Hydrothermal Synthesis of Layered Titanium Phosphate Ti 2 O 2 H(PO 4 )[(NH 4 ) 2 PO 4 ] 2 and Its Potential Application in Cosmetics

: Titanium phosphates were recently revealed as promising cosmetic pigments; however, their photocatalytic activity and sun protective factor (SPF) levels have not been investigated in detail. In this study, we used hydrothermal conditions to prepare nanocrystalline anatase, brookite, and layered titanium phosphate using the titanium lactate complex, NH 4 H 2 PO 4 , and urea as precursors. The samples were characterized by powder X-ray di ﬀ raction (XRD) in addition to Raman spectroscopy, transmission and scanning electron microscopy (TEM, SEM), energy-dispersive X-ray spectroscopy (EDX), and UV-Vis spectroscopy. Furthermore, the photocatalytic activity, sun protective factor, and moisture retention ability were determined for the samples. Brookite exhibited the highest SPF value and anatase the lowest, while Ti 2 O 2 H(PO 4 )[(NH 4 ) 2 PO 4 ] 2 displayed highly promising UV protection and moisture retention properties and, therefore, represents a polyfunctional pigment that is particularly well suited for cosmetic applications.


Introduction
Titanium oxide has been widely used as a cosmetic pigment and physical UV-filter for more than 40 years [1]. However, some titanium oxide samples, as well as related materials, exhibit photocatalytic activity that leads to reactive oxygen species (ROS) formation [2][3][4][5]. ROS can damage the organic components of cosmetic products and the skin sebum, both of which contribute to irritating inflammatory reactions on the skin surface [6,7]. Therefore, current efforts toward developing new pigments also involve monitoring and suppressing their potential photocatalytic activity.
There are several approaches to suppressing the photocatalytic reactions on the surface of titania, including using amorphous titanium oxide, which has less photocatalytic activity than the crystalline phases [8]. Additionally, photocatalytic activity may be suppressed by doping the titanium oxide with metal ions [9][10][11] or modifying its surface composition [12][13][14]. Matsukura and Onoda showed that phosphate modification leads to photocatalytic activity suppression in TiO 2 -containing pigments [12]. In addition, Onoda et al. [13,14] found that amorphous titanium phosphate possesses less photocatalytic activity than titanium oxide and helps to retain the skin moisture, however, they did not determine the sun-protective factor (SPF) of the samples and thus could not evaluate titanium phosphates as UV-filters.
In the present study, we prepared two nanocrystalline modifications of TiO 2 ; anatase and brookite and crystalline titanium phosphate, by hydrothermal treatment of titanium lactate complex in the For photocatalytic measurements, an original commercially available quartz reactor of the AceGlass Inc was used. The scheme of the setup is shown in Supplementary Materials Figure S1A. The photocatalytic activity of the samples was measured through the discoloration of 4.4 × 10 −5 M methyl orange (Sigma-Aldrich, 85%) aqueous solution due to the first stage of the methyl orange degradation results in destruction of a diazo groups inducing a light absorption. During the experiment the reaction mixture was irradiated with UV illumination of the high-pressure Hg bulb (5.5 W). Illumination spectrum of the bulb is presented on the Figure S1B.
Continuous sampling was provided using a peristaltic pump. A flow of the reaction mixture passed through a U-shaped cell, where absorption spectra were collected using an HRX-2000 xenon lamp and a QE65000 spectrophotometer (Ocean Optics, Largo, FL, USA) every 5 s.
The photocatalytic activity was calculated as a first order constant of the reaction of discoloration. In order to make a comparison with other experimental data, the photocatalytic activity of Evonic P25 was also determined. The measurement procedure and calculation are described in detail in our previous work [8].

Results and Discussion
To study the formation of titanium phosphates under hydrothermal conditions, we varied the concentration of NH 4 H 2 PO 4 in the reaction mixture. In all cases, white powders were obtained. XRD patterns are presented in Figure 1. According to powder XRD, the variation in NH 4 H 2 PO 4 concentration leads to a change in the phase composition of the samples ( Figure 1).  [22,23]. Khainakova et al. prepared this phase using TiCl 4 as the titanium-containing precursor, H 3 PO 4 as the P source, and urea as the ammonium source [22]. In [23] (Figure 2). According to [24], the phase consists of layers which are stacked in the direction of the b-axis. The layers consist of TiO 6 octahedra and PO 4  1003, and 1102 cm −1 , respectively. Finally, the two bands of the ammonium group are also present in the spectrum (1411 and 1675 cm −1 ) [26].   TiP4). Bands were assigned according to [25,26].   TiP4). Bands were assigned according to [25,26]. Bands were assigned according to [25,26]. To get more information about the influence of the increase of Ti:P ratio on the structure and composition the samples were examined using SEM, TEM, and EDX. The results of the EDX analysis are presented in Table 1 and Table S1. In TiP0 no phosphate was determined, and in TiP0.5 phosphate concentration is much lower than in the reaction mixture. For the TiP0.5 sample, this is likely because the phosphate ions only adsorb on the surface of the TiO 2 particles and this affects the crystallization process enough to yield an anatase phase, which is in contrast to the TiP0 sample. In the TiP0.75, TiP1, and TiP1.5 samples, the Ti:P ratio is less than 1.5, due to the presence of a second phase, in which the Ti:P ratio is lower than Ti:P = 1:1.5. An increase in the Ti:P ratios determined for the TiP0.75-TiP1-TiP1.5 sample series indicates an increase in the fraction of Ti 2 O 2 H(PO 4 )[(NH 4 ) 2 PO 4 ] 2 in the composition of the samples. From the XRD patterns, the intensity of the reflections of (NH 4 )TiOPO 4 decrease with the increase in Ti:P ratio. Thus, the EDX data fit nicely with the XRD data. For the TiP2 and TiP4 samples, the Ti:P ratio is 1.5. Therefore, even in presence of excess phosphate ions in the reaction mixture the Ti 2 O 2 H(PO 4 )[(NH 4 ) 2 PO 4 ] 2 phase with a Ti:P ratio of 1.5 is formed.
The morphology of the samples was examined by SEM and TEM. When no phosphate was added to the reaction mixture, rod-like brookite particles formed (Figure 3a,b). This particle shape was observed in [17] for brookite under similar synthetic conditions. TiP0.5 consisted of small aggregated particles around 10 nm in size (Figure 3c,d). Thus, the presence of phosphate ions on the reaction mixture prevented the formation of the brookite phase.
Plate-like particles formed in the samples with Ti:P ratios higher than 0.75 ( Figures S2-S6). For all phosphate concentrations, the thickness of the plates was from 0.2 to 0.5 µm and the lateral size was about 3.6-9.3 µm (Table S2). The thickness and lateral sizes of the particles tend to decrease with the increase of phosphate concentration in the reaction mixture. The plate-like morphology observed in Ti 2 O 2 H(PO 4 )[(NH 4 ) 2 PO 4 ] 2 phase was also described in [22]. With increasing concentrations of phosphate ions and decreasing of (NH 4 )TiOPO 4 content the surfaces seem to become smoother. The samples TiP4 and TiP2, which contain only Ti 2 O 2 H(PO 4 )[(NH 4 ) 2 PO 4 ] 2 phase consist of smooth plate-like particles (Figure 4a,b, Figures S2 and S3), so this phase has a smooth plate-like morphology. In the samples TiP1 and TiP0.75 the plate-like particles are covered by small particles (Figure 4c,d, Figures S5 and S6). According XRD analysis these samples contain (NH 4 )TiOPO 4 impurity, so (NH 4 )TiOPO 4 phase consists of small particles which cover Ti 2 O 2 H(PO 4 )[(NH 4 ) 2 PO 4 ] 2 plates. TiP1.5 sample also contains (NH 4 )TiOPO 4 according XRD, but small particles of this phase are not clearly seen in SEM images ( Figure S4), probably, because of low content of this phase in the sample.
Optical properties of the pure phases were studied by diffuse reflectance UV-Vis spectroscopy ( Figure 5). To determine the optical band gap, the Kubelka-Munk function F was calculated as follows [19]:  2 vs. hν and the energy of direct transition was found to be 3.53 eV ( Figure S7). For the pure brookite and commercial rutile samples, the band gap values agree well with those in the literature [27]. For the commercial anatase and TiP0.5 samples, which also includes anatase, the optical band gaps were found to be at 3.23 and 3.09 eV, respectively. In the literature, the optical band gap for anatase varies from 3.09 to 3.2 eV [27][28][29][30]. This discrepancy cannot be explained by the quantum size effect because the exciton radius in anatase is very small [28,29]. However, the formation of energy levels due to defects in anatase crystalline structure may shift the absorption edge and could cause the apparent change of the band gap.  2 vs. hν and the energy of direct transition was found to be 3.53 eV ( Figure S7). For the pure brookite and commercial rutile samples, the band gap values agree well with those in the literature [27]. For the commercial anatase and TiP0.5 samples, which also includes anatase, the optical band gaps were found to be at 3.23 and 3.09 eV, respectively. In the literature, the optical band gap for anatase varies from 3.09 to 3.2 eV [27][28][29][30]. This discrepancy cannot be explained by the quantum size effect because the exciton radius in anatase is very small [28,29]. However, the formation of energy levels due to defects in anatase crystalline structure may shift the absorption edge and could cause the apparent change of the band gap.      The photocatalytic and photoprotective properties of the commercial and prepared within this study samples are presented in Table 2. Commercial cosmetic pigments and nanocrystalline brookite possessed high photocatalytic activity. The TiP0.5 sample was likely less active because it contained amorphous anatase, or its surface may be modified by phosphate groups. According the literature, these factors may decrease photocatalytic activity [8,12]. The Ti 2 O 2 H(PO 4 )[(NH 4 ) 2 PO 4 ] 2 was also photocatalytically active. The photocatalytic activity of crystalline titanium phosphate phases has been previously described (Ti 2 O(PO 4 ) 2 (H 2 O) 2 in [31] and Bi-containing composite with Ti(HPO 4 ) 2 (H 2 O) in [32]), however, this is the first report to determine the photocatalytic activity of The nanocrystalline brookite (TiP0) possessed the highest SPF of the samples included in this study. However, its photocatalytic activity is also high, so for potential application the suppression of photocatalytic activity is necessary. In contrast, the SPF exhibited by titanium phosphate (TiP4) was the same as the commercial samples and its photocatalytic activity was lower than the commercial samples. Anatase (TiP0.5) displayed the lowest photocatalytic activity, however, its SPF was very low, thus negating its benefit in sun-protective cosmetics. UVAPF (UVA protection factor) values of the samples show the level of protection against UVA part of the spectrum (320-400 nm). The highest value corresponds to brookite sample (TiP0), whereas Ti 2 O 2 H(PO 4 )[(NH 4 ) 2 PO 4 ] 2 does not demonstrate high value. It is clearly seen in the absorption spectra of the samples ( Figure S8), that brookite and Ti 2 O 2 H(PO 4 )[(NH 4 ) 2 PO 4 ] 2 phases possess the highest values of protection from UVB rays (290-320 nm), whereas commercial anatase and rutile actively protect from UVA rays. The combination of different phases in cosmetics may provide broad spectrum protection from both UVA and UVB types of solar radiation.
Another advantage of titanium phosphate is its ability to retain moisture. Figure 6 depicts the rate of moisture loss specific to each sample over time. Among all of the samples investigated, titanium phosphate tended to retain the most moisture for the longest period of time. Therefore, it may be superior to traditional titanium oxide. Furthermore, due to its sufficiently high SPF and lower level of photocatalytic activity than brookite and commercial samples, titanium phosphate may be used as multifunctional cosmetic pigment that confers both moisture retention and sun protection. The photocatalytic and photoprotective properties of the commercial and prepared within this study samples are presented in Table 2. Commercial cosmetic pigments and nanocrystalline brookite possessed high photocatalytic activity. The TiP0.5 sample was likely less active because it contained amorphous anatase, or its surface may be modified by phosphate groups. According the literature, these factors may decrease photocatalytic activity [8,12]. The Ti2O2H(PO4)[(NH4)2PO4]2 was also photocatalytically active. The photocatalytic activity of crystalline titanium phosphate phases has been previously described (Ti2O(PO4)2(H2O)2 in [31] and Bi-containing composite with Ti(HPO4)2(H2O) in [32]), however, this is the first report to determine the photocatalytic activity of The nanocrystalline brookite (TiP0) possessed the highest SPF of the samples included in this study. However, its photocatalytic activity is also high, so for potential application the suppression of photocatalytic activity is necessary. In contrast, the SPF exhibited by titanium phosphate (TiP4) was the same as the commercial samples and its photocatalytic activity was lower than the commercial samples. Anatase (TiP0.5) displayed the lowest photocatalytic activity, however, its SPF was very low, thus negating its benefit in sun-protective cosmetics. UVAPF (UVA protection factor) values of the samples show the level of protection against UVA part of the spectrum (320-400 nm). The highest value corresponds to brookite sample (TiP0), whereas Ti2O2H(PO4)[(NH4)2PO4]2 does not demonstrate high value. It is clearly seen in the absorption spectra of the samples ( Figure S8), that brookite and Ti2O2H(PO4)[(NH4)2PO4]2 phases possess the highest values of protection from UVB rays (290-320 nm), whereas commercial anatase and rutile actively protect from UVA rays. The combination of different phases in cosmetics may provide broad spectrum protection from both UVA and UVB types of solar radiation.
Another advantage of titanium phosphate is its ability to retain moisture. Figure 6 depicts the rate of moisture loss specific to each sample over time. Among all of the samples investigated, titanium phosphate tended to retain the most moisture for the longest period of time. Therefore, it may be superior to traditional titanium oxide. Furthermore, due to its sufficiently high SPF and lower level of photocatalytic activity than brookite and commercial samples, titanium phosphate may be used as multifunctional cosmetic pigment that confers both moisture retention and sun protection.   ] 2 was found to be both UV-protective, at levels comparable with the commercial samples, and moisture-retentive and is therefore an attractive multifunctional pigment for use in the cosmetic industry.
Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4352/9/7/332/s1, Table S1: Data of the energy dispersive X-ray analysis; Table S2: particle size parameters for the samples, which contain Ti 2 O 2 H(PO 4 )[(NH 4 ) 2 PO 4 ] 2 ; Figure S1: (a) Scheme of the photocatalytic measurements setup, (b) Spectrum of the Hg bulb; Figure S2: The microstructure of the TiP4, sample prepared by hydrothermal treatment of titanium lactate complex at 180 • C in the presence of urea and phosphate ions and with Ti:P ratio of 1:4; Figure S3: The microstructure of the TiP2, sample prepared by hydrothermal treatment of titanium lactate complex at 180 • C in the presence of urea and phosphate ions and with Ti:P ratio of 1:2; Figure S4: The microstructure of the TiP1.5, sample prepared by hydrothermal treatment of titanium lactate complex at 180 • C in the presence of urea and phosphate ions and with Ti:P ratio of 1:1.5 (a,b) magnification 50,000 and 5000 respectively; Figure S5: The microstructure of the TiP1, sample prepared by hydrothermal treatment of titanium lactate complex at 180 • C in the presence of urea and phosphate ions and with Ti:P ratio of 1:1; Figure S6: The microstructure of the TiP0.75, sample prepared by hydrothermal treatment of titanium lactate complex at 180 • C in the presence of urea and phosphate ions and with Ti:P ratio of 1:0.75; Figure