- Email: [email protected]
Preparation of nanocrystalline titania thin films by using pure and water-modified supercritical carbon dioxide
J. of Supercritical Fluids 117 (2016) 289–296
Contents lists available at ScienceDirect
The Journal of Supercritical Fluids journal homepage: www.elsevier.com/locate/supflu
Preparation of nanocrystalline titania thin films by using pure and water-modified supercritical carbon dioxide Marie Sajfrtová a,∗, Marie Cerhová a, Vladislav Dˇrínek a, Stanislav Daniˇs b, Lenka Matˇejová c a b c
Institute of Chemical Process Fundamentals of the ASCR, v.v.i., Rozvojová 135, 165 02 Prague 6, Czech Republic Department of Condensed Matter Physics, Faculty of Mathematics and Physics, Charles University in Prague, Ke Karlovu 5, 121 16 Prague 2, Czech Republic Institute of Environmental Technology, VSˇ B-Technical University of Ostrava, 17. listopadu 15/2172, 708 33 Ostrava-Poruba, Czech Republic
a r t i c l e
i n f o
Article history: Received 7 April 2016 Received in revised form 10 July 2016 Accepted 11 July 2016 Available online 12 July 2016 Keywords: Titania thin film Anatase Supercritical carbon dioxide Subcritical water Crystallization Microstructure
a b s t r a c t Processing by pure and water-modified (30 wt.%) supercritical carbon dioxide and by subcritical water were utilized for the direct preparation of highly pure TiO2 anatase thin films without any subsequent thermal treatment. One step processing was compared with the two or three step processing combining pure and modified CO2 . The effect of temperature (40–150 ◦ C) and the amount of CO2 (100–200 g) passed through the high-pressure column on the (micro)structure and the purity of TiO2 thin films were examined at pressure of 30 MPa. Prepared thin films were characterized with respect to the structural properties and purity by Raman spectroscopy. The most promising thin films were analysed with respect to microstructural properties by means of X-ray diffraction to determine the phase composition, the crystallite-size and the crystallite-size distribution. High temperature had a positive effect on the crystallization as well as the purity of TiO2 thin films during the one step and multi-step processing. When TiO2 thin films were exposed to water-modified supercritical CO2 and temperature of 150 ◦ C under pressure of 30 MPa, the desired crystalline structure of anatase was obtained. The anatase crystallites growth was mainly influenced by the presence of water. Anatase crystallites sizes of 2–12 nm were obtained depending on the processing method on both investigated substrates (soda-lime glass and monocrystalline Si) on which the TiO2 thin films were deposited. Using one step or multi-step processing by water-modified supercritical CO2 any undesirable effects such as Na+ ions diffusion from the soda-lime glass substrate to the one-layer TiO2 film, having negative effect on crystallization of anatase, did not take place. The universality of developed processing by pure and water-modified supercritical CO2 for preparation of TiO2 anatase thin films was successfully confirmed for two different substrates. © 2016 Elsevier B.V. All rights reserved.
1. Introduction TiO2 (Titanium dioxide, titania) belongs thanks to its excellent photochemical performance and other photo-induced phenomena under UV light (with wavelength < 365 nm) among materials under keen scientific interest. It has been explored in a form of thin films in many application areas; as sensing films of gas sensors, coatings for self-cleaning surfaces or with antimicrobial activity, an electrode material or photocatalyst promising in waste water and air treatment technologies. Titania is a semiconductor with the band gap energy of 3.2 eV, 3.0 eV and 3.1 eV that belong to its crystal structure of anatase, rutile a brookite, respectively. Despite
∗ Corresponding author. E-mail addresses: [email protected] (M. Sajfrtová), [email protected] (M. Cerhová), [email protected] (V. Dˇrínek), [email protected] (S. Daniˇs), [email protected] (L. Matˇejová). http://dx.doi.org/10.1016/j.supflu.2016.07.007 0896-8446/© 2016 Elsevier B.V. All rights reserved.
the fact that anatase shows higher band gap energy than rutile or brookite, anatase exhibits higher photoactivity. The reason of this performance is a different structure of energy bands; the energy of anatase conductive band is higher than that of rutile or brookite [1,2]. Practically it means the enhanced reduction capability of excited electrons which are needed for the formation of free radicals which finally participate in the degradation e.g. of organic pollutants [3–5]. Since microstructural properties such as the crystallite size on nanosize level and shape, crystallite size distribution, phase composition, oxygen vacancy, microstrain etc. crucially affect photo-electrochemical response and photocatalytic performance of titania thin films [6], e.g. via the effect on their electronic and optical properties, it is very important to know and control all preparation and processing steps to avoid undesirable effects within microstructure of thin films. Commonly used method for the preparation of pure and crystalline TiO2 thin films is calcination [7,8]. However, this approach
290
M. Sajfrtová et al. / J. of Supercritical Fluids 117 (2016) 289–296
has some disadvantages such as excessive sintration, crystallite growth or recrystallization. Moreover, during calcination the microstructural properties such as crystallite-size and phase composition, having the influence on the photocatalytic activity, cannot be controlled. In last years this standard thermal processing was overcome by supercritical carbon dioxide drying combined with thermal processing. It was found that the pre-treatment using supercritical carbon dioxide proposed for lowering the process temperature of sol-gel derived metal oxide thin films as well as powders helps to enlarge surface area significantly and improve electronic properties showing enhanced photocatalytic performance [9–11]. Wei et al. [12] engaged in the effect of the post-treatment of mesoporous crystalline TiO2 thin films by a supercritical CO2 (scCO2 ) on the photovoltaic performance. In their study, a serie of TiO2 thin films composed of mesopores and nanoparticles were prepared by a spin-coating process and then statically treated by scCO2 for 30 min at pressures of 34.5 MPa and 58.6 MPa and at temperatures ranging from 60 to 100 ◦ C. The films were calcined at temperature of 400 ◦ C for 3 h before the characterization and solar cell fabrication. The mesopores in the TiO2 thin film without the scCO2 treatment collapsed after calcination at temperatures above 400 ◦ C, while the films treated by scCO2 exhibited an enhanced thermal stability and well-preserved porosity. Film processing using supercritical CO2 accomplished with a batch-type reaction system was compared with conventional calcination by Asai et al. [13]. The TiO2 precursor films on soda-glass substrates prepared by sol-gel coating using Ti-alkoxide solution were converted to crystalline TiO2 (anatase with 12 nm crystal size) successfully by treatment in scCO2 at a fluid pressure of 15 MPa and a substrate temperature of 300 ◦ C whereas no crystallization occurred by conventional heat treatment at 400 ◦ C. Besides that, several works reported on utilization of pressurized water and supercritical/subcritical methanol in a flow regime which resulted in nanocrystalline titania powders with significantly enlarged surface area and purity comparable to usually thermally treated titania powders [14,15]. It was shown that utilization of this solvent combination leads to direct crystallization of titania, namely by the effect of pressurized water [14–17]. The motivation of this work was to design the improved supercritical fluid process which can lead to purification and direct crystallization of TiO2 thin films without any subsequent thermal treatment. The processing by pure or water-modified supercritical CO2 in a flow regime was investigated and compared with subcritical water processing. All the experiments were performed over precursor titania gel thin films prepared by reverse micelles assisted sol-gel method, using hardly removable nonionic surfactant Triton X-114, however, forming uniform nano-domains. The effect of various processing conditions (i.e. temperature, pressure, volume and flow rate of solvents) as well as the type of substrate on microstructure and purity of TiO2 thin films was thoroughly studied by means of Raman spectroscopy, X-ray diffraction, and contact angle measurements.
was treated ultrasonically for 30 min before utilization to remove bubbles. For deposition of the films two types of substrate were used: soda-lime glass and monocrystalline Si (non-diffracting substrate). 2.2. Sol preparation and deposition of thin films Gel titania thin films were deposited on substrates (soda-lime glass and monocrystalline Si) by dip-coating method, using the sol prepared by sol-gel process controlled within reverse micelles of non-ionic surfactant Triton X-114 in cyclohexane. The molar composition of titania sol was following; cyclohexane: Triton X-114:water:Ti(OCH2 (CH3 )2 )4 = 11:1:1:1 [14]. In a shortcut, proper amounts of cyclohexane, Triton X-114 and water were mixed and vigorously stirred for 15 min for homogenization and formation of reverse micelles. Then, titanium (IV) isopropoxide was injected to micellar solution under vigorous stirring. After addition of isopropoxide the sol was stirred for next 20 min. A prepared sol was left standing in a closed glass bottle for 4 h to stabilize. Ultrasonically cleaned and dried soda-lime glasses and monocrystalline Si were dipped into the sol by using a dip-coater 4 idLab. The deposition conditions were following: the immersion velocity 15 cm min−1 , the delay in the sol 30 s, the emergence velocity 6 cm min−1 . After deposition the substrates with gel thin films were left on air overnight and then were processed by proper investigated method using pure and/or water-modified scCO2 or subcritical water. 2.3. Supercritical fluid crystallization (SFC) Air dried glasses with deposited gel thin films were fixed in a steel holder and immersed in the high pressure column (volume 150 ml; inner diameter 30 mm) filled in the bottom part with glass beads serving as solvent flow distributors. The column was connected with stainless steel capillaries and placed into the airconditioned box. The SFC experiments were carried out in Spe-ed SFE apparatus (Applied Separations, USA) whose schema is shown in Fig. 1. Carbon dioxide was sucked from a pressure container using a high-pressure pump cooled by water to 5 ◦ C. ScCO2 entered the lower end of the column at pressure of 30 MPa. The temperature of scCO2 was maintained using a hot air ventilator in an oven. In experiments with modified CO2 , water was supplied at a constant flow rate by a high-pressure LCP 4020.3 (ECOM s.r.o.) and mixed with CO2 before entering the column to reach concentration in scCO2 30 wt.%. The solution flowing from its upper end was expanded to the ambient pressure in a heated micrometer valve and the extract was collected in empty glass vial at the ambient temperature. The quantity of the gaseous CO2 leaving the trap and its flow rate was measured using a gas meter. The CO2 flow rate was adjusted to 0.8 g min−1 using the micrometer valve. Experimental conditions and design were changed as shown in Table 1.
2.1. Materials
2.3.1. One step processing Experiment with pure CO2 was carried out at constant pressure of 30 MPa, temperature of 150 ◦ C and solvent amount of 100 g. When modified CO2 was used, different temperature (40–150 ◦ C) and pressure (10–30 MPa) were applied.
Cyclohexane (p.a., Lachner, Neratovice, CR), Triton X-114 ((1,1,3,3,-Tetramethylbutyl)phenyl-polyethylene glycol, SigmaAldrich, USA), Ti(OCH2 (CH3 )2 )4 (Titanium(IV) isopropoxide, SigmaAldrich, USA) and distilled water were used for the preparation of precursor solution of TiO2 gel films. Carbon dioxide (>99.9%) for supercritical fluid processing was purchased from Linde Technoplyn (Prague, CR). Distilled water for subcritical water processing
2.3.2. Multi-step processing Combination of several steps with pure and water-modified scCO2 was tested. Order of individual steps and temperature varied as shown in Table 1. In the first step, the pure or water-modified CO2 at the temperature of 40 or 150 ◦ C was used. After 100 g of CO2 flowed through the column the conditions were changed and experiment continued. Second step was finished when 100 g of CO2
2. Materials and methods
M. Sajfrtová et al. / J. of Supercritical Fluids 117 (2016) 289–296
291
Fig. 1. Illustration of experimental set-up: (1) CO2 pressure bottle, (2) bottle with modifier, (3) cooling bath, (4) high pressure pumps, (5) air-heated box, (6) high pressure column, (7) glass beads, (8) glasses with deposited TiO2 thin films, (9) micrometer valve, (10) separator and (11) gas meter. Table 1 Experimental design and conditions of supercritical fluid crystallization. Experiment Nr.
Experimental design (temperature, ◦ C/solvent)
1 2 3 4 5 6 7 8 9 10 11 12
150 40 100 150 150a 40 40 150 40 40 150 40
1st step
a
2nd step CO2 CO2 + W CO2 + W CO2 + W CO2 + W CO2 CO2 CO2 CO2 + W CO2 CO2 CO2 + W
40 150 150 150 150 150 150
focused to a spot of 0.6 mm in diameter. The incident energy was minimized as necessary to avoid thermal effects as a consequence of the laser irradiation. Spectral resolution of the instrument was 2 cm−1 and the spectrometer was calibrated using crystalline silicon standard with line at 520 cm−1 .
3rd step
CO2 + W CO2 + W CO2 + W CO2 CO2 + W CO2 + W CO2 + W
150 150 150
CO2 CO2 CO2
at 10 MPa.
was consumed. In order to remove residual moisture from the films surface the third step followed in some experiments until 200 g of pure CO2 flowed at 150 ◦ C. The CO2 flow rate at this scCO2 drying was adjusted to 1.6 g min−1 . 2.4. Subcritical water crystallization The experiments with subcritical water (SW) were performed in the same apparatus as the SFC. Heated water was pumped to highpressure column by high-pressure LCP 4020.3 (ECOM s.r.o.) with a set flow rate of 2 ml min−1 . After 50 ml of water flowed through the expansion valve, the experiment was terminated.
2.5.2. X-ray diffraction XRD investigations were provided on a Panalytical MPD diffractometer in the parallel beam geometry. X-rays were generated using Cu-xray tube (40 kV/35 mA). Primary beam was conditioned by means of x-ray mirror and diffracted beam have been collected using proportional gas-filled detector equipped with parallel-plate collimator. X-ray patterns were measured at constant angle of incidence of x-rays (1◦ with respect to the layer surface). Diffraction angle, 2\, varied within the range 10–120◦ Crystal structure, crystallite-size as a median of the log-normal crystallite-size distribution and log-normal crystallite-size distribution were evaluated using the Rietveld fitting program for thin film analysis − MSTRUCT, using the sphere model of the crystallites [18]. 2.5.3. Contact angle measurements The device for the measurement of contact angles consisted of a microscope with quadruple lens, table with a micro-displacements and camera Sony SSC-C370. The image was processed using software Drop Analysis (Ecole Polytechnique Federale de Lausanne). 3. Results and discussion 3.1. Structure and purity of TiO2 thin films
2.5. Characterization of thin films The prepared thin films were characterized with respect to their microstructural properties by Raman spectroscopy as well as Xray diffraction (XRD). Raman spectroscopy helped to identify the presence of anatase, the desired crystal structure of TiO2 and the presence of impurities from used organic precursors. Thin films with confirmed crystallinity was subsequently subjected to XRD, which allowed to determine the crystallite size, crystallite-size distribution and the specific composition of the crystalline phases. 2.5.1. Raman spectroscopy Raman spectra were acquired on a dispersive Nicolet Almega XR spectrometer equipped with Olympus BX 51 microscope. Excitation laser source (473 nm) with an incident power of 50 mW with 256 expositions with 0.5 s duration was used. The laser beam was
Raman spectroscopy proved to be a very sensitive technique for determining the crystallinity and purity of prepared thin films. Raman spectra and the appearance of TiO2 thin films deposited on soda-lime glasses processed by scCO2 under various conditions mentioned in Table 1 and subcritical water are compared in Figs. 2 and 3. Signal intensities in the range of 3300–700 cm−1 of Raman shift indicates the occurrence of undissolved surfactant and other organic impurities, vibrations in the range of 700–100 cm−1 of Raman shift (the grey field) show the presence of TiO2 anatase peaks. 3.1.1. One step processing Fig. 2 compares Raman spectra of TiO2 thin films processed in one step by subcritical water and pure or water-modified scCO2 at different temperatures and pressures.
292
M. Sajfrtová et al. / J. of Supercritical Fluids 117 (2016) 289–296
Fig. 2. Raman spectra and appearance of TiO2 thin films deposited on soda-lime glasses processed at 30 MPa using one step processing by: subcritical water at 150 ◦ C (SW), pure scCO2 at 150 ◦ C (1), water-modified scCO2 at 40 ◦ C (2), 100 ◦ C (3) and 150 ◦ C (4) and at 10 MPa and 150 ◦ C (5). Valence and deformation vibrations: ( C H, 3073 cm−1 ), ( C H, 2913 cm−1 ), (C C, 1613 cm−1 ), ␦(H C H, 1465 cm−1 ) a (C C aliphatic, 1288 cm−1 ), region of anatase peaks: 700–100 cm−1 .
at 30 MPa and 150 ◦ C (Exp. 1) were
Films treated with pure scCO2 amorphous as expected. Moreover, the presence of organic carbon was proved in these thin films. The surfactant bands which are attributed in Raman spectra to valence and deformation vibrations ( C H, 3073 cm−1 ), ( C H, 2913 cm−1 ), (C C, 1613 cm−1 ), ␦(H C H, 1465 cm−1 ) and (C C aliphatic, 1288 cm−1 ) indicate that the solubility of surfactant and other organic residues of the sol production in pure scCO2 is very poor. Contrary to that in Raman spectra of thin films processed by subcritical water (Exp. SW) at temperature of 150 ◦ C and pressure of 30 MPa only bending C H peak, actually G (graphitic-like) peak, in 1106 cm−1 and negligible valence vibrations ( C H, 3080 cm−1 ) and ( C H, 2935 cm−1 ) were observed. It demonstrates higher solubility of surfactant in subcritical water. The small peaks in the range of 100–700 cm−1 of Raman shift indicate the beginning crystallization of anatase in subcritical water as well. However, this solvent itself caused unwanted removal of the film from the soda-lime glass substrate, thus, the scCO2 was enriched by water to combine advantages of both processing. When 30 wt.% of water was added to the scCO2 as modifier at low temperature of 40 ◦ C (Exp. 2), no crystallization of TiO2 was observed. Increasing temperature to 100 ◦ C (Exp. 3) caused the formation of anatase crystallites and better solubility of organic compounds as evident from the characteristic anatase bands at 168, 418, 532 and 653 cm−1 and decreased stretching C H vibrations at 2800–3050 cm−1 . Increased bending C H vibration at ∼1620 cm−1 (D-peak) can be attributed to partial graphitization of the surfactant residues in thin film. At the highest tested temperature of 150 ◦ C (Exp. 4) almost all the surfactant and organic substances were removed from thin film and the anatase crystal structure was achieved. Only thin films prepared at 100 ◦ C and 150 ◦ C showed the compact layer on the soda-lime glass surface. Reducing the pressure bellow 30 MPa at 150 ◦ C (Exp. 5) decreased the solubility of the surfactant and associated purity of thin film. Increased bending C H vibration at ∼1620 cm−1 proves the graphitization of the surfactant residues. Although the anatase crystallized, the partial elution of TiO2 film from glass substrate was observed as evident in the photograph in Fig. 2.
Fig. 3. Raman spectra and appearance of TiO2 thin films deposited on soda-lime glasses processed at 30 MPa using multi-step processing by: pure scCO2 at 40 ◦ C followed by water-modified scCO2 at 40 ◦ C (6), pure scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (7, 10*), pure scCO2 at 150 ◦ C followed by watermodified scCO2 at 150 ◦ C (8, 11*), water-modified scCO2 at 40 ◦ C followed by pure scCO2 at 150 ◦ C (9) and water-modified scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (12*), *with subsequent drying with pure scCO2 at 150 ◦ C. Valence and deformation vibrations: ( C H, 3073 cm−1 ), ( C H, 2913 cm−1 ), (C C, 1613 cm−1 ), ␦(H C H, 1465 cm−1 ) a (C C aliphatic, 1288 cm−1 ), region of anatase peaks: 700–100 cm−1 . (For interpretation of the references to colour in the text, the reader is referred to the web version of this article.)
3.1.2. Multi-step processing The effect of experimental design, namely the different temperature and the solvent composition (pure and water-modified scCO2 ) combined in two or three steps on crystallinity and purity of TiO2 thin films is shown in Raman spectra in Fig. 3. Anatase was detected in all the thin films prepared by the multistep processing, however, the conditions and order of individual steps affected the purity of the films. Although no crystallization took place in the one step processing of thin films using pure scCO2 or water-modified scCO2 at 40 ◦ C (see Fig. 2), when both solvents were combined the anatase crystallized (Fig. 3, Exp. 6). Nevertheless, the film contained a lot of impurities as evident from the presence of valence and deformation vibrations ( C H, 3071 cm−1 ), ( C H, 2938 cm−1 ), (C C, 1598 cm−1 ), ␦(H C H, 1468 cm−1 ), (C C aliphatic, 1286 cm−1 ) and bending C H vibration (1620 cm−1 ) in spectrum of this sample. It is evident that the subsequent increase in temperature from 40 ◦ C to 150 ◦ C of water-modified scCO2 in the second step has a positive effect on the crystallization as well as the purity of the sample (Exp. 7, 10 and 12). Using high density CO2 (30 MPa and 40 ◦ C) in the first step of processing, there may be a partial cleaning of films from non-polar organic impurities. Water and high temperature in the second step of processing acted as a crystallization initiator. The experiment with reverse solvent arrangement when the pure scCO2 in the first step was replaced by the water-modified scCO2 (Exp. 9) and the experiments conducted at 150 ◦ C (Exp. 8, 11) failed to effectively remove the organic contamination from films, however, were accompanied by successful crystallization of anatase. The appear-
M. Sajfrtová et al. / J. of Supercritical Fluids 117 (2016) 289–296
293
Fig. 4. XRD spectra of TiO2 thin films deposited on soda-lime glasses processed at 30 MPa using one step processing by water-modified scCO2 at 100 ◦ C (3) and 150 ◦ C (4); and multi-step processing by: pure scCO2 at 40 ◦ C followed by water-modified scCO2 at 40 ◦ C (6), pure scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (7, 10*), pure scCO2 at 150 ◦ C followed by water-modified scCO2 at 150 ◦ C (8, 11*), water-modified scCO2 at 40 ◦ C followed by pure scCO2 at 150 ◦ C (9) and water-modified scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (12*). *with additional drying with pure scCO2 at 150 ◦ C. Dashed lines represent crystal structure of anatase.
Fig. 5. TiO2 anatase crystallite-size of thin films on soda-lime glasses processed at 30 MPa using one step processing by water-modified scCO2 at 100 ◦ C (3) and 150 ◦ C (4); and multi-step processing by: pure scCO2 at 40 ◦ C followed by water-modified scCO2 at 40 ◦ C (6), pure scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (7, 10*), pure scCO2 at 150 ◦ C followed by water-modified scCO2 at 150 ◦ C (8, 11*), water-modified scCO2 at 40 ◦ C followed by pure scCO2 at 150 ◦ C (9) and water-modified scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (12*). *with additional drying with pure scCO2 at 150 ◦ C.
ance of all thin films with crystallized anatase in the multi-step processing is shown in Fig. 3 as well. As obvious from Fig. 3 thin films create a compact layers without any significant inhomogeneity and/or defects on soda-lime glass surface and show violet up to green colour indicating the different anatase crystallite-size and thickness of thin films. 3.2. Crystal structure and microstructure of thin films determined from XRD XRD analysis of all prepared titania thin films confirmed the presence of anatase crystal structure as evident from the typical bands of the anatase structure in Fig. 4. Information on the crystallite-size and crystallite-size distribution of nanocrystalline TiO2 anatase play an important role in the physico-chemical behaviour of the material in terms of chemical and phase stability and chemical reactivity [19]. The comparison of anatase crystallite-sizes determined for TiO2 thin films deposited
on soda-lime glasses prepared by pure and modified scCO2 under different experimental design and conditions is summarized in Fig. 5. The size of anatase crystallites of thin films deposited on sodalime glasses varied according to the processing procedure and applied temperature in the range of 1–12 nm. Using the one step processing with water-modified scCO2 the anatase crystallite-size was positively affected by increasing temperature (see Fig. 5, Exp. 3 and 4). The anatase crystallite-size increased from ∼3 to ∼6.3 nm. When experiments began with pure scCO2 in multi-step processing and were followed with water-modified scCO2 , only smaller anatase crystallites under ∼4 nm were formed. Change of the order of solvents in multi-step processing, when the water-modified scCO2 at 40 ◦ C was used in the first step (see Fig. 5, Exp. 9), resulted in doubling the size of anatase crystallites. Due to the presence of water the crystallization of anatase was initiated already in the first step of processing and the crystal growth could continue in the second step during slow removal of moisture, moreover, more
294
M. Sajfrtová et al. / J. of Supercritical Fluids 117 (2016) 289–296
intensively as a consequence of higher temperature of 150 ◦ C. The positive effect of the prolonged action of water and of high temperature on crystal growth was also confirmed in Exp. 12, where the largest anatase crystallites (∼12 nm) were obtained. 3.3. Aspects related to the mechanism of crystallization of TiO2 thin films in water-modified supercritical CO2 Principally three procedures were compared; (1.) processing by pure scCO2 , (2.) processing by water-modified scCO2 (combined with processing by pure scCO2 in multi-steps) and (3.) processing by subcritical H2 O. The effectiveness of individual procedures differed a lot concerning the crystallization of single-layer TiO2 thin films, since the crystallization of TiO2 thin films is affected significantly by solubility of the nonionic surfactant Triton X-114 used for sol-gel synthesis in individual media. It is worth to mention that on one hand, Triton X-114 belongs among surfactants enabling to form reverse micelles and, thus, the preparation of TiO2 crystallites in nano-size due to limited hydrolysis and polycondensation reactions taking place in the core of the micelles, creating Ti O Ti O Ti bonds in the amorphous gel matrix. On the other hand, it is a complicated molecule from organic point of view, composed of polar and nonpolar part: (1,1,3,3-tetramethylbutyl)phenyl-polyethylene glycol. This chemical composition makes the surfactant Triton X-114 hardly totally degradable even by calcination, therefore higher calcination temperature (∼400◦ C) and more deposited layers were needed to reach crystallization of TiO2 anatase thin films [20]. Moreover, the calcination of TiO2 thin films deposited on soda-lime glasses is accompanied with a negative diffusion of Na+ ions [21,22], making the crystallization of TiO2 thin films more complicated from crystallographic point of view. Therefore, just the dissolution of the surfactant Triton X-114 in medium/mixture of media possessing optimal dissolving power opens up the way not only for the low-temperature preparation of anatase thin films, but also for its recovery. In our study the processing by pure scCO2 and the processing by subcritical H2 O were firstly examined separately. It was revealed that neither scCO2 , nor subcritical H2 O shows the sufficient dissolving power to remove sufficiently the used surfactant from thin films. Each medium dissolved only part of the surfactant. On the other side, processing by subcritical water resulted in crystallization of TiO2 thin films contrary to processing by scCO2 . Therefore the processing by water-modified scCO2 at various experimental conditions and experimental design was investigated to reveal the best/optimal conditions for maximum dissolution/removal of the
Fig. 6. XRD spectra of TiO2 thin films deposited on monocrystalline Si processed at 30 MPa using one step processing by water-modified scCO2 at 150 ◦ C (4); and multistep processing by: pure scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (7, 10*), water-modified scCO2 at 40 ◦ C followed by pure scCO2 at 150 ◦ C (9) and water-modified scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (12*). *with additional drying with pure scCO2 at 150 ◦ C. Dashed lines represent crystal structure of anatase.
surfactant. As it was mentioned above, the dissolution of the nonionic surfactant Triton X-114 is a crucial initial step affecting further the crystallization of TiO2 thin films. The dissolution/removal of the surfactant leads to revelation of the Ti O Ti bonds, reacting with subcritical water (i.e. in our tests with liquid water under pressure at relatively mild temperatures of 40 ◦ C or 150 ◦ C). Pressurized water attacks and cleaves the Ti O Ti bonds, allowing the atomic rearrangement and the crystallization of TiO2 anatase thin films. In the presence of pressurized water, which was always in excess during the processing, the rapid nucleation and the crystallite growth take place. While the dissolution of the surfactant strongly depends on both used temperature and pressure which must be optimal for water-modified scCO2 to dissolve the surfactant sufficiently, the nucleation and the anatase crystallites growth are affected mainly by applied temperature (namely temperature of water) and the exposure time to water. Our findings can be supported partially by already reported works from Imai et al. [23,24] and Uchiyama et al. [25] who studied different, however, partially similar systems. Imai et al. [23,24] observed the crystallization of alkoxide-derived amorphous TiO2 gel films during exposure to water vapour produced during the heat treatment to anatase films at low temperature of 180 ◦ C and they suggested the similar crys-
Fig. 7. TiO2 anatase crystallite-size of thin films deposited on monocrystalline Si processed at 30 MPa using one step processing by water-modified scCO2 at 150 ◦ C (4); and multi-step processing by: pure scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (7, 10*), water-modified scCO2 at 40 ◦ C followed by pure scCO2 at 150 ◦ C (9) and water-modified scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (12*). *with additional drying with pure scCO2 at 150 ◦ C.
M. Sajfrtová et al. / J. of Supercritical Fluids 117 (2016) 289–296
295
Fig. 8. Contact angles of TiO2 thin films deposited on soda-lime glasses processed at 30 MPa using one step processing by: subcritical water at 150 ◦ C (SW), pure scCO2 at 150 ◦ C (1), water-modified scCO2 at 40 ◦ C (2), 100 ◦ C (3) and 150 ◦ C (4) and at 10 MPa and 150 ◦ C (5), and using multi-step processing by: pure scCO2 at 40 ◦ C followed water-modified scCO2 at 40 ◦ C (6), pure scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (7, 10*), pure scCO2 at 150 ◦ C followed by water-modified scCO2 at 150 ◦ C (8, 11*), water-modified scCO2 at 40 ◦ C followed by pure scCO2 at 150 ◦ C (9) and water-modified scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (12*), *with additional drying with pure scCO2 at 150 ◦ C. Appearance of the drop on the surface of selected TiO2 thin films is shown.
tallization mechanism as in our study. Water as a vapour was the crystallization agent. Uchiyama et al. [25], who studied the preparation of TiO2 thin films from titanium alkoxide-derived solutions containing acetylacetone, agreed with the mechanism of crystallization from Imai et al. [23,24]. In addition, they reported on the effect of higher content of water present in the titanium alkoxidederived solutions containing acetylacetone. They reported that larger amount of water present in the solutions and, therefore, subsequently larger amount of water vapour present during the heat treatment provides higher nucleation rate, resulting in the generation of a larger number of small crystallites (in their case of smaller TiO2 anatase crystallites). Within our study the following conclusions about the processing of reverse micelles-derived TiO2 gel thin films by water-modified scCO2 may be stated; (1.) pressurized hot water as a part of the water-scCO2 mixture dissolves the part of the used surfactant and causes the crystallization of TiO2 thin films, (2.) scCO2 serves as a carrier of the pressurized hot water and dissolves a part of the surfactant insoluble in the pressurized hot water, and (3.) longer exposure of the revealed Ti O Ti bonds to pressurized water and higher temperature promotes the TiO2 anatase crystallite growth. 3.4. Effect of substrate on crystal structure and microstructure of TiO2 thin films
The crystal structure and microstructural results from one step and multi-step processing by pure/water-modified scCO2 of TiO2 thin films (composed of one layer) deposited on monocrystalline Si are shown in Figs. 6 and 7. Concerning the crystal structure, it is evident that the substrate had no effect on the crystal structure and TiO2 anatase crystallized in all investigated runs on monocrystalline Si as well. It reveals that during developed processing by pure and water-modified scCO2 the negative diffusion of Na+ ions from soda-lime glass to TiO2 films does not occur due to low processing temperatures, and it results in direct crystallization of TiO2 anatase thin films (composed of one layer) on both types of substrates. Concerning the anatase crystallite-size, one step processing by water-modified scCO2 at 150 ◦ C of TiO2 films deposited on monocrystalline Si (Fig. 7, Exp. 4) led to anatase crystallites of similarly small, but not so uniform size as on soda-lime glass (Fig. 5, Exp. 4). In contrast to anatase thin films deposited on soda-lime glasses the processing of films deposited on monocrystalline Si by water-modified scCO2 at 40 ◦ C in the first step led to smaller anatase crystallites (Fig. 7, Exp. 9) and the anatase crystallites growth was not influenced by additional drying with pure scCO2 at 150 ◦ C (Fig. 7, Exp. 7 and 10*). The multi-step processing by watermodified scCO2 in both steps and with additional drying with pure scCO2 at 150 ◦ C (Fig. 7, Exp. 12) led to anatase crystallites of similar size as well as the size-distribution as on soda-lime glasses. 3.5. Surface wettability
The most promising found experimental design and conditions (i.e. Exp. 4, 7, 9, 10 and 12, see Table 1) were investigated with respect to crystal structure and microstructure for TiO2 thin films deposited on monocrystalline Si as well. The effect of substrate on quality of titania thin films was keenly investigated, since it was revealed in some previous studies [21,22] that the type of substrate affects significantly the crystallization of TiO2 films. The diffusion of Na+ ions from the soda-lime glass to TiO2 multi-layer films occurred as a consequence of high-temperature calcination (∼500 ◦ C), and this phenomenon inhibited the crystallization of the first layer and influenced the crystal structure in the other layers of TiO2 multilayer thin films, affecting their photoactivity negatively (decreasing the photoactivity) as well. The pre-coating of soda-lime glass with SiO2 layer was necessary in such case to get TiO2 anatase films.
On the basis of information on values of the contact angles it is possible to evaluate the surface wettability of prepared TiO2 thin films. Based on the measured contact angles in the range from 8 to 30◦ , as shown in Fig. 8, the hydrophilic character of the majority of thin films can be confirmed. The hydrophobic character was observed only in the case of films prepared using pure scCO2 (Exp. 1) where the contact angle reached 95◦ . No correlation between the contact angles and the anatase crystallite-size was observed. 4. Conclusions TiO2 anatase thin films were successfully prepared using processing by pure and water-modified supercritical carbon dioxide
296
M. Sajfrtová et al. / J. of Supercritical Fluids 117 (2016) 289–296
without additional thermal treatment. When the processing was performed at lower temperatures than 150 ◦ C and pressure bellow 30 MPa the greater amount of surfactant residue in thin films was observed. In terms of TiO2 anatase crystallization in a form of thin film the one step processing with pure scCO2 and with subcritical water proved to be unsuitable due to poor solubility of used surfactant and low adhesion of thin film to soda-lime glass substrate, respectively. However, it was found that the addition of the small amount of water to scCO2 during the one step and multi-step processing led to crystallization of TiO2 anatase, accompanied with high homogeneity of thin films. Two and three-step processing using the high density CO2 (30 MPa and 40 ◦ C) in the first step and subsequently the temperature of 150 ◦ C in the next step/s were advantageous in terms of the film purity as well as the crystallization of anatase. When scCO2 was modified by water in two steps followed by additional drying with pure scCO2 at 150 ◦ C (Exp. 12) the largest anatase crystallites (∼12 nm) were obtained thanks to the prolonged action of water and higher temperature in the second and third step of processing. Concerning the effect of used substrate (soda-lime glass and monocrystalline Si) on the crystal structure and microstructure of TiO2 thin films, it was revealed that using the one step or multi-step processing by water-modified scCO2 any undesirable phenomena within the substrates, having the negative effects on crystallization of TiO2 thin films, did not take place; on both substrates the anatase thin films directly crystallized. The largest anatase crystallites of ∼11 nm size were obtained similarly as on soda-lime glasses, using scCO2 modified by water in two steps followed by additional drying with scCO2 at 150 ◦ C (Exp. 12). Thus, the universality of developed processing of TiO2 gel thin films by pure and water-modified scCO2 was successfully confirmed. The assessed values of contact angles of TiO2 anatase thin films in the range from 8 to 30◦ confirmed the hydrophilic character of most thin films (except those prepared using pure scCO2 ). Acknowledgement The financial support of the Grant Agency of the Czech Republic (project reg. no. 14-23274S) is gratefully acknowledged. References [1] M. Mohamad, B.U. Haq, R. Ahmed, A. Shaari, N. Ali, R. Hussain, A density functional study of structural, electronic and optical properties of titanium dioxide: characterization of rutile, anatase and brookite polymorphs, Mater. Sci. Semicond. Process. 31 (2015) 405–414. [2] D. Reyes-Coronado, G. Rodriguez-Gattorno, M.E. Espinosa-Pesqueira, C. Cab, R. de Coss, G. Oskam, Phase-pure TiO(2) nanoparticles: anatase, brookite and rutile, Nanotechnology 19 (2008). [3] V. Augugliaro, M. Bellardita, V. Loddo, G. Palmisano, L. Palmisano, S. Yurdakal, Overview on oxidation mechanisms of organic compounds by TiO2 in heterogeneous photocatalysis, J. Photochem. Photobiol. C 13 (2012) 224–245. [4] F. Han, V.S.R. Kambala, M. Srinivasan, D. Rajarathnam, R. Naidu, Tailored titanium dioxide photocatalysts for the degradation of organic dyes in wastewater treatment: a review, Appl. Catal. A: Gen. 359 (2009) 25–40.
[5] K. Nakata, T. Ochiai, T. Murakami, A. Fujishima, Photoenergy conversion with TiO2 photocatalysis: new materials and recent applications, Electrochim. Acta 84 (2012) 103–111. [6] L.Q. Ye, J.Y. Liu, L.H. Tian, T.Y. Peng, L. Zan, The replacement of {101} by {010} facets inhibits the photocatalytic activity of anatase TiO2, Appl. Catal. B: Environ. 134 (2013) 60–65. [7] Y.H. Hsien, C.F. Chang, Y.H. Chen, S.F. Cheng, Photodegradation of aromatic pollutants in water over TiO(2) supported on molecular sieves, Appl. Catal. B: Environ. 31 (2001) 241–249. [8] Y.D. Wang, C.L. Ma, X.D. Sun, H.D. Li, Synthesis and characterization of amorphous TiO2 with wormhole-like framework mesostructure, J. Non Cryst. Solids 319 (2003) 109–116. [9] S. Boujday, F. Wunsch, P. Portes, J.F. Bocquet, C. Colbeau-Justin, Photocatalytic and electronic properties of TiO2 powders elaborated by sol-gel route and supercritical drying, Sol. Energy Mater Sol. Cells 83 (2004) 421–433. [10] P. Kluson, H. Luskova, T. Cajthaml, O. Solcova, Non thermal preparation of photoactive titanium(IV) oxide thin layers, Thin Solid Films 495 (2006) 18–23. [11] S. Tursiloadi, Y. Yamanaka, H. Hirashima, Thermal evolution of mesoporous titania prepared by CO2 supercritical extraction, J. Sol-Gel Sci. Technol. 38 (2006) 5–12. [12] M.D. Wei, K.X. Wang, M. Yanagida, H. Sugihara, M.A. Morris, J.D. Holmes, H.S. Zhou, Supercritical fluid processing of mesoporous crystalline TiO2 thin films for highly efficient dye-sensitized solar cells, J. Mater. Chem. 17 (2007) 3888–3893. [13] Y. Asai, S. Narishige, K. Fujioka, H. Uchida, S. Koda, Chemical processing for sol-gel derived metal oxide thin films using supercritical carbon dioxide fluid, ICC3: Symposium 2A: Novel Chemical Processing; Sol-Gel and Solution-Based Processing (2011), IOP Publishing. [14] L. Matejova, T. Cajthaml, Z. Matej, O. Benada, P. Kluson, O. Solcova, Super/subcritical fluid extractions for preparation of the crystalline titania, J. Supercrit. Fluids 52 (2010) 215–221. [15] L. Matejova, Z. Matej, R. Fajgar, T. Cajthaml, O. Solcova, TiO2 powders synthesized by pressurized fluid extraction and supercritical drying: effect of water and methanol on structural properties and purity, Mater. Res. Bull. 47 (2012) 3573–3579. [16] H. Bartkova, P. Kluson, L. Bartek, M. Drobek, T. Cajthaml, J. Krysa, Photoelectrochemical and photocatalytic properties of titanium (IV) oxide nanoparticulate layers, Thin Solid Films 515 (2007) 8455–8460. [17] P. Kluson, P. Kacer, T. Cajthaml, M. Kalaji, Titania thin films and supported nanostructured membranes prepared by the surfactant assisted sol-gel method, Chem. Biochem. Eng. Q. 17 (2003) 183–190. [18] Z. Matej, R. Kuzel, L. Nichtova, XRD total pattern fitting applied to study of microstructure of TiO2 films, Powder Diffr. 25 (2010) 125–131. [19] K. Koci, L. Obalova, L. Matejova, D. Placha, Z. Lacny, J. Jirkovsky, O. Solcova, Effect of TiO2 particle size on the photocatalytic reduction of CO2, Appl. Catal. B: Environ. 89 (2009) 494–502. [20] O. Solcova, L. Matejova, S. Krejcikova, Z. Matej, R. Kuzel, Z. Stryhal, O. Benada, Structural study of tailored titania thin layers, Collect. Czech. Chem. Commun. 73 (2008) 1222–1230. [21] J. Krysa, P. Novotna, S. Kment, A. Mills, Effect of glass substrate and deposition technique on the properties of sol gel TiO2 thin films, J. Photochem. Photobiol. A 222 (2011) 81–86. [22] P. Novotna, J. Krysa, J. Maixner, P. Kluson, P. Novak, Photocatalytic activity of sol-gel TiO2 thin films deposited on soda lime glass and soda lime glass precoated with a SiO2 layer, Surf. Coat. Tech. 204 (2010) 2570–2575. [23] H. Imai, H. Hirashima, Preparation of porous anatase coating from sol-gel-derived titanium dioxide and titanium dioxide-silica by water-vapor exposure, J. Am. Ceram. Soc. 82 (1999) 2301–2304. [24] H. Imai, H. Morimoto, A. Tominaga, H. Hirashima, Structural changes in sol-gel derived SiO2 and TiO3 films by exposure to water vapor, J. Sol-Gel Sci. Technol. 10 (1997) 45–54. [25] H. Uchiyama, M. Okada, H. Kozuka, Effect of the amount of H2O on the crystallization of TiO2 films prepared from alkoxide solutions containing acetylacetone, J. Am. Ceram. Soc. 97 (2014) 1091–1095.
Contents lists available at ScienceDirect
The Journal of Supercritical Fluids journal homepage: www.elsevier.com/locate/supflu
Preparation of nanocrystalline titania thin films by using pure and water-modified supercritical carbon dioxide Marie Sajfrtová a,∗, Marie Cerhová a, Vladislav Dˇrínek a, Stanislav Daniˇs b, Lenka Matˇejová c a b c
Institute of Chemical Process Fundamentals of the ASCR, v.v.i., Rozvojová 135, 165 02 Prague 6, Czech Republic Department of Condensed Matter Physics, Faculty of Mathematics and Physics, Charles University in Prague, Ke Karlovu 5, 121 16 Prague 2, Czech Republic Institute of Environmental Technology, VSˇ B-Technical University of Ostrava, 17. listopadu 15/2172, 708 33 Ostrava-Poruba, Czech Republic
a r t i c l e
i n f o
Article history: Received 7 April 2016 Received in revised form 10 July 2016 Accepted 11 July 2016 Available online 12 July 2016 Keywords: Titania thin film Anatase Supercritical carbon dioxide Subcritical water Crystallization Microstructure
a b s t r a c t Processing by pure and water-modified (30 wt.%) supercritical carbon dioxide and by subcritical water were utilized for the direct preparation of highly pure TiO2 anatase thin films without any subsequent thermal treatment. One step processing was compared with the two or three step processing combining pure and modified CO2 . The effect of temperature (40–150 ◦ C) and the amount of CO2 (100–200 g) passed through the high-pressure column on the (micro)structure and the purity of TiO2 thin films were examined at pressure of 30 MPa. Prepared thin films were characterized with respect to the structural properties and purity by Raman spectroscopy. The most promising thin films were analysed with respect to microstructural properties by means of X-ray diffraction to determine the phase composition, the crystallite-size and the crystallite-size distribution. High temperature had a positive effect on the crystallization as well as the purity of TiO2 thin films during the one step and multi-step processing. When TiO2 thin films were exposed to water-modified supercritical CO2 and temperature of 150 ◦ C under pressure of 30 MPa, the desired crystalline structure of anatase was obtained. The anatase crystallites growth was mainly influenced by the presence of water. Anatase crystallites sizes of 2–12 nm were obtained depending on the processing method on both investigated substrates (soda-lime glass and monocrystalline Si) on which the TiO2 thin films were deposited. Using one step or multi-step processing by water-modified supercritical CO2 any undesirable effects such as Na+ ions diffusion from the soda-lime glass substrate to the one-layer TiO2 film, having negative effect on crystallization of anatase, did not take place. The universality of developed processing by pure and water-modified supercritical CO2 for preparation of TiO2 anatase thin films was successfully confirmed for two different substrates. © 2016 Elsevier B.V. All rights reserved.
1. Introduction TiO2 (Titanium dioxide, titania) belongs thanks to its excellent photochemical performance and other photo-induced phenomena under UV light (with wavelength < 365 nm) among materials under keen scientific interest. It has been explored in a form of thin films in many application areas; as sensing films of gas sensors, coatings for self-cleaning surfaces or with antimicrobial activity, an electrode material or photocatalyst promising in waste water and air treatment technologies. Titania is a semiconductor with the band gap energy of 3.2 eV, 3.0 eV and 3.1 eV that belong to its crystal structure of anatase, rutile a brookite, respectively. Despite
∗ Corresponding author. E-mail addresses: [email protected] (M. Sajfrtová), [email protected] (M. Cerhová), [email protected] (V. Dˇrínek), [email protected] (S. Daniˇs), [email protected] (L. Matˇejová). http://dx.doi.org/10.1016/j.supflu.2016.07.007 0896-8446/© 2016 Elsevier B.V. All rights reserved.
the fact that anatase shows higher band gap energy than rutile or brookite, anatase exhibits higher photoactivity. The reason of this performance is a different structure of energy bands; the energy of anatase conductive band is higher than that of rutile or brookite [1,2]. Practically it means the enhanced reduction capability of excited electrons which are needed for the formation of free radicals which finally participate in the degradation e.g. of organic pollutants [3–5]. Since microstructural properties such as the crystallite size on nanosize level and shape, crystallite size distribution, phase composition, oxygen vacancy, microstrain etc. crucially affect photo-electrochemical response and photocatalytic performance of titania thin films [6], e.g. via the effect on their electronic and optical properties, it is very important to know and control all preparation and processing steps to avoid undesirable effects within microstructure of thin films. Commonly used method for the preparation of pure and crystalline TiO2 thin films is calcination [7,8]. However, this approach
290
M. Sajfrtová et al. / J. of Supercritical Fluids 117 (2016) 289–296
has some disadvantages such as excessive sintration, crystallite growth or recrystallization. Moreover, during calcination the microstructural properties such as crystallite-size and phase composition, having the influence on the photocatalytic activity, cannot be controlled. In last years this standard thermal processing was overcome by supercritical carbon dioxide drying combined with thermal processing. It was found that the pre-treatment using supercritical carbon dioxide proposed for lowering the process temperature of sol-gel derived metal oxide thin films as well as powders helps to enlarge surface area significantly and improve electronic properties showing enhanced photocatalytic performance [9–11]. Wei et al. [12] engaged in the effect of the post-treatment of mesoporous crystalline TiO2 thin films by a supercritical CO2 (scCO2 ) on the photovoltaic performance. In their study, a serie of TiO2 thin films composed of mesopores and nanoparticles were prepared by a spin-coating process and then statically treated by scCO2 for 30 min at pressures of 34.5 MPa and 58.6 MPa and at temperatures ranging from 60 to 100 ◦ C. The films were calcined at temperature of 400 ◦ C for 3 h before the characterization and solar cell fabrication. The mesopores in the TiO2 thin film without the scCO2 treatment collapsed after calcination at temperatures above 400 ◦ C, while the films treated by scCO2 exhibited an enhanced thermal stability and well-preserved porosity. Film processing using supercritical CO2 accomplished with a batch-type reaction system was compared with conventional calcination by Asai et al. [13]. The TiO2 precursor films on soda-glass substrates prepared by sol-gel coating using Ti-alkoxide solution were converted to crystalline TiO2 (anatase with 12 nm crystal size) successfully by treatment in scCO2 at a fluid pressure of 15 MPa and a substrate temperature of 300 ◦ C whereas no crystallization occurred by conventional heat treatment at 400 ◦ C. Besides that, several works reported on utilization of pressurized water and supercritical/subcritical methanol in a flow regime which resulted in nanocrystalline titania powders with significantly enlarged surface area and purity comparable to usually thermally treated titania powders [14,15]. It was shown that utilization of this solvent combination leads to direct crystallization of titania, namely by the effect of pressurized water [14–17]. The motivation of this work was to design the improved supercritical fluid process which can lead to purification and direct crystallization of TiO2 thin films without any subsequent thermal treatment. The processing by pure or water-modified supercritical CO2 in a flow regime was investigated and compared with subcritical water processing. All the experiments were performed over precursor titania gel thin films prepared by reverse micelles assisted sol-gel method, using hardly removable nonionic surfactant Triton X-114, however, forming uniform nano-domains. The effect of various processing conditions (i.e. temperature, pressure, volume and flow rate of solvents) as well as the type of substrate on microstructure and purity of TiO2 thin films was thoroughly studied by means of Raman spectroscopy, X-ray diffraction, and contact angle measurements.
was treated ultrasonically for 30 min before utilization to remove bubbles. For deposition of the films two types of substrate were used: soda-lime glass and monocrystalline Si (non-diffracting substrate). 2.2. Sol preparation and deposition of thin films Gel titania thin films were deposited on substrates (soda-lime glass and monocrystalline Si) by dip-coating method, using the sol prepared by sol-gel process controlled within reverse micelles of non-ionic surfactant Triton X-114 in cyclohexane. The molar composition of titania sol was following; cyclohexane: Triton X-114:water:Ti(OCH2 (CH3 )2 )4 = 11:1:1:1 [14]. In a shortcut, proper amounts of cyclohexane, Triton X-114 and water were mixed and vigorously stirred for 15 min for homogenization and formation of reverse micelles. Then, titanium (IV) isopropoxide was injected to micellar solution under vigorous stirring. After addition of isopropoxide the sol was stirred for next 20 min. A prepared sol was left standing in a closed glass bottle for 4 h to stabilize. Ultrasonically cleaned and dried soda-lime glasses and monocrystalline Si were dipped into the sol by using a dip-coater 4 idLab. The deposition conditions were following: the immersion velocity 15 cm min−1 , the delay in the sol 30 s, the emergence velocity 6 cm min−1 . After deposition the substrates with gel thin films were left on air overnight and then were processed by proper investigated method using pure and/or water-modified scCO2 or subcritical water. 2.3. Supercritical fluid crystallization (SFC) Air dried glasses with deposited gel thin films were fixed in a steel holder and immersed in the high pressure column (volume 150 ml; inner diameter 30 mm) filled in the bottom part with glass beads serving as solvent flow distributors. The column was connected with stainless steel capillaries and placed into the airconditioned box. The SFC experiments were carried out in Spe-ed SFE apparatus (Applied Separations, USA) whose schema is shown in Fig. 1. Carbon dioxide was sucked from a pressure container using a high-pressure pump cooled by water to 5 ◦ C. ScCO2 entered the lower end of the column at pressure of 30 MPa. The temperature of scCO2 was maintained using a hot air ventilator in an oven. In experiments with modified CO2 , water was supplied at a constant flow rate by a high-pressure LCP 4020.3 (ECOM s.r.o.) and mixed with CO2 before entering the column to reach concentration in scCO2 30 wt.%. The solution flowing from its upper end was expanded to the ambient pressure in a heated micrometer valve and the extract was collected in empty glass vial at the ambient temperature. The quantity of the gaseous CO2 leaving the trap and its flow rate was measured using a gas meter. The CO2 flow rate was adjusted to 0.8 g min−1 using the micrometer valve. Experimental conditions and design were changed as shown in Table 1.
2.1. Materials
2.3.1. One step processing Experiment with pure CO2 was carried out at constant pressure of 30 MPa, temperature of 150 ◦ C and solvent amount of 100 g. When modified CO2 was used, different temperature (40–150 ◦ C) and pressure (10–30 MPa) were applied.
Cyclohexane (p.a., Lachner, Neratovice, CR), Triton X-114 ((1,1,3,3,-Tetramethylbutyl)phenyl-polyethylene glycol, SigmaAldrich, USA), Ti(OCH2 (CH3 )2 )4 (Titanium(IV) isopropoxide, SigmaAldrich, USA) and distilled water were used for the preparation of precursor solution of TiO2 gel films. Carbon dioxide (>99.9%) for supercritical fluid processing was purchased from Linde Technoplyn (Prague, CR). Distilled water for subcritical water processing
2.3.2. Multi-step processing Combination of several steps with pure and water-modified scCO2 was tested. Order of individual steps and temperature varied as shown in Table 1. In the first step, the pure or water-modified CO2 at the temperature of 40 or 150 ◦ C was used. After 100 g of CO2 flowed through the column the conditions were changed and experiment continued. Second step was finished when 100 g of CO2
2. Materials and methods
M. Sajfrtová et al. / J. of Supercritical Fluids 117 (2016) 289–296
291
Fig. 1. Illustration of experimental set-up: (1) CO2 pressure bottle, (2) bottle with modifier, (3) cooling bath, (4) high pressure pumps, (5) air-heated box, (6) high pressure column, (7) glass beads, (8) glasses with deposited TiO2 thin films, (9) micrometer valve, (10) separator and (11) gas meter. Table 1 Experimental design and conditions of supercritical fluid crystallization. Experiment Nr.
Experimental design (temperature, ◦ C/solvent)
1 2 3 4 5 6 7 8 9 10 11 12
150 40 100 150 150a 40 40 150 40 40 150 40
1st step
a
2nd step CO2 CO2 + W CO2 + W CO2 + W CO2 + W CO2 CO2 CO2 CO2 + W CO2 CO2 CO2 + W
40 150 150 150 150 150 150
focused to a spot of 0.6 mm in diameter. The incident energy was minimized as necessary to avoid thermal effects as a consequence of the laser irradiation. Spectral resolution of the instrument was 2 cm−1 and the spectrometer was calibrated using crystalline silicon standard with line at 520 cm−1 .
3rd step
CO2 + W CO2 + W CO2 + W CO2 CO2 + W CO2 + W CO2 + W
150 150 150
CO2 CO2 CO2
at 10 MPa.
was consumed. In order to remove residual moisture from the films surface the third step followed in some experiments until 200 g of pure CO2 flowed at 150 ◦ C. The CO2 flow rate at this scCO2 drying was adjusted to 1.6 g min−1 . 2.4. Subcritical water crystallization The experiments with subcritical water (SW) were performed in the same apparatus as the SFC. Heated water was pumped to highpressure column by high-pressure LCP 4020.3 (ECOM s.r.o.) with a set flow rate of 2 ml min−1 . After 50 ml of water flowed through the expansion valve, the experiment was terminated.
2.5.2. X-ray diffraction XRD investigations were provided on a Panalytical MPD diffractometer in the parallel beam geometry. X-rays were generated using Cu-xray tube (40 kV/35 mA). Primary beam was conditioned by means of x-ray mirror and diffracted beam have been collected using proportional gas-filled detector equipped with parallel-plate collimator. X-ray patterns were measured at constant angle of incidence of x-rays (1◦ with respect to the layer surface). Diffraction angle, 2\, varied within the range 10–120◦ Crystal structure, crystallite-size as a median of the log-normal crystallite-size distribution and log-normal crystallite-size distribution were evaluated using the Rietveld fitting program for thin film analysis − MSTRUCT, using the sphere model of the crystallites [18]. 2.5.3. Contact angle measurements The device for the measurement of contact angles consisted of a microscope with quadruple lens, table with a micro-displacements and camera Sony SSC-C370. The image was processed using software Drop Analysis (Ecole Polytechnique Federale de Lausanne). 3. Results and discussion 3.1. Structure and purity of TiO2 thin films
2.5. Characterization of thin films The prepared thin films were characterized with respect to their microstructural properties by Raman spectroscopy as well as Xray diffraction (XRD). Raman spectroscopy helped to identify the presence of anatase, the desired crystal structure of TiO2 and the presence of impurities from used organic precursors. Thin films with confirmed crystallinity was subsequently subjected to XRD, which allowed to determine the crystallite size, crystallite-size distribution and the specific composition of the crystalline phases. 2.5.1. Raman spectroscopy Raman spectra were acquired on a dispersive Nicolet Almega XR spectrometer equipped with Olympus BX 51 microscope. Excitation laser source (473 nm) with an incident power of 50 mW with 256 expositions with 0.5 s duration was used. The laser beam was
Raman spectroscopy proved to be a very sensitive technique for determining the crystallinity and purity of prepared thin films. Raman spectra and the appearance of TiO2 thin films deposited on soda-lime glasses processed by scCO2 under various conditions mentioned in Table 1 and subcritical water are compared in Figs. 2 and 3. Signal intensities in the range of 3300–700 cm−1 of Raman shift indicates the occurrence of undissolved surfactant and other organic impurities, vibrations in the range of 700–100 cm−1 of Raman shift (the grey field) show the presence of TiO2 anatase peaks. 3.1.1. One step processing Fig. 2 compares Raman spectra of TiO2 thin films processed in one step by subcritical water and pure or water-modified scCO2 at different temperatures and pressures.
292
M. Sajfrtová et al. / J. of Supercritical Fluids 117 (2016) 289–296
Fig. 2. Raman spectra and appearance of TiO2 thin films deposited on soda-lime glasses processed at 30 MPa using one step processing by: subcritical water at 150 ◦ C (SW), pure scCO2 at 150 ◦ C (1), water-modified scCO2 at 40 ◦ C (2), 100 ◦ C (3) and 150 ◦ C (4) and at 10 MPa and 150 ◦ C (5). Valence and deformation vibrations: ( C H, 3073 cm−1 ), ( C H, 2913 cm−1 ), (C C, 1613 cm−1 ), ␦(H C H, 1465 cm−1 ) a (C C aliphatic, 1288 cm−1 ), region of anatase peaks: 700–100 cm−1 .
at 30 MPa and 150 ◦ C (Exp. 1) were
Films treated with pure scCO2 amorphous as expected. Moreover, the presence of organic carbon was proved in these thin films. The surfactant bands which are attributed in Raman spectra to valence and deformation vibrations ( C H, 3073 cm−1 ), ( C H, 2913 cm−1 ), (C C, 1613 cm−1 ), ␦(H C H, 1465 cm−1 ) and (C C aliphatic, 1288 cm−1 ) indicate that the solubility of surfactant and other organic residues of the sol production in pure scCO2 is very poor. Contrary to that in Raman spectra of thin films processed by subcritical water (Exp. SW) at temperature of 150 ◦ C and pressure of 30 MPa only bending C H peak, actually G (graphitic-like) peak, in 1106 cm−1 and negligible valence vibrations ( C H, 3080 cm−1 ) and ( C H, 2935 cm−1 ) were observed. It demonstrates higher solubility of surfactant in subcritical water. The small peaks in the range of 100–700 cm−1 of Raman shift indicate the beginning crystallization of anatase in subcritical water as well. However, this solvent itself caused unwanted removal of the film from the soda-lime glass substrate, thus, the scCO2 was enriched by water to combine advantages of both processing. When 30 wt.% of water was added to the scCO2 as modifier at low temperature of 40 ◦ C (Exp. 2), no crystallization of TiO2 was observed. Increasing temperature to 100 ◦ C (Exp. 3) caused the formation of anatase crystallites and better solubility of organic compounds as evident from the characteristic anatase bands at 168, 418, 532 and 653 cm−1 and decreased stretching C H vibrations at 2800–3050 cm−1 . Increased bending C H vibration at ∼1620 cm−1 (D-peak) can be attributed to partial graphitization of the surfactant residues in thin film. At the highest tested temperature of 150 ◦ C (Exp. 4) almost all the surfactant and organic substances were removed from thin film and the anatase crystal structure was achieved. Only thin films prepared at 100 ◦ C and 150 ◦ C showed the compact layer on the soda-lime glass surface. Reducing the pressure bellow 30 MPa at 150 ◦ C (Exp. 5) decreased the solubility of the surfactant and associated purity of thin film. Increased bending C H vibration at ∼1620 cm−1 proves the graphitization of the surfactant residues. Although the anatase crystallized, the partial elution of TiO2 film from glass substrate was observed as evident in the photograph in Fig. 2.
Fig. 3. Raman spectra and appearance of TiO2 thin films deposited on soda-lime glasses processed at 30 MPa using multi-step processing by: pure scCO2 at 40 ◦ C followed by water-modified scCO2 at 40 ◦ C (6), pure scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (7, 10*), pure scCO2 at 150 ◦ C followed by watermodified scCO2 at 150 ◦ C (8, 11*), water-modified scCO2 at 40 ◦ C followed by pure scCO2 at 150 ◦ C (9) and water-modified scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (12*), *with subsequent drying with pure scCO2 at 150 ◦ C. Valence and deformation vibrations: ( C H, 3073 cm−1 ), ( C H, 2913 cm−1 ), (C C, 1613 cm−1 ), ␦(H C H, 1465 cm−1 ) a (C C aliphatic, 1288 cm−1 ), region of anatase peaks: 700–100 cm−1 . (For interpretation of the references to colour in the text, the reader is referred to the web version of this article.)
3.1.2. Multi-step processing The effect of experimental design, namely the different temperature and the solvent composition (pure and water-modified scCO2 ) combined in two or three steps on crystallinity and purity of TiO2 thin films is shown in Raman spectra in Fig. 3. Anatase was detected in all the thin films prepared by the multistep processing, however, the conditions and order of individual steps affected the purity of the films. Although no crystallization took place in the one step processing of thin films using pure scCO2 or water-modified scCO2 at 40 ◦ C (see Fig. 2), when both solvents were combined the anatase crystallized (Fig. 3, Exp. 6). Nevertheless, the film contained a lot of impurities as evident from the presence of valence and deformation vibrations ( C H, 3071 cm−1 ), ( C H, 2938 cm−1 ), (C C, 1598 cm−1 ), ␦(H C H, 1468 cm−1 ), (C C aliphatic, 1286 cm−1 ) and bending C H vibration (1620 cm−1 ) in spectrum of this sample. It is evident that the subsequent increase in temperature from 40 ◦ C to 150 ◦ C of water-modified scCO2 in the second step has a positive effect on the crystallization as well as the purity of the sample (Exp. 7, 10 and 12). Using high density CO2 (30 MPa and 40 ◦ C) in the first step of processing, there may be a partial cleaning of films from non-polar organic impurities. Water and high temperature in the second step of processing acted as a crystallization initiator. The experiment with reverse solvent arrangement when the pure scCO2 in the first step was replaced by the water-modified scCO2 (Exp. 9) and the experiments conducted at 150 ◦ C (Exp. 8, 11) failed to effectively remove the organic contamination from films, however, were accompanied by successful crystallization of anatase. The appear-
M. Sajfrtová et al. / J. of Supercritical Fluids 117 (2016) 289–296
293
Fig. 4. XRD spectra of TiO2 thin films deposited on soda-lime glasses processed at 30 MPa using one step processing by water-modified scCO2 at 100 ◦ C (3) and 150 ◦ C (4); and multi-step processing by: pure scCO2 at 40 ◦ C followed by water-modified scCO2 at 40 ◦ C (6), pure scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (7, 10*), pure scCO2 at 150 ◦ C followed by water-modified scCO2 at 150 ◦ C (8, 11*), water-modified scCO2 at 40 ◦ C followed by pure scCO2 at 150 ◦ C (9) and water-modified scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (12*). *with additional drying with pure scCO2 at 150 ◦ C. Dashed lines represent crystal structure of anatase.
Fig. 5. TiO2 anatase crystallite-size of thin films on soda-lime glasses processed at 30 MPa using one step processing by water-modified scCO2 at 100 ◦ C (3) and 150 ◦ C (4); and multi-step processing by: pure scCO2 at 40 ◦ C followed by water-modified scCO2 at 40 ◦ C (6), pure scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (7, 10*), pure scCO2 at 150 ◦ C followed by water-modified scCO2 at 150 ◦ C (8, 11*), water-modified scCO2 at 40 ◦ C followed by pure scCO2 at 150 ◦ C (9) and water-modified scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (12*). *with additional drying with pure scCO2 at 150 ◦ C.
ance of all thin films with crystallized anatase in the multi-step processing is shown in Fig. 3 as well. As obvious from Fig. 3 thin films create a compact layers without any significant inhomogeneity and/or defects on soda-lime glass surface and show violet up to green colour indicating the different anatase crystallite-size and thickness of thin films. 3.2. Crystal structure and microstructure of thin films determined from XRD XRD analysis of all prepared titania thin films confirmed the presence of anatase crystal structure as evident from the typical bands of the anatase structure in Fig. 4. Information on the crystallite-size and crystallite-size distribution of nanocrystalline TiO2 anatase play an important role in the physico-chemical behaviour of the material in terms of chemical and phase stability and chemical reactivity [19]. The comparison of anatase crystallite-sizes determined for TiO2 thin films deposited
on soda-lime glasses prepared by pure and modified scCO2 under different experimental design and conditions is summarized in Fig. 5. The size of anatase crystallites of thin films deposited on sodalime glasses varied according to the processing procedure and applied temperature in the range of 1–12 nm. Using the one step processing with water-modified scCO2 the anatase crystallite-size was positively affected by increasing temperature (see Fig. 5, Exp. 3 and 4). The anatase crystallite-size increased from ∼3 to ∼6.3 nm. When experiments began with pure scCO2 in multi-step processing and were followed with water-modified scCO2 , only smaller anatase crystallites under ∼4 nm were formed. Change of the order of solvents in multi-step processing, when the water-modified scCO2 at 40 ◦ C was used in the first step (see Fig. 5, Exp. 9), resulted in doubling the size of anatase crystallites. Due to the presence of water the crystallization of anatase was initiated already in the first step of processing and the crystal growth could continue in the second step during slow removal of moisture, moreover, more
294
M. Sajfrtová et al. / J. of Supercritical Fluids 117 (2016) 289–296
intensively as a consequence of higher temperature of 150 ◦ C. The positive effect of the prolonged action of water and of high temperature on crystal growth was also confirmed in Exp. 12, where the largest anatase crystallites (∼12 nm) were obtained. 3.3. Aspects related to the mechanism of crystallization of TiO2 thin films in water-modified supercritical CO2 Principally three procedures were compared; (1.) processing by pure scCO2 , (2.) processing by water-modified scCO2 (combined with processing by pure scCO2 in multi-steps) and (3.) processing by subcritical H2 O. The effectiveness of individual procedures differed a lot concerning the crystallization of single-layer TiO2 thin films, since the crystallization of TiO2 thin films is affected significantly by solubility of the nonionic surfactant Triton X-114 used for sol-gel synthesis in individual media. It is worth to mention that on one hand, Triton X-114 belongs among surfactants enabling to form reverse micelles and, thus, the preparation of TiO2 crystallites in nano-size due to limited hydrolysis and polycondensation reactions taking place in the core of the micelles, creating Ti O Ti O Ti bonds in the amorphous gel matrix. On the other hand, it is a complicated molecule from organic point of view, composed of polar and nonpolar part: (1,1,3,3-tetramethylbutyl)phenyl-polyethylene glycol. This chemical composition makes the surfactant Triton X-114 hardly totally degradable even by calcination, therefore higher calcination temperature (∼400◦ C) and more deposited layers were needed to reach crystallization of TiO2 anatase thin films [20]. Moreover, the calcination of TiO2 thin films deposited on soda-lime glasses is accompanied with a negative diffusion of Na+ ions [21,22], making the crystallization of TiO2 thin films more complicated from crystallographic point of view. Therefore, just the dissolution of the surfactant Triton X-114 in medium/mixture of media possessing optimal dissolving power opens up the way not only for the low-temperature preparation of anatase thin films, but also for its recovery. In our study the processing by pure scCO2 and the processing by subcritical H2 O were firstly examined separately. It was revealed that neither scCO2 , nor subcritical H2 O shows the sufficient dissolving power to remove sufficiently the used surfactant from thin films. Each medium dissolved only part of the surfactant. On the other side, processing by subcritical water resulted in crystallization of TiO2 thin films contrary to processing by scCO2 . Therefore the processing by water-modified scCO2 at various experimental conditions and experimental design was investigated to reveal the best/optimal conditions for maximum dissolution/removal of the
Fig. 6. XRD spectra of TiO2 thin films deposited on monocrystalline Si processed at 30 MPa using one step processing by water-modified scCO2 at 150 ◦ C (4); and multistep processing by: pure scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (7, 10*), water-modified scCO2 at 40 ◦ C followed by pure scCO2 at 150 ◦ C (9) and water-modified scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (12*). *with additional drying with pure scCO2 at 150 ◦ C. Dashed lines represent crystal structure of anatase.
surfactant. As it was mentioned above, the dissolution of the nonionic surfactant Triton X-114 is a crucial initial step affecting further the crystallization of TiO2 thin films. The dissolution/removal of the surfactant leads to revelation of the Ti O Ti bonds, reacting with subcritical water (i.e. in our tests with liquid water under pressure at relatively mild temperatures of 40 ◦ C or 150 ◦ C). Pressurized water attacks and cleaves the Ti O Ti bonds, allowing the atomic rearrangement and the crystallization of TiO2 anatase thin films. In the presence of pressurized water, which was always in excess during the processing, the rapid nucleation and the crystallite growth take place. While the dissolution of the surfactant strongly depends on both used temperature and pressure which must be optimal for water-modified scCO2 to dissolve the surfactant sufficiently, the nucleation and the anatase crystallites growth are affected mainly by applied temperature (namely temperature of water) and the exposure time to water. Our findings can be supported partially by already reported works from Imai et al. [23,24] and Uchiyama et al. [25] who studied different, however, partially similar systems. Imai et al. [23,24] observed the crystallization of alkoxide-derived amorphous TiO2 gel films during exposure to water vapour produced during the heat treatment to anatase films at low temperature of 180 ◦ C and they suggested the similar crys-
Fig. 7. TiO2 anatase crystallite-size of thin films deposited on monocrystalline Si processed at 30 MPa using one step processing by water-modified scCO2 at 150 ◦ C (4); and multi-step processing by: pure scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (7, 10*), water-modified scCO2 at 40 ◦ C followed by pure scCO2 at 150 ◦ C (9) and water-modified scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (12*). *with additional drying with pure scCO2 at 150 ◦ C.
M. Sajfrtová et al. / J. of Supercritical Fluids 117 (2016) 289–296
295
Fig. 8. Contact angles of TiO2 thin films deposited on soda-lime glasses processed at 30 MPa using one step processing by: subcritical water at 150 ◦ C (SW), pure scCO2 at 150 ◦ C (1), water-modified scCO2 at 40 ◦ C (2), 100 ◦ C (3) and 150 ◦ C (4) and at 10 MPa and 150 ◦ C (5), and using multi-step processing by: pure scCO2 at 40 ◦ C followed water-modified scCO2 at 40 ◦ C (6), pure scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (7, 10*), pure scCO2 at 150 ◦ C followed by water-modified scCO2 at 150 ◦ C (8, 11*), water-modified scCO2 at 40 ◦ C followed by pure scCO2 at 150 ◦ C (9) and water-modified scCO2 at 40 ◦ C followed by water-modified scCO2 at 150 ◦ C (12*), *with additional drying with pure scCO2 at 150 ◦ C. Appearance of the drop on the surface of selected TiO2 thin films is shown.
tallization mechanism as in our study. Water as a vapour was the crystallization agent. Uchiyama et al. [25], who studied the preparation of TiO2 thin films from titanium alkoxide-derived solutions containing acetylacetone, agreed with the mechanism of crystallization from Imai et al. [23,24]. In addition, they reported on the effect of higher content of water present in the titanium alkoxidederived solutions containing acetylacetone. They reported that larger amount of water present in the solutions and, therefore, subsequently larger amount of water vapour present during the heat treatment provides higher nucleation rate, resulting in the generation of a larger number of small crystallites (in their case of smaller TiO2 anatase crystallites). Within our study the following conclusions about the processing of reverse micelles-derived TiO2 gel thin films by water-modified scCO2 may be stated; (1.) pressurized hot water as a part of the water-scCO2 mixture dissolves the part of the used surfactant and causes the crystallization of TiO2 thin films, (2.) scCO2 serves as a carrier of the pressurized hot water and dissolves a part of the surfactant insoluble in the pressurized hot water, and (3.) longer exposure of the revealed Ti O Ti bonds to pressurized water and higher temperature promotes the TiO2 anatase crystallite growth. 3.4. Effect of substrate on crystal structure and microstructure of TiO2 thin films
The crystal structure and microstructural results from one step and multi-step processing by pure/water-modified scCO2 of TiO2 thin films (composed of one layer) deposited on monocrystalline Si are shown in Figs. 6 and 7. Concerning the crystal structure, it is evident that the substrate had no effect on the crystal structure and TiO2 anatase crystallized in all investigated runs on monocrystalline Si as well. It reveals that during developed processing by pure and water-modified scCO2 the negative diffusion of Na+ ions from soda-lime glass to TiO2 films does not occur due to low processing temperatures, and it results in direct crystallization of TiO2 anatase thin films (composed of one layer) on both types of substrates. Concerning the anatase crystallite-size, one step processing by water-modified scCO2 at 150 ◦ C of TiO2 films deposited on monocrystalline Si (Fig. 7, Exp. 4) led to anatase crystallites of similarly small, but not so uniform size as on soda-lime glass (Fig. 5, Exp. 4). In contrast to anatase thin films deposited on soda-lime glasses the processing of films deposited on monocrystalline Si by water-modified scCO2 at 40 ◦ C in the first step led to smaller anatase crystallites (Fig. 7, Exp. 9) and the anatase crystallites growth was not influenced by additional drying with pure scCO2 at 150 ◦ C (Fig. 7, Exp. 7 and 10*). The multi-step processing by watermodified scCO2 in both steps and with additional drying with pure scCO2 at 150 ◦ C (Fig. 7, Exp. 12) led to anatase crystallites of similar size as well as the size-distribution as on soda-lime glasses. 3.5. Surface wettability
The most promising found experimental design and conditions (i.e. Exp. 4, 7, 9, 10 and 12, see Table 1) were investigated with respect to crystal structure and microstructure for TiO2 thin films deposited on monocrystalline Si as well. The effect of substrate on quality of titania thin films was keenly investigated, since it was revealed in some previous studies [21,22] that the type of substrate affects significantly the crystallization of TiO2 films. The diffusion of Na+ ions from the soda-lime glass to TiO2 multi-layer films occurred as a consequence of high-temperature calcination (∼500 ◦ C), and this phenomenon inhibited the crystallization of the first layer and influenced the crystal structure in the other layers of TiO2 multilayer thin films, affecting their photoactivity negatively (decreasing the photoactivity) as well. The pre-coating of soda-lime glass with SiO2 layer was necessary in such case to get TiO2 anatase films.
On the basis of information on values of the contact angles it is possible to evaluate the surface wettability of prepared TiO2 thin films. Based on the measured contact angles in the range from 8 to 30◦ , as shown in Fig. 8, the hydrophilic character of the majority of thin films can be confirmed. The hydrophobic character was observed only in the case of films prepared using pure scCO2 (Exp. 1) where the contact angle reached 95◦ . No correlation between the contact angles and the anatase crystallite-size was observed. 4. Conclusions TiO2 anatase thin films were successfully prepared using processing by pure and water-modified supercritical carbon dioxide
296
M. Sajfrtová et al. / J. of Supercritical Fluids 117 (2016) 289–296
without additional thermal treatment. When the processing was performed at lower temperatures than 150 ◦ C and pressure bellow 30 MPa the greater amount of surfactant residue in thin films was observed. In terms of TiO2 anatase crystallization in a form of thin film the one step processing with pure scCO2 and with subcritical water proved to be unsuitable due to poor solubility of used surfactant and low adhesion of thin film to soda-lime glass substrate, respectively. However, it was found that the addition of the small amount of water to scCO2 during the one step and multi-step processing led to crystallization of TiO2 anatase, accompanied with high homogeneity of thin films. Two and three-step processing using the high density CO2 (30 MPa and 40 ◦ C) in the first step and subsequently the temperature of 150 ◦ C in the next step/s were advantageous in terms of the film purity as well as the crystallization of anatase. When scCO2 was modified by water in two steps followed by additional drying with pure scCO2 at 150 ◦ C (Exp. 12) the largest anatase crystallites (∼12 nm) were obtained thanks to the prolonged action of water and higher temperature in the second and third step of processing. Concerning the effect of used substrate (soda-lime glass and monocrystalline Si) on the crystal structure and microstructure of TiO2 thin films, it was revealed that using the one step or multi-step processing by water-modified scCO2 any undesirable phenomena within the substrates, having the negative effects on crystallization of TiO2 thin films, did not take place; on both substrates the anatase thin films directly crystallized. The largest anatase crystallites of ∼11 nm size were obtained similarly as on soda-lime glasses, using scCO2 modified by water in two steps followed by additional drying with scCO2 at 150 ◦ C (Exp. 12). Thus, the universality of developed processing of TiO2 gel thin films by pure and water-modified scCO2 was successfully confirmed. The assessed values of contact angles of TiO2 anatase thin films in the range from 8 to 30◦ confirmed the hydrophilic character of most thin films (except those prepared using pure scCO2 ). Acknowledgement The financial support of the Grant Agency of the Czech Republic (project reg. no. 14-23274S) is gratefully acknowledged. References [1] M. Mohamad, B.U. Haq, R. Ahmed, A. Shaari, N. Ali, R. Hussain, A density functional study of structural, electronic and optical properties of titanium dioxide: characterization of rutile, anatase and brookite polymorphs, Mater. Sci. Semicond. Process. 31 (2015) 405–414. [2] D. Reyes-Coronado, G. Rodriguez-Gattorno, M.E. Espinosa-Pesqueira, C. Cab, R. de Coss, G. Oskam, Phase-pure TiO(2) nanoparticles: anatase, brookite and rutile, Nanotechnology 19 (2008). [3] V. Augugliaro, M. Bellardita, V. Loddo, G. Palmisano, L. Palmisano, S. Yurdakal, Overview on oxidation mechanisms of organic compounds by TiO2 in heterogeneous photocatalysis, J. Photochem. Photobiol. C 13 (2012) 224–245. [4] F. Han, V.S.R. Kambala, M. Srinivasan, D. Rajarathnam, R. Naidu, Tailored titanium dioxide photocatalysts for the degradation of organic dyes in wastewater treatment: a review, Appl. Catal. A: Gen. 359 (2009) 25–40.
[5] K. Nakata, T. Ochiai, T. Murakami, A. Fujishima, Photoenergy conversion with TiO2 photocatalysis: new materials and recent applications, Electrochim. Acta 84 (2012) 103–111. [6] L.Q. Ye, J.Y. Liu, L.H. Tian, T.Y. Peng, L. Zan, The replacement of {101} by {010} facets inhibits the photocatalytic activity of anatase TiO2, Appl. Catal. B: Environ. 134 (2013) 60–65. [7] Y.H. Hsien, C.F. Chang, Y.H. Chen, S.F. Cheng, Photodegradation of aromatic pollutants in water over TiO(2) supported on molecular sieves, Appl. Catal. B: Environ. 31 (2001) 241–249. [8] Y.D. Wang, C.L. Ma, X.D. Sun, H.D. Li, Synthesis and characterization of amorphous TiO2 with wormhole-like framework mesostructure, J. Non Cryst. Solids 319 (2003) 109–116. [9] S. Boujday, F. Wunsch, P. Portes, J.F. Bocquet, C. Colbeau-Justin, Photocatalytic and electronic properties of TiO2 powders elaborated by sol-gel route and supercritical drying, Sol. Energy Mater Sol. Cells 83 (2004) 421–433. [10] P. Kluson, H. Luskova, T. Cajthaml, O. Solcova, Non thermal preparation of photoactive titanium(IV) oxide thin layers, Thin Solid Films 495 (2006) 18–23. [11] S. Tursiloadi, Y. Yamanaka, H. Hirashima, Thermal evolution of mesoporous titania prepared by CO2 supercritical extraction, J. Sol-Gel Sci. Technol. 38 (2006) 5–12. [12] M.D. Wei, K.X. Wang, M. Yanagida, H. Sugihara, M.A. Morris, J.D. Holmes, H.S. Zhou, Supercritical fluid processing of mesoporous crystalline TiO2 thin films for highly efficient dye-sensitized solar cells, J. Mater. Chem. 17 (2007) 3888–3893. [13] Y. Asai, S. Narishige, K. Fujioka, H. Uchida, S. Koda, Chemical processing for sol-gel derived metal oxide thin films using supercritical carbon dioxide fluid, ICC3: Symposium 2A: Novel Chemical Processing; Sol-Gel and Solution-Based Processing (2011), IOP Publishing. [14] L. Matejova, T. Cajthaml, Z. Matej, O. Benada, P. Kluson, O. Solcova, Super/subcritical fluid extractions for preparation of the crystalline titania, J. Supercrit. Fluids 52 (2010) 215–221. [15] L. Matejova, Z. Matej, R. Fajgar, T. Cajthaml, O. Solcova, TiO2 powders synthesized by pressurized fluid extraction and supercritical drying: effect of water and methanol on structural properties and purity, Mater. Res. Bull. 47 (2012) 3573–3579. [16] H. Bartkova, P. Kluson, L. Bartek, M. Drobek, T. Cajthaml, J. Krysa, Photoelectrochemical and photocatalytic properties of titanium (IV) oxide nanoparticulate layers, Thin Solid Films 515 (2007) 8455–8460. [17] P. Kluson, P. Kacer, T. Cajthaml, M. Kalaji, Titania thin films and supported nanostructured membranes prepared by the surfactant assisted sol-gel method, Chem. Biochem. Eng. Q. 17 (2003) 183–190. [18] Z. Matej, R. Kuzel, L. Nichtova, XRD total pattern fitting applied to study of microstructure of TiO2 films, Powder Diffr. 25 (2010) 125–131. [19] K. Koci, L. Obalova, L. Matejova, D. Placha, Z. Lacny, J. Jirkovsky, O. Solcova, Effect of TiO2 particle size on the photocatalytic reduction of CO2, Appl. Catal. B: Environ. 89 (2009) 494–502. [20] O. Solcova, L. Matejova, S. Krejcikova, Z. Matej, R. Kuzel, Z. Stryhal, O. Benada, Structural study of tailored titania thin layers, Collect. Czech. Chem. Commun. 73 (2008) 1222–1230. [21] J. Krysa, P. Novotna, S. Kment, A. Mills, Effect of glass substrate and deposition technique on the properties of sol gel TiO2 thin films, J. Photochem. Photobiol. A 222 (2011) 81–86. [22] P. Novotna, J. Krysa, J. Maixner, P. Kluson, P. Novak, Photocatalytic activity of sol-gel TiO2 thin films deposited on soda lime glass and soda lime glass precoated with a SiO2 layer, Surf. Coat. Tech. 204 (2010) 2570–2575. [23] H. Imai, H. Hirashima, Preparation of porous anatase coating from sol-gel-derived titanium dioxide and titanium dioxide-silica by water-vapor exposure, J. Am. Ceram. Soc. 82 (1999) 2301–2304. [24] H. Imai, H. Morimoto, A. Tominaga, H. Hirashima, Structural changes in sol-gel derived SiO2 and TiO3 films by exposure to water vapor, J. Sol-Gel Sci. Technol. 10 (1997) 45–54. [25] H. Uchiyama, M. Okada, H. Kozuka, Effect of the amount of H2O on the crystallization of TiO2 films prepared from alkoxide solutions containing acetylacetone, J. Am. Ceram. Soc. 97 (2014) 1091–1095.