications, TiO2 has been widely investigated for photocatalysis and has been accepted as a promising technology for the complete degradation and mineralization of organic contaminants in the environment (Gunti et al. 2018; Shayegan et al. 2018).

A relevant aspect of the preparation of TiO2 catalysts is the development of particles of very small size, with high specific surface area, controlled porosity, and pore size distribution, seeking to enhance their catalytic activity and, therefore, the efficiency of the process (Yoo et al. 2005). Titanium dioxide particles have been synthesized through several methods, including sol-gel (Yoo et al. 2005; Agartan et al. 2015), hydrothermal (Mamaghani et al. 2019), emulsion (Dong et al. 2014), ultrasound (Ghows et al. 2010), and microwave heating (Cui et al. 2012).

Among the proposed synthesis methods, supercritical fluid (SCF) media are among the most recent. The use of supercritical CO2 (scCO2) as an alternative route for preparing new materials can considerably reduce environmental risks, as it is environmentally friendly, non-toxic, non-inflammable, cheap, and a recyclable technology. In the precipitation process using CO2, the sol-gel reaction takes place in a supercritical environment, favoring rapid hydrolysis of the precursor and a fast condensation process. Hence, the dissolution of CO2 in the precursor solution leads to a rapid expansion of the phase liquid, causing rapid precipitation of solutes (Silva et al. 2019). Likewise, precipitation of catalytic inorganic materials with particular characteristics such as particle size, size distribution, crystallinity, polymorphism, and specifically high surface area can be obtained by employing carbon dioxide as the reaction medium or an antisolvent because of its tunable thermodynamic properties (Aaymonier et al. 2006; Cansell and Aaymonier 2009; Adschiri and Yoko 2018). In this context, several studies have reported the preparation of TiO2 as micro and nanoparticles with specific catalytic properties employing supercritical carbon dioxide as reaction medium (Tadros et al. 1996; Stallings and Lamb 2003; Alonso et al. 2007; Larder et al. 2021) or as an antisolvent (Tang et al. 2006; Silva et al. 2014; Marin et al. 2015).

Tang et al. (2006) precipitated a TiO2 precursor, namely, titanium(IV) oxyacetylacetonate, using scCO2 as antisolvent by the SAS technique from a methanol solution at fixed temperature and pressure values. The aim was to prepare titanium oxide as a support for gold catalysts by calcination of precipitated particles from the precursor. After calcination, gold catalysts were supported into TiO2 particles. According to the authors, these catalysts have a high activity toward the oxidation of carbon monoxide compared with the same catalyst prepared using other methods.

Silva et al. (2014) synthesized TiO2 using an anti-solvent expansion with scCO2 and used the titanium (IV) isopropoxide - (TTIP) precursor with 250 bar pressure and 60 °C. They obtained an average particle size of 316.7 ± 2.7 nm and a surface area of 418.5 m2 g–1. According to the authors, using scCO2 expansion technology as the reaction medium produced a nanomaterial with a large surface area, which is attractive for adsorption processes. The same authors (Silva et al. 2019) later used the SAS technique to produce TiO2 nanoparticles with an anatase-rutile phase from the same precursor (TTIP). The precipitated material was tested as a catalyst for dye photodegradation and compared with a commercial mixed-phase TiO2. They observed that TiO2 precipitated from SAS displayed higher activity than the commercial photocatalyst. Marin et al. (2015) reported similar results in the precipitation of TiO2 particles from different titanium alkoxides as precursors employing the SAS technique. After calcination, they obtained a precipitated powder that presented a mixed anatase/rutile phase. The authors stated that the catalytic activity of SAS-TiO2 obtained in their work is equivalent to a commercial one.

On the other hand, ionic liquids (ILs) have received significant attention and have been used in different synthesis reactions. They are widely used as organic solvents or reagents in chemical reactions. ILs consist of cations and anions linked through ionic bonds, showing a liquid state around room temperature. ILs are ideal media that combine two advantages: the liquid state of molecular liquids and the ionic character of ionic compounds (Blanchard et al. 1999; Chen and Mu 2021).

They can act as stabilizers due to their low surface tension, resulting from their composition (positive and negative ions) and low vapor pressure. Consequently, ILs reduce air pollution risk, making them ideal substitutes for conventional organic solvents. Besides, theregeneration of ILs is easy since the bond between the cation and the anion in ILs is strong; thus, they can be recovered and reused repeatedly (Chen and Mu 2021). ILs have also been studied to obtain TiO2 with a high surface area through the sol-gel synthesis method (Yoo et al. 2005; Shahi et al. 2015) studied the use of different ILs to synthesize TiO2 particles via the sol-gel technique. They showed that among the ILs examined, the 1-butyl-3-methylimidazolium hexafluorophosphate ([Bmim][PF6]) was the most suitable to prepare mesoporous TiO2 particles with a BET area of approximately 478 m2 g–1.

The ILs are excellent solvents for various substances; inorganic, organic compounds, biomolecules, organometallics, and metallic ions. Considering their potential as solvents, ILs can naturally replace conventional organic solvents used in large quantities in chemical processing industries (Keskin et al. 2007). Still, they are water-soluble, and if they enter the aquatic environment through accidental spills or effluents, they can cause environmental problems (Freire et al. 2008). Furthermore, systematic studies on the toxicity and biodegradability of IL must be conducted for its safe use in industry (Chen and Mu 2021).

The low volatility of the ILs also causes product separation and recovery complications. Thus, if the product is volatile, retrodistillation can be used to remove the product from the IL; if a hydrophilic product is obtained in a hydrophobic IL, water can be used to remove the product from the IL. But if the product is not volatile or thermosensitive, it becomes impracticable to recover (Huddleston et al. 1998; Zhao et al. 2005). Therefore, scCO2 is used because it recovers different solutes from ILs without cross-contamination (Keskin et al. 2007).

The scCO2 is volatile and non-polar, forming attractive two-phase systems with non-volatile and polar ILs. The product recovery procedure with these systems is based on the premise that scCO2 is soluble in ILs, but ILs are insoluble in scCO2. As most organic compounds are soluble in scCO2 and scCO2 is soluble in ILs, these products are transferred from the IL to the supercritical phase, causing a complete separation within the systems (Blanchard et al. 2001; Zhao et al. 2005; Keskin et al. 2007).

Thus, here we report a route to synthesize mesoporous TiO2 particles using supercritical carbon dioxide as antisolvent in an ambient saturated with water in the presence of the ionic liquid [C8mim][NTf2]. Further, the synthesized particles were characterized by X-ray diffraction (XRD), thermogravimetric analysis (TG/DTA), Brunauer-Emmet-Teller (BET) using the Barrett-Joyner-Halenda (BJH) method, and field emission gun scanning electron microscopy (FEG-SEM) measurements to understand the influence of the use of the IL and operational conditions on the size and other structural characteristics of the particles. The amount of IL and the proportion between the Ti precursor and the solvent used affected the size and properties of the developed particles.

Experimental

Materials

Isopropanol (iPrOH) (99.5%) and TTIP (97.0%) were purchased from Sigma-Aldrich, the IL 1-methyl-3-octylimidazolium bis[trifluoromethylsulfonyl] imide, [C8mim][NTf2] was purchased from Ionic Liquids Technologies (Iolitec) with a purity of 99.0%, and carbon dioxide was obtained from White Martins (99.0%).

Reactive TiO2 precipitation by SAS technique

The precipitation of TiO2 particles was carried out using an experimental apparatus based on a technique that uses supercritical carbon dioxide as an anti-solvent (SAS), according to the scheme displayed in Fig. 1. More details of the experimental apparatus, with some modifications, and its operation can be found elsewhere (Franceschi et al. 2008).

Fig. 1
figure 1

Adapted from Franceschi et al. (2008)

Experimental apparatus used in the precipitation of TiO2 particles by the SAS technique employing carbon dioxide as antisolvent. V1—ball valve; V2 and V3—needle valve; SP—Syringe Pump; PT—pressure transducer; PI—pressure indicator; TI—temperature indicator; TC—temperature control.

Full size image

For precipitation experiments, initially, a solution was prepared with TTIP on isopropanol (iPrOH) under an inert atmosphere with TTIP:iPrOH molar ratios of 0.01, 0.03, and 0.05. The IL [C8mim][NTF2] was then added to this solution in the molar percentage of 1, 3, and 5% IL related to TTIP and stirred for 10 min, given a homogeneous solution. Hence, the nomenclature used in this study for the synthesized particles is TiO2 x_y% IL, where x represents the TTIP:iPrOH molar ratio and y is the IL molar percentage.

Briefly, the experimental apparatus was mainly composed of a syringe pump (ISCO, model 500D) and a positive displacement HPLC pump (Acuflow Series III) that are used to flow the liquid CO2 and the solutions, respectively, into the precipitation chamber that consists of a cylindrical stainless steel vessel (homemade) with an internal volume of 500 mL. The precipitation chamber was built to support pressures and temperatures up to 30 MPa and 100 °C.

Precipitation of TiO2 was conducted at a constant pressure and temperature of 8.0 MPa, and 40°C. Carbon dioxide was pumped into the precipitation chamber through a stainless-steel tube of 1/16” external diameter at a flow rate of 20 mL min–1 and a pressure of 10 MPa. The flow inlet and outlet of the precipitation chamber were controlled by two micrometering valves (V2–V3). The solution was pumped into the top of the precipitation chamber through a silica capillary tube with an internal diameter of 150 μm at a flow rate of 2 mL min–1 and pressure of 14 MPa controlled by a backpressure regulator (GO-Regulator, Series BP-66, Model 1A11QEQ151). The pressure inside the chamber was monitored using a pressure transducer (PT) (HUBA CONTROL, model 691). The internal temperature of the precipitation chamber and the needle valve (V3) was kept through heating tapes (FISATOM, Model 5, at 200 W) coupled with temperature controllers (TC) (NOVUS, Model N1200).

The experimental procedure starts with the addition of 10 mL of ultrapure water (purified by a Milli-Q system from Millipore®) to the precipitation chamber. The chamber is then pressurized with CO2 at a constant flow of 20 mL min–1 while the temperature is raised to experimental values. The water added to the precipitation chamber promotes a CO2 water-saturated environment, thereby accelerating the hydrolysis reaction and TiO2 precipitation. When pressure and temperature are constant at the desired value, the solution is pumped into the precipitation chamber at 2 mL min–1 simultaneously with the liquid flow of CO2. After that, scCO2 continued to flow through the chamber at 40 °C under 20 mL min–1 and 8.0 MPa for 30 min more to remove non-reacted materials. Then, the precipitation chamber was slowly depressurized, and the suspension of TiO2 particles in the water was collected, filtered, and then dried at 60 °C. Further, the solid particles were calcined at 450 °C for 2 h with a heating rate of 3 °C/min in airflow. The precipitated TiO2 yield in each experiment was around 85% regarding the theoretical yield (100% of the reacted TTIP). This difference is related to unreacted TTIP and TiO2 losses during collection at the end of each experimental condition.

Characterization

X-ray diffraction analyses were conducted in a PANalytical diffractometer Empyrean, operating with CuKα radiation (λ = 0.15406 nm), with a 0.02° step and a scanning speed of 0.5° per min, in the range 20° < 2θ < 80°. The crystallite size of the particles was estimated using the Debye Scherrer equation (Eq. 1):

$$D=\frac{\text{0,94} * \lambda }{\beta * \text{cos}\,\theta },$$
(1)

where in (D) is the crystallite size in nanometers, λ is the wavelength of the X-ray radiation (0.1548 nm for CuKα radiation), and β is the full width at half-peak height (FWHM). The full width values at half-maximum were obtained using the X’Pert HighScore software.

The thermal stability of the materials was tested using thermogravimetric analysis on a Shimadzu Simultaneous DTA-TG apparatus. Nearly 10 mg of each sample was analyzed from room temperature up to 1000 °C at a heating rate of 10 °C min–1, with a flow of 50 mL min–1 N2 using tin recipients. The physisorption of N2 at 77 K was measured and used for the textural characterization of the catalysts. The specific surface area and pore size distribution were obtained from the adsorption-desorption isotherms by BET and BJH methods. The physical adsorption of pure nitrogen was carried out in a Quantachrome Nova 1200 Multistation Instrument, model Autosorb 3B. Samples were pretreated under a vacuum at 100 °C for 2 h. SEM-FEG micrographs were obtained in a scanning electron microscope (JSM-6510LV) equipped with a microanalysis by energy-dispersive X-ray spectroscopy (EDS) system. No pretreatment or gold recovery was necessary. The average particle size was estimated from SEM-FEG images (counting about 400 particles in each image) by using the image analysis software Solutions Size Meter 1.1.

Results and discussion

X-ray diffraction

Figure 2a and b show the XRD patterns of the samples precipitated by SAS before and after calcination at 450 °C. The XRD peaks can be associated with the crystalline phase of TiO2 anatase, with the most intense peak at 2θ = 25.4 °C (JCPDS 73-1764), even though the samples were not calcined (Fig. 2a). Yoo et al. (2005) also observed that as-prepared TiO2 particles show an anatase crystalline phase before thermal treatment. They used ionic liquids as templates in the sol-gel method to prepare mesoporous TiO2 particles. This fact prevents the fast hydrolysis of the titanium precursor when in excessive amounts of water. Hence, crystalline TiO2 particles were directly formed. However, the crystallinity of the synthesized particles was enhanced after calcination. Additionally, peaks associated with the brookite crystalline phase at 2θ = 30.80° were also observed (Fig. 2b). The XRD patterns obtained in our study are similar to those obtained by Paszkiewicz et al. (2016), Shahi et al. (2016), and Liu et al. (2009), who synthesized TiO2 using IL using hydrothermal methods. The simultaneous addition of water and (TTIP) to the supercritical carbon dioxide (scCO2) environment allows controlled precipitation of the TiO2 anatase phase due to the rapid hydrolysis of (TTIP) (Silva et al. 2019).

Fig. 2
figure 2

XRD patterns of the TiO2 samples before (a) and after calcination (b)

Full size image

Reducing the hydrolysis rate provides a longer aging time for forming the sol-gel network. It makes possible an ordered array, thus resulting in the formation of highly porous TiO2 particles. The capping effect exerted by the ionic liquid resulted in an anhydrous area located between the ionic liquid and the precursor TTIP. These conditions can generally suppress the formation of hydroxide and oxyhydrate, generating amorphous species. In this case, IL also favored the formation of TiO2 in the anatase phase, as seen previously.

Thermogravimetric analysis

The weight loss of the TiO2 samples obtained from the TG thermographs is shown in Fig. 3a. The TiO2 sample precipitated without IL presented a total weight loss of approximately 17%, with no significant loss above 500 °C. Similar behaviors were observed for TiO2 0.01_1% IL and TiO2 0.01_5% IL, with total weight losses of 15% and 16%, respectively. The other three samples (i.e., TiO2 0.03_3% IL, TiO2 0.05_1% IL, and TiO2 0.05_5% IL) showed higher total weight losses of 28%, 29%, and 32%, respectively. This effect was induced by the increase in the TTIP/iPrOH ratio.

Fig. 3
figure 3

Thermographs (a) and DTA (b) of samples synthesized in different proportions of TTIp/iPrOH and IL/TTIP before calcination

Full size image

Between 25 and 120 °C, nearly 10% of weight loss occurred in the TiO2 0.03 without IL, TiO2 0.01_1% IL, and TiO2 0.01_5% IL samples, related to the evaporation of water, the physically adsorbed, and organic solvent residues that remained in the TiO2. The weight loss was 21% for both TiO2 0.03_3% IL and TiO2 0.05_1% IL and 12% for the TiO2 0.05_5% IL sample. The DTA curves of these samples (Fig. 3b) showed endothermic peaks around 78 °C that refer to both the dehydration of the samples and the vaporization of isopropanol.

The next stage of weight loss, between 121 and 300 °C, accounted for a mass loss of around 5% for samples TiO2 0.03 without IL, TiO2 0.01_1% IL, TiO2 0.01_5% IL, and TiO2 0.05_5% IL. Besides, DTA patterns showed exothermic peaks in this temperature range associated with the evaporation of water chemically adsorbed, resulting in TiOH condensation and thermal decomposition of non-hydrolyzed TTIP residues (Choi et al. 2006; Silva et al. 2014).

In the last stage of the sample transformation, all samples have a minor weight loss. Between 301 and 500 °C, a mass loss of 2–3% is associated with the thermal decomposition of organic residues strongly bound to the TiO2 structure. Above this temperature range, the sample weights remained unchanged up to 1000 °C, and a residual of 80–70% of the mass is associated with pure TiO2 (Verma et al. 2012; Silva et al. 2014). The transition phase from anatase to rutile was not observed for any sample, which agrees with a previous report that showed high thermal stability of TiO2 synthesized using an IL (Choi et al. 2008).

The DTA curve for the sample TiO2 0.03 without IL exhibits an exothermal peak at 412 °C, corresponding to the transition from the amorphous to the crystalline anatase phase of TiO2. The samples made using the IL did not present this phase transition because the anatase phase is fully structured at the end of the synthesis process, even before thermal treatment, as seen in Fig. 2a. For samples with higher proportions of IL, TiO2 0.01_5% IL, and TiO2 0.05_5% IL, there is an exothermal peak at 314 °C, resulting in an amorphous state transformation from the anatase/brookite phases. Furthermore, it can also be attributed to energy losses, which occur in a combustion reaction, as seen in the combustion of organic matter (Bu et al. 2012; Leyva-Porras et al. 2015).

Specific surface area

The N2 adsorption isotherms (BET method) and pore size distribution (BJH method) obtained for the samples synthesized with and without IL are shown in Fig. 4. All samples presented isotherms that can be identified as type IV, according to IUPAC classifications, characteristic of mesoporous materials (Sing et al. 1985). The pore size distribution of the samples is similar and relatively narrow, in the range of 2.80–3.21 nm (Table 1), which indicates a high homogeneity of the pores, associated with the high mechanical stability of the materials, which is an appreciated prerequisite for its use as a photocatalyst (Yoo et al. 2005; Miao et al. 2006). However, note that the addition of IL increases both the pore volume and pore size of the powders; therefore, the TiO2 0.05_5% IL showed a pore size of 5.8 nm due to the higher IL content (Zhang et al. 2015b).

Fig. 4
figure 4

Adsorption-desorption isotherms of N2 and pore size distribution graphs (BJH) (inset) of particles calcined at 450 °C for samples synthesized in different proportions of TTIP/iPrOH and IL concentration through scCO2: a TiO2_0.03 without IL, b TiO2_0.01_1% IL, c TiO2_0.01_5% IL, d TiO2_0.03_3% IL, e TiO2_0.05_1% IL, and f TiO2_0.05_5% IL

Full size image
Table 1 Summary of the specific surface area, pore volume, pore size distribution, and crystallite size determined for the TiO2 particles synthesized using IL in scCO2
Full size table

The BET surface area of the TiO2 particles was reduced by the thermal treatment conducted at 450 °C, whereas the final area was highly similar (around 100 m2 g–1), independent of the IL content. The particles prepared without IL or low IL content showed the greatest reduction in surface area after calcination, around 50%. Alternatively, TiO2 powder prepared with 3% IL presented higher thermal stability, with less than 40% surface area reduction. The sample with the highest IL content (TiO2 0.05_5% IL) lost only 26% of its surface area, maintaining a highly porous structure after calcination. These findings agree with previous studies by Yoo et al. (2005) and Hu et al. (Hu et al. 2008), who prepared TiO2 in the presence of ILs by the sol-gel method under ambient temperature and pressure conditions.

The total volume of pores of TiO2 particles prepared without ionic liquid was 0.225 cm3/g before calcination and 0.224 cm3/g after calcination at 450 °C; hence, there was an insignificant decrease in the total volume of pores, and these results agree with those observed by Yoo et al. (2005). Conversely, the total pore volume was 0.208 cm3/g and 0.240 cm3/g before and after heat treatment at 450 °C for the samples prepared with 3% ionic liquid (TiO2_0.03_3% IL). This increase in total pore volume may be related to the presence of ionic liquid under supercritical CO2 conditions. This behavior is contrary to those found in the reports of Yoo et al. (2005), who prepared TiO2 in the presence of IL using the sol-gel method, and Sui et al. (2011), who prepared TiO2 particles via supercritical CO2. Notably, the total pore volume increased significantly (twice) in samples with the maximum proportion of IL (TiO2_0.05_5% IL), which changed from 0.173 cm3/g before calcination to 0.346 cm3/g after calcination, indicating a positive effect of the use of the ionic liquid in the synthesis of TiO2 particles under supercritical conditions.

Scanning electron microscopy

FEG-SEM images of all samples are shown in Fig. 5. The developed TiO2 particles have a spherical morphology and a wide particle size distribution (Fig. 6). The mean particle size of the particles is summarized in Table 1.

Fig. 5
figure 5

FEG-SEM images of the particles synthesized via scCO2 in different ratios of TTIP/iPrOH and IL/TTIP: a TiO2_0.03 without IL, b TiO2_0.03_3% IL, c TiO2_0.01_1% IL, d TiO2_0.05_1% IL e TiO2_0.01_5% IL, and f TiO2_0.05_5% IL

Full size image
Fig. 6
figure 6

Particle size distribution of SAS processed samples obtained from SEM images

Full size image

The particles made without IL (Fig. 5a) have an average diameter of 1.12 ± 0.8 μm, while the samples made with IL have an average diameter size of 0.536 ± 0.350 μm. Additionally, TiO2 particles made with IL showed uniform morphology and smooth surfaces, indicating the strong capacity of IL as a structuring agent (Sui et al. 2011; Paszkiewicz et al. 2016).

Although the use of IL decreases the particle size, the higher the IL concentration, the higher the particle size. The particles made using 1% IL have an average diameter of 0.519 ± 0.370 μm, while the powder made with 5% IL has an average diameter of 0.657 ± 0.440 μm. Similar behavior occurs for the smaller ratio of TTIP/iPrOH but with lower intensity. The results found here are in accordance with previous studies by Paszkiewicz et al. (2016), who synthesized TiO2 particles using the solvothermal method in the presence of different ILs ([BMIM][Cl] and [DMIM][Cl]) and with different concentrations of IL. These authors concluded that an increase in the IL concentration increases the particle size, independent of the IL used.

The particle size distribution for all experimental conditions can be observed in Fig. 6. Note that samples with smaller ratios of TTIP/iPrOH (Fig. 5c; Table 1) have the smallest average particle sizes (i.e., 0.326 ± 130 μm and 0.328 ± 0.120 μm) and narrow granulometric distribution. It indicates that the initial concentration of the Ti precursor plays a vital role in crystallization and, consequently, the size of the particles. Furthermore, the average crystallite sizes estimated using the Scherrer equation for all samples range from 6.1 to 8.0 nm. These values of crystallite sizes demonstrate that the TiO2 particles are formed of nanocrystalline grains.

The dissolution of titanium alkoxide and the stabilization of aqueous dispersions in supercritical CO2 are necessary to form spherical particles. The injector nozzle produces very small liquid droplets of precursor solutions that, when in contact with scCO2, rapidly expand the liquid phase. The low amount of water can deplete the hydrolysis of the titania precursor and favor the formation of spherical particles (Silva et al. 2019; Tadros et al. 1996).

Conclusions

Here, we proposed a novel method for TiO2 particles by reactive precipitation using scCO2 as an antisolvent saturated with water and an IL as an adjuvant. The influence of the TTIP/IPrOH molar ratio (0.01, 0.03, and 0.05) and the molar percentage of IL regarding TTIP (1, 3, and 5%) on the characteristics of TiO2 particles were investigated. The synthesized TiO2 nanoparticles are crystalline, in the anatase and brookite phases, after thermal treatment at 450 °C for 2 h. Notably, the samples synthesized using the IL displayed the crystalline phase of anatase without thermal treatment, indicating that the IL suppresses hydrolysis and results in crystalline particles.

Thermogravimetric data indicate that the transition from amorphous to crystalline anatase phase of TiO2 does not occur for samples made using the IL, which agrees with the XRD data. The synthesized particles have high values of the BET surface area. Nevertheless, the higher the amount of IL used, the smaller the surface area after the calcination step. The calcined samples showed high surface area values, ranging from 100 to 118 m2 g–1. The pore size increases as the IL concentration increases, shifting from 2.80 for the sample synthesized without IL to 5.8 nm for a sample made using the IL.

Spherical particles were produced using the developed method. The average particle size also increases as the TTIP/iPrOH ratio increases. Nevertheless, the use of ionic liquid to synthesize TiO2 particles decreases the mean particle size compared to those synthesized without ionic liquid. However, the higher the ionic liquid concentration, the larger the mean particle size, showing that the IL concentration is an essential parameter in the control of precipitated TiO2 particles. Therefore, the developed method opens up the opportunity to synthesize spherical mesoporous TiO2 nanoparticles with high surface area and controlled size.