Urea incorporated into ordered mesoporous silica for potential cosmetic application

Urea’s thermal instability and burning on sensitive skin can cause problems for cosmetic formulations. To overcome these drawbacks, urea was incorporated into ordered mesoporous silica (SBA-15). SBA15 was synthesized using tetraethyl orthosilicate and Pluronic P12 in an acid medium. Urea (20 wt.%) was incorporated into calcined SBA-15 by the incipient wetness impregnation method. Several techniques were used to characterize the samples. Skin hydration and transepidermal water loss were measured using Corneometer CM 825 PC and Tewameter 300 TM. Results showed that the structural properties of SBA-15Urea were similar to pure SBA-15, indicating that SBA-15 remained structured even after urea incorporation. Nitrogen physisorption data showed the volume and surface area of the pores in SBA-15Urea were much lower than those in SBA-15, demonstrating that urea was deposited inside the mesopores. In vivo moisturization studies revealed that SBA-15Urea was not able to reduce transepidermal water loss compared to the other products and control, while forming a non-occlusive surface film on the skin. We conclude that incorporation of urea in the pores of the inorganic SBA-15 matrix is a promising approach to enhancing its stability and providing a prolonged moisturizing effect.


INTRODUCTION
Chemical substances with moisturizing properties are widely used as cosmetic actives. These, urea (CO(NH 2 ) 2 ) is one the most valuable, due to its hydrophilic properties that promote water retention (Kapuscinska, 2014). Urea also known as carbamide is a colorless, odorless solid, highly soluble in water and is a natural constituent of the skin because it is produced by the human body as a result of metabolism of proteins and other compounds containing nitrogen.
It is practically non-toxic (LD 50 is 15 g kg -1 for rats) and it acts as a natural moisturizing factor (approximately 7wt.%) and the risk of allergic reactions is very low (Decaux et al., 2014;Kapuscinska, 2014). Despite its advantages, however, urea's thermal instability and a possible sensation of tingling or burning on sensitive skin can cause problems for cosmetic formulations mainly when urea is used in high concentration (Kapuscinska, 2014).
Formulations with up to 10wt% of urea are considered medicine, due to its keratolytic effects (Harivel, Besse, 2011;Sore, Hansenne, 2004). As urea readily cross the placental barrier, regulatory agencies such as the Food and Drug Administration (FDA), Cosmetic Ingredient Review (CIR) and Canada Health recommend that urea-based cosmetics and medicines are not used during pregnancy (Anvisa, 2010;Yamarik et al., 2005). Besides that, its stability decreases with increasing temperature. This is a problem for cosmetic formulations because depending on the reaction medium, temperature and pH conditions used, urea can be decomposed into ammonia and isocyanic acid (Eq. 1) that react with water by hydrolysis to form ammonia (Eq. 2) (Panyachariwat et al., 2014;Jiao at al., 2016): CO(NH 2 ) 2 (s) → NH 3 (g) + HNCO(g) Eq. 1 HNCO(g) + H 2 O(g) → NH 3 (g) + CO 2 (g) Eq. 2 In order to improve the moisturizing action of the urea and its thermal properties, and also prevent skin irritation, the incorporation of urea into an ordered mesoporous silica (OMS) matrix such as SBA-15 (amorphous SiO 2 named as Santa Barbara Amorphous-15) can be a good option. We believe that dispersal of urea into these nanopores would lead a gradual urea release from within the pores, with a consequently prolonged hydration action providing unique benefits for a cosmetic product. SBA-15 is a promising candidate due to unique properties, such as high specific surface area (above 700 m 2 g -1 ), pore size around 10 nm which can be expanded up to 50 nm with swelling agents, high pore volume (up to 2.5 cm 3 g -1 ) and a hydrophilic surface (Zhao et al., 1998a;Zhao et al., 1998b;Johansson et al., 2010;Pal, Bhaumik, 2013;Vinu et al., 2006;Mody et al., 2013;Alothman, 2012;Jardim et al., 2016).
Ordered mesoporous silica has attracted greatly interest for several applications, mainly for drug delivery and catalysis, and many research groups have been studying them (Jiao et al., 2016;Jardim et al., 2016;Martins et al., 2010;Pérez-Esteve et al., 2016;Braun et al., 2016;Yu et al., 2016;Ayad et al., 2016;Vieira et al. 2019;Takamori, et al. 2019). Most of these studies focused both in vitro and in vivo tests and concluded that these silicas have good thermal and chemical stability and biocompatibility and low toxicity. A variety of chemical modifications have been made to their external and internal surface, according to the need, to make them either hydrophobic or hydrophilic (Ayad et al., 2016;Tarn et al., 2013;da Silva et al., 2006). These are valuable properties for cosmetic applications.
In this work, urea was incorporated into SBA-15 by an incipient wetness impregnation method to yield an SBA-15Urea material. Cosmetic formulations containing the characterized material were then used for in vivo studies of skin moisturization. The main aim of this study was to demonstrate improved thermal stability of urea and to promote long-lasting skin hydration by gradual urea release from SBA-15.

SBA-15 Synthesis
SBA-15 was synthesized using a procedure reported by Zhao (Zhao et al., 1998b). TEOS was used as silica precursor and Pluronic ® P123 as the structuredirecting agent in an acid medium (HCl, 2.0 mol L -1 ). 4.0 g of Pluronic ® P123 were dissolved in 30.0 g of water and 120.0 g of HCl and stirred at room temperature to produce a homogeneous mixture and then 8.32 g of TEOS was added. The mixture was stirred vigorously at 40 °C for 24 h and heated in a Teflon-lined autoclave at 100 °C for 48 h. The SBA-15 product was separated from the solution by filtration, washed with deionized water, dried at 60 °C and calcinated at 540 °C for about 4 h under a nitrogen atmosphere. Soon after the atmosphere was switched to air, while keeping the furnace temperature at 540 °C for another three hours.

Preparation of SBA-15Urea
Urea was incorporated into SBA-15 by incipient wetness impregnation method, according to the following procedure: Urea (0.38 g) dissolved in ethanol (5.5 mL), was added to SBA-15 (1.5 g) suspended in ethanol (25.0 mL). The mixture was stirred at room temperature for 48 h. A rotary evaporator was then used to remove solvent. The material obtained was dried under vacuum over anhydrous calcium chloride and the sample was designated as SBA-15Urea. The schematic representation of the synthesis of this material is shown in Figure 1.
Small angle X-ray diffractometry of the samples was recorded on a Nanostar Bruker diffractometer by using CuKα radiation (λ = 0.15418 nm) operating at 40 kV and 30 mA, with a multivalent Hi-STAR bidimensional detector. The sample to detector distance was 65 cm, with q vector values ranging from 0.2 to 3.5 nm 1 . All the data were normalized by measuring time and being corrected for absorption effects.
Nitrogen adsorption-desorption isotherms were performed with a NOVA (Quantachrome) porosimeter. Prior thermal treatment at 60 ºC was made in order to dry the samples and the isotherms were taken at 77 K (N 2 ). The pore size distribution, pore volume and pore radius were calculated using the BJH (Barrett-Joyner-Halenda) method (Barret at al., 1951). The surface area was calculated using the BET (Brunauer-Emmett-Teller) method (Brunauer et al., 1938).
Infrared absorption spectra (IR) were recorded in the range 4000 -400 cm −1 in KBr pellets, using an IR Prestige-21 Fourier Transformation (FT) Spectrophotometer, Shimadzu. Thermogravimetric analysis and differential thermal analysis curves were obtained on a Shimadzu simultaneous TGA/DTA analyzer DTG-60H in the temperature range 25 to 900 o C with heating rate of 20 °C min -1 , under air atmosphere (100 cm 3 min -1 ) and using alumina crucibles with approximately 5.0 mg of the sample.
The morphology of SBA-15 was examined by scanning electron microscopy (SEM). The SBA-15 powder directly onto double-sided carbon tape on an aluminum stub was coated with gold and then SEM images were recorded on a JEOL microscope, model JSM 6610LV, operating with a Secondary Electron Imaging (SEI) detector.

Moisturizing cosmetic formulation containing SBA-15Urea
For the base gel 3.0 g of hydroxyethylcellulose (HEC), 1.6 g of ProTeg SL preservative (phenoxyethanol + ethyl hexyl glycerin) and 140.3 g of distilled water were weighed. HEC was added to the water, the mixture being stirred with a glass rod until dissolution. The mixture was then heated to 60 °C, when the gel became clear and viscous. The gel was then cooled to 45 °C in a water bath and the ProTeg SL preservative was added and the gel homogenized.
After the base gel cooled to room temperature, 75.1 g of gel were weighed and diluted with distilled water (25.1 g) to obtain 100.0 g of base gel.
The cosmetic moisturizer formulation, containing urea 0.1% by weight, was prepared by mixing the basic gel (100 g) and SBA-15Urea (0.5 g). The mixture was homogenized on a Vortex mixer for 15 minutes. The other products were prepared in a similar manner using the same amount of basic gel and 0.5 g of SBA-15 and 0.5 g of urea, obtaining BG+SBA-15 and BG+Urea, respectively ( Figure 2).

In vivo moisturize studies
This was a blind and randomized study to measure skin hydration and transepidermal water loss (Chaves at al., 2014). A sensorial test was carried out under controlled conditions (25±2 °C and 55±5 % relative humidity) where volunteers were instructed not to leave the testing site, not expose themselves to the sun, not consume caffeine, not use 24-hour moisturizing creams prior to testing. The 6 volunteers were between 18 and 40 years old. Study was approved by the Ethics Committee of the Federal University of São Paulo -Number 1.113.126.
Five different formulations were applied at 2.0 mg cm -2 in different areas of the forearm and named by 1, 2, 3, 4 and 5.
The measurements were performed immediately before application and after 1, 2, 3, 4 and 5 h (Chaves et al., 2014;Choi et al., 2013;Leite-Silva, et al., 2009). Hydration and transepidermal water loss data were analyzed by one way ANOVA with Tukey test, using Statistica 13.0 (TIBCO Software Inc., CA, USA) software to evaluate differences or any relationship between formulations and the obtained results.

SAXS analysis
Structural properties of ordered mesoporous silica with and without urea were evaluated by SAXS. Figure  3 illustrates the curves SAXS of SBA-15 and SBA-15Urea.
As evident in Figure 3, both diffractograms contain five intense diffraction peaks, indexable as (100), (110) (200), (210) and (300) reflections, which are characteristics of SBA-15 that has a highly ordered 2D hexagonal mesostructure, space group p6mm (Zhao et al., 1998b;Jardim at al., 2016). However, it was observed that the intensity of the diffraction peaks of SBA-15Urea decreased compared to pure SBA-15, indicating inclusion of urea in the SBA-15 pores. Table I shown the interplanar spacing (d (hkl) ) and lattice parameter (a (hkl) ) values. The lattice parameter values (Table I) of SBA-15Urea slightly decreased but this was expected due to the heating of silica for water removal before the incorporation process.

SEM analysis
The morphology of the SBA-15 sample analyzed by SEM (Figure 4) was similar to that observed by Zhao et al. (Zhao at al., 1998b). The SBA-15 powder shows many rope-like domains with uniform particle size of about 1 mm, while retaining the well-defined wheat-like macrostructure clusters of SBA-15.

Nitrogen physisorption analysis
Nitrogen adsorption-desorption isotherms and pore size distribution are shown in Figure 5 and Figure  6, respectively. The textural properties obtained by N 2 sorption of SBA-15 and SBA-15Urea are summarized in Table II. For both samples, the isotherms exhibited type-IV behavior and H1 hysteresis loops (Sing at al., 1985) typical of SBA-15 indicating that even after urea incorporation the hexagonal pore structure of SBA-15 was preserved, corroborating the SAXS results. The surface area and pore volume (Table II)

FTIR spectroscopy
The FTIR spectra of the samples are shown in Figure 7. In SBA-15 and SBA-15Urea samples, a broad absorption band characteristic of silanol groups (vSi-O-H) was observed at 3100 and 3700 cm -1 , which was attributed to OH stretching of silanol groups and adsorbed water. In the region around 1080-1210 cm -1 an intense broad band was attributed to Si-O-Si asymmetric stretching (v as (Si-O-Si)). The band at 960 cm -1 was attributed to the stretching vibration of the vSi-OH. The characteristic bands of the siloxanes groups noted around 800 and 470cm -1 were attributed to Si-O-Si symmetric stretching (v s (Si-O-Si)) and δ(O-Si-O), respectively. The band at 1632 cm -1 was assigned to a water bending vibration (Jardim at al., 2016;Prado et al., 2005).
In the FTIR spectrum of the SBA-15Urea sample (Figure 7), the main characteristic infrared bands of urea (v(NH 2 ) vibration at 3500-3000 cm -1 , v(CO) and δ(NH 2 ) vibrations around 1680-1620 cm -1 and v s (NH 2 ) and v as (NH 2 ) vibrations at 1150 and 1058 cm -1 , respectively) were overlapped with the OH stretching of silanol groups and adsorbed water, water bending vibration and siloxanes groups. Only the ν as (CN) vibration band, at 1460 cm -1 appeared to be resolved. Therefore we have not observed significant band shifts to indicate that silica interacts slightly with the urea (Keuleers at al. 1999).

TG/DTA analysis
The TG/DTA curves for all samples are shown in Figure 8 and Figure 9. The TG curve of urea ( Figure  8) show three events of weight loss that corresponds to thermal decomposition of urea ((NH 2 ) 2 CO (s) → NH 3 (g) + HNCO (g)) in the temperature range of 150-430 o C, which are accompanied by three endothermic peaks in the DTA curve. At 450 o C the decomposition products of urea were completely eliminated and no residue remained. An endothermic peak, at 141 o C, corresponds to the melting point of urea (Khattab at al., 1984). The TG curve of calcined SBA-15 (Figure 9) indicates that Pluronic ® P123 was completely removed. The two events of weight loss (from 25 to 700 o C) could be attributed to elimination of water (from 25 to 90 o C, 7% weight loss) and dehydroxylation of surface silanols (from 90 to 700 o C, 6% weight loss). The TG curve of SBA-15Urea (Figure 9) shows two events of weight loss, in the temperature range from 25 to 700 o C. The first event (25-130 o C), with 8% weight loss, corresponds to the loss of physically adsorbed water and the second event (130-700 o C), with 21% weight loss, was designated to the urea loss and some residual silanol. The percentage of mass loss obtained by TG is similar the nominal amount of urea used. Comparing the TG curves of urea and SBA-15Urea, it was noted that the onset temperature for urea decomposition was higher (173 o C) for pure urea than for SBA-15Urea (151 o C). However, the final temperature of thermal decomposition was greater for SBA-15Urea than for pure urea.

In vivo moisturization studies
A degree of whitening was observed in the in vivo moisturization studies after the natural drying of samples 3 (BG+SBA-15Urea) and 4 (BG+SBA-15) as shown in Figure 10. This phenomenon is due to the presence of SBA-15 in which individual particles (approximately 1 mm) are aggregate into blocks. Whitening was not observed for sample 2 (BG+Urea) which lacked SBA-15.  Figure 11 shows skin hydration measured at 0, 1, 2, 3, 4 and 5 hours (T0, T1, T2, T3, T4 and T5, respectively) after application of basic gel (BG), basic gel with urea (BG+Urea), basic gel with urea in ordered mesoporous silica SBA-15 (BG+SBA-15Urea) and basic gel with ordered mesoporous silica SBA-15 without urea (BG+SBA-15), as well as an untreated control area (C).
The variation of hydration values between T0 to T5 was not significant for BG (p = 0.3025) by one-way ANOVA or C (p = 0.1141). These results were expected, as BG is the basic gel vehicle without a moisturizing agent (urea) and C is the untreated control.
Tukey test showed that T0 was significantly different from T1, T2, T3, T4 and T5 for BG+SBA-15Urea and for BG+SBA-15. It is observed in Figure  12 that there is a reduction in hydration values from T0 to T1, T2, T3, T4 and T5 for both products. However, there was an increase in the degree of whitening over the experiment time period.
Comparing the average hydration values ( Figure  12) from T1 to T5 using one-way ANOVA, p-value was 2.29 x 10 -17 indicating a high significantly difference between the products, considering p < 0.05. By means of Tukey test, it was possible to separate the products into groups (Table III).   The hydration state of the skin surface is commonly evaluated by an indirect method based on the measurement of electrical conductivity of percentage of water present in the horny layer (Tagami, 1995). This is the method used by the Corneometer ® . The presence of whitening areas at the forearm squares with samples containing SBA-15 indicates film formation, which might have masked the correct reading of the degree of hydration of the Corneometer ® . A standardized wash procedure should be performed on the skin to obtain a correct reading with the Corneometer ® . In addition, the skin surface hydration should be measured within the next day as recommended by Imokawa for nontransparent cosmetic (Imokawa, 1989). Although most studies required an overnight washout period prior to measurement, this could have had an influence on the results (Lodén, 2012).
Since the variation found for transepidermal water loss was not statistically significant during five hours of measurement for each product and control, the average TEWL between T0 to T5 were calculated ( Figure 14).
Average TEWL of each product (Figure 14) was compared using one-way ANOVA. The results showed significant differences (p = 0.014) between the products. Tukey test showed clusters of the products into three groups: 1(BG+SBA-15 and BG), 2 (BG+Urea and C) and 3 (BG+SBA-15Urea). BG+SBA-15Urea was effective in reducing transepidermal water loss compared to the other products and control.
A review on the clinical effectiveness of moisturizers suggested that urea-based preparations may be preferable as a first-line treatment, but there was an unmet need for well-powered comparisons between individual moisturizers (Lindh, 2015).
Many papers have shown that urea increase skin moisturization but does not reduce TEWL corroborating this study. In two small-randomized controlled trials, urea-containing moisturizers were not significantly superior to any other treatment in reducing TEWL [Lodén, at al., 1999;Hagströmer, at al. 2001).

CONCLUSIONS
The synthesis of the SBA-15 matrix was reproducible and the estimated values of surface area, pore size and pore volume were similar to those reported in the literature. Urea was incorporated into SBA-15 at approximately 21 wt.%, and results indicated that it was deposited inside the mesopores and caused little change to the structural integrity of SBA-15.
The presence of whitened regions on the forearm where SBA-15 and SBA-15Urea were applied suggests that these materials formed a film on the skin surface, probably due to the nanotubes being aggregated. This film was not occlusive, however, no change in transepidermal water loss was observed on treated areas over five hours after application. On the other hand, skin hydration readings at these times for the SBA-15 and SBA-15Urea treated skin were only 50% of basal or control values and we believe that the film prevented good contact of the corneometer probe with the skin surface, resulting in measurements that were not truly indicative of the underlying skin properties. This scenario is likely, as the skin below the film is expected to be well hydrated in the presence of urea.
With their five hours retention on the skin surface, the SBA15 materials used in this work are promising vehicles for providing long-lasting moisturizing, in combination with urea or other raw materials. Future studies by this group will investigate new methods of decreasing aggregation of the nanotubes, to promote better dispersion of SBA15 in the formulation.
Estado de São Paulo (FAPESP) and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq). We owe special thanks to Dr Jeff Grice (a native English speaker) from University of Queensland, Australia, for his careful and excellent English language revision of our manuscript.