# Fileset

[-Revised-Articlehigh_quality_epitaxial_thin_films_by_on-without_highlight.pdf](https://mdr.nims.go.jp/filesets/fa5d2cdc-47a9-4e4d-8cad-5249e7122ccd/download)

## Creator

Sudu Hakuruge Dilan Priyankara Wijekoon, Kosuke Ono, Masaru Shimomura, Takahiko Kawaguchi, Naonori Sakamoto, Naoki Wakiya

## Rights

[Creative Commons BY-NC Attribution-NonCommercial 4.0 International](https://creativecommons.org/licenses/by-nc/4.0/)

## Other metadata

[High quality epitaxial, homogeneous anatase thin films by on-site controlled hydrolysis on LaAlO3 substrates and characterization](https://mdr.nims.go.jp/datasets/8a8706dd-eded-4bd7-a37d-78d8c570d4d5)

## Fulltext

1  High quality epitaxial, homogeneous anatase thin films by on-site controlled 1 hydrolysis on LaAlO3 substrates and characterization 2 Sudu Hakuruge Dilan Priyankara Wijekoona, Kosuke Onoa, Masaru Shimomuraa, 3 Takahiko Kawaguchib, Naonori Sakamoto a,b,c, Naoki Wakiyaa,b,c* 4 a Graduate School of Science and Technology, Shizuoka University, 3 5 1 Johoku, Chuo Ku, 5 Hamamatsu 432 8561, Japan 6 b Graduate School of Integrated Science and Technology, Department of Engineering, Shizuoka 7 University, 3 5 1, Johoku, Chuo Ku, Hamamatsu 432 8561, Japan 8 c Research Institute of Electronics, Shizuoka University, 3 5 1, Johoku, Chuo Ku, Hamamatsu 9 432 8561, Japan 10 *Corresponding author’s name: Naoki Wakiya, PhD  11 Present address: Professor, Shizuoka University, Hamamatsu, 432-8561, Japan  12 Email address: wakiya.naoki@shizuoka.ac.jp 13 14 2  High quality epitaxial, homogeneous anatase thin films by on-site 1 controlled hydrolysis on LaAlO3 substrates and characterization 2  3 The anatase form of TiO₂ is a widely studied material due to its broad range of applications. 4 Epitaxial anatase thin films have attracted significant attention because of their enhanced 5 electrical and optical properties. However, fabricating anatase thin films remains challenging 6 due to their metastability and the need for highly sophisticated fabrication techniques. On-7 site controlled hydrolysis is a simple, cost-effective, and rapid method for producing smooth, 8 compact thin films on various surfaces. In this study, we demonstrate a straightforward 9 approach to fabricating highly oriented epitaxial anatase thin films on LaAlO₃ substrates 10 using different solvent mixtures. The epitaxial orientation and film quality were analyzed 11 using X-ray diffraction pole figures and rocking curves, while surface morphology was 12 characterized by Scanning electron microscopy and atomic force microscopy. Our results 13 indicate that thin film quality and morphology are primarily influenced by the annealing 14 temperature rather than the choice of solvent or titanium precursor, confirming the feasibility 15 of a scalable, low-cost epitaxial fabrication technique for anatase thin films. 16 Keywords: Anatase titanium dioxide, Solid state epitaxy, on-site controlled hydrolysis 17 High-quality epitaxial anatase thin films were fabricated via a solvent-assisted method, with 18 annealing temperature being the key factor, while solvent variations had minimal impact on 19 epitaxial properties or surface morphology. 20  21 Classification: 306 Thin film/ Coatings; 504 X-ray diffraction. 22 1. Introduction 23 Titanium dioxide (TiO₂) is one of the most intensively studied, stable, inert, and dielectric materials, 24 with applications in photocatalysis solar cells (dye-sensitized solar cells, perovskite solar cells), 25 transparent conductors, gas sensors, photodetectors, and UV-protectors, among other metal oxides 26 [1–10]. TiO₂ naturally exists in three polymorphic forms: rutile, anatase, and brookite. Among 27 3  these, rutile is the most thermodynamically stable phase, while anatase is a metastable phase that 1 transforms into rutile at temperatures above 700°C [11–13]. The fabrication of high-quality 2 epitaxial thin films of anatase has garnered significant attention due to their enhanced electrical 3 and optical properties, which result from a lower density of grain boundaries compared to 4 polycrystalline films [14]. Various thin film deposition techniques have been employed, including 5 pulsed laser deposition (PLD) [15], molecular beam epitaxy (MBE) [15,16], atomic layer 6 deposition (ALD) [17,18], sol-gel spin coating [14,19], dip coating [20], hydrothermal [21], and 7 solvothermal methods [22–24]. However, vacuum-based and hydrothermal techniques are less 8 energy-efficient and limit large-scale film production due to their high instrumentation costs and 9 complexity. On the other hand, as reviewed in a study by K. Ono at al. [23] the epitaxial quality 10 of the thin films is comparatively poor in solvent based techniques. Altogether fabrication of high-11 quality anatase single-crystalline thin films particularly challenging. Also reports on high quality 12 epitaxial anatase thin film fabrication by simple cost-effective rapid techniques is rare. 13 LaAlO₃ (LAO) substrates are widely used for heteroepitaxial thin film synthesis due to 14 their favorable properties such as high-temperature stability, chemical inertness, and the ability to 15 achieve atomically flat surfaces [15,25,26]. At high temperatures, LAO transforms into a cubic 16 structure while in most cases, the (001) orientation of the pseudo-cubic unit cell (a = b = c ≈ 3.791 17 Å, α ≈ β ≈ γ ≈ 90°) has been considered during fabricating (001)-oriented anatase thin films [27,28]. 18 Its pseudo-cubic structure has a lattice constant that closely matches the horizontal lattice 19 parameter (a = b ~3.784 Å) of tetragonal lattice anatase TiO₂ resulting in an extremely low lattice 20 mismatch of approximately 0.16%. This near-perfect lattice match can promote coherent epitaxial 21 growth with minimal strain and defect formation [23,29]. 22 4  The on-site controlled hydrolysis is a simple, cost effective and efficient new type of 1 concept that allows to make continuous compact thin films of TiO2. This technique enables precise 2 control over the composition and morphology of the films while utilizing relatively simple 3 equipment [30]. Although the on-site controlled hydrolysis process has been studied for the 4 deposition of polycrystalline TiO₂ films, its potential for producing epitaxial anatase films has not 5 been realized. 6 In this study, we explored the fabrication of high-quality epitaxial anatase TiO₂ thin films 7 on LAO substrates using the on-site controlled hydrolysis method with different solvent mixtures. 8 Our results demonstrate the feasibility of producing epitaxial anatase films with excellent 9 crystallinity and surface smoothness. Detailed, structural, morphological analyses were conducted 10 to elucidate the quality of the films and their potential for advanced applications. This work 11 represents a significant step toward the scalable and cost-effective fabrication of epitaxial oxide 12 thin films for next-generation technologies. 13 2. Experimental 14 2.1. Materials and method 15 Titanium tetraisopropoxide (TTIP) (Fujifilm Wako Chemicals, Wako 1st Grade, 95%) and titanium 16 tetraisobutoxide (TTBu) (Fujifilm Wako Chemicals, Wako 1st Grade, 95%) were used as titanium 17 precursors. 1-propanol (nPr) (Fujifilm Wako Chemicals, Guaranteed Reagent, 99.5%), 1-butanol 18 (nBu) (MP Biomedicals, Molecular Biology Reagent), were used as alcohols.  001 oriented, single 19 side polished LAO crystals purchased from SHINKOSHA Co. LTD. (0.5 mm thick, Ra≦1.0 nm) 20 were used as substrates. 21 5   1 Four thin films were fabricated on 8x8 mm2, 001 oriented LAO substrate using four 2 different solvent combinations: TTIP in nPr, TTIP in nBu, TTBu in nPr, TTBu in nBu maintaining 3 the water bath temperature at 0°C and temperature around the spin coater at 50 °C by onsite 4 controlled hydrolysis described in a previous study[30]. Five cycles of spin coating were repeated 5 to obtain approximately 150 nm of final thickness. After each cycle, the films were annealed at 6 150 °C for 3 minutes, followed by a final annealing step at 250 °C for 5 minutes to ensure complete 7 drying of the film. The concentration of titanium alkoxides in all solvent mixtures were fixed at 8 0.12 moldm-3. Ultimately fabricated films were annealed at 700 °C and 800 °C for 10h to obtain 9 epitaxial thin films (Figure 1).   10 2.2.Characterization 11 All films were characterized using the X'Pert Materials Research Diffractometer (MRD, Malvern 12 PANalytical Ltd., UK) with Cu Kα radiation (λ = 1.5406 Å). The scans included 2θ/ω scans, 13 rocking curve (ω) measurements, azimuthal scans (ϕ), and pole figure measurements. The tension 14 voltage was set to 45 kV, and the current to 40 mA. 2θ/ω scans were performed over a range of 15 18° to 55°, covering the most intense XRD peaks of anatase, with 0.02° intervals and a dwell time 16 of 0.5 seconds per step. Rocking curve (ω) scans were conducted over a 5° range, with 0.005° 17 intervals and a dwell time of 0.5 seconds per step. For azimuthal (ϕ) scans, measurements were 18 taken for the LAO (101) plane and the anatase (101) plane at 2θ = 33.4°, ψ = 45°, and 2θ = 25.3°, 19 ψ = 68.3°, respectively, using a 0.1° step size and a 0.1-second dwell time per step. Pole figure 20 measurements were performed for the LAO substrate (101) plane over a ψ range of 40°–50° at 2θ 21 = 33.4°, and for the anatase thin film over a ψ range of 0°– 90° at 2θ = 25.3°, with a 0.5° step size 22 and a 0.1-second dwell time per step. The morphology of the thin film was observed using a JSM-23 6  7001F Analytical Field Emission Scanning Electron Microscope (FE-SEM) at an acceleration 1 voltage of 15 kV and a probe current of 8 mA, after sputtering a 10 Å gold layer to enhance the 2 specimen’s conductivity. The surface morphology data over a 2500 µm² area was collected with a 3 50/1024 µm lateral resolution using a VN-8010 Simple AFM and analyzed with Gwyddion 2.67 4 (open-source software). Survey spectrum of X-ray Photoelectron Spectroscopy (XPS) was 5 recorded using an AXIS-ULTRA DLD photoelectron spectrograph with 1 eV spectral resolution 6 and three accumulations, while narrow spectra were collected with 0.1 eV spectral resolution 7 employing an Al Kα X-ray source. The sample surface was cleaned by argon (Ar) sputtering at 3 8 kV for 2 minutes before the XPS measurement. Thickness of thin films were estimated by cross-9 sectional images taken by FE-SEM.  10 3. Results and discussion 11 The XPS survey spectra of all thin films shown in Figure 2(a) consist only of the characteristic 12 signals of Ti and O, apart from the adventitious C 1s peak. The Figure 2(b) shows the deconvoluted 13 high-resolution XPS spectra of the thin film made by TTIP-nBu solvent mixture, exhibiting the 14 spin–orbit doublets of Ti 2p at approximately 458.5 eV (2p₃/₂) and 464.2 eV (2p₁/₂), corresponding 15 to area percentages of 32.8% and 67.2%, respectively. The energy separation between each doublet 16 is ~5.8 eV, which is consistent with previously reported values. These deconvoluted spectra 17 confirm the presence of Ti in the +4 oxidation state. The Figure 2(c) represent the deconvoluted 18 oxygen peak of the thin film made by TTIP-nBu solvent mixture, consists of two components with 19 different peak intensities: lattice oxygen (OL) at ~529.5 eV and hydroxides (OH) at ~531.2 eV, 20 characteristic of TiO₂. [22,23,31]. The C 1s spectrum (Figure 2(d)) can be deconvoluted into three 21 peaks: C-CH (284.6 eV), C-O (285.9 eV), and O=C (288.3 eV). The C-CH and C-O peaks are 22 attributed to adventitious carbon, likely from environmental contamination. The peak at 288.3 eV 23 7  provides indirect evidence of the presence of Ti-substituted carbon in the film [32]. However, at 1 high annealing temperatures, the likelihood of carbon atoms remaining in the thin film is very low 2 confirming that carbon impurities in the crystal are not significant [33,34]. Overall, XPS that all 3 the thin films are composed of Ti+4 state and O-2 state as well as are of high purity. Additionally, 4 the absence of characteristic signals from substrate elements serves as an indirect indication that 5 the thin films are continuous without cracks [31]. Moreover, the nearly identical high-resolution 6 XPS spectra suggest that the surface chemical properties of all thin films are almost identical 7 (Figure S1).  8 The XRD analysis of all TiO2 thin films on LAO substrates was conducted using pole 9 figures, omega scans, and 2θ/ω scans to evaluate crystallographic orientation, texture quality, and 10 structural properties. The 2θ/ω scan in Figure 3 shows only the characteristic peaks of LaAlO₃ 11 (001), (002), and anatase (004), confirming that all fabricated TiO₂ thin films are in the anatase 12 phase, with LaAlO₃ (001) || anatase (001). A shift in peak positions of anatase (004) was observed 13 from the stress-free position at 36.801° to 37.79° at 700 °C and to 37.72° at 800 °C. Using Bragg's 14 law and the relationship between lattice spacing, this shift corresponds to an increase in the lattice 15 parameters over c direction from 9.512 Å (stress-free) to 9.515 Å (at 700 °C) and 9.532 Å (at 16 800 °C) and the. The calculated theoretical in-plane strain is 0.20%, which, according to Poisson’s 17 relationship, indicates an out-of-plane expansion corresponding to in-plane compressive strain 18 [19,27,28,35].  19 Figure 4 compares the (004) anatase peaks of thin films obtained at 700 °C and 800 °C 20 using omega scan 2θ at 37.72° and 37.79°, respectively. As clearly seen, the Full Width at Half 21 Maximum (FWHM) improves at 800 °C, indicating an enhancement in crystal quality compared 22 to films annealed at 700 °C. Furthermore, the narrow peaks observed at 800 °C, with FWHM 23 8  values below 0.48° for all thin films, suggest excellent crystallinity, minimal mosaicity in the 1 epitaxial layer, and strong substrate-film interaction during growth. However, the FWHM values 2 for thin films fabricated using different solvent mixtures exhibit notable deviations. The minimum 3 FWHM observed is 0.37° for the TTBu-nBu-based solvent mixture, while the maximum FWHM 4 is 0.48° for the TTBu-nPr solvent mixture (Table 1). This variation follows a parallel trend with 5 the FWHM of the substrate (002) peaks (Figure S3), which can be attributed to the anisotropic 6 dependency of FWHM rocking curves due to the pseudo cubic nature of the substrate [36]. 7 Moreover, the mean FWHM of the rocking curve is the best-reported value obtained via solvent-8 assisted synthesis, as confirmed by a recent research review [23]. 9 As shown in Figure 5, the {101} planes of anatase (101, 011, -101, 0-11) are approximately 10 90° apart and align with the corresponding reflections of the LaAlO₃ substrate (101, 011, -101, 0-11 11). This indicates that the anatase film is not only oriented in-plane but also oriented out-of-plane 12 on the LaAlO₃ substrate, following the epitaxial relationship LaAlO₃ [001] || anatase [001], 13 consistent with previous reports. However, as seen in the magnified image, a 0.2° misalignment is 14 observed, which can be attributed to strain in the film. Furthermore, no additional peaks beyond 15 the {101} reflections or evidence of twinning were detected, further confirming the high-quality 16 crystal orientation. The average FWHM of the ϕ scan is 0.35° where substrate is 0.18° the 17 representing a considerable improvement in in-plane alignment compared to previously reported 18 solvent-based synthesis methods [19,37]. 19 The pole figures (Figure 6) for anatase (101) reflections further confirm the epitaxial 20 relationship between the anatase film and the LaAlO₃ substrate, as the intensity maxima of the 21 anatase {101} planes well align with those of the substrate. The absence of additional intensity 22 regions suggests the absence of secondary phases or random orientations. Furthermore, the similar 23 9  characteristics of the pole figures for each film are in good agreement with other types XRD of 1 analyses, providing evidence of identical thin film formation (Figure S2). 2 As previously reported by our group, smooth, compact, and continuous thin films can be 3 fabricated on surfaces through on-site controlled hydrolysis, achieved via the partial hydrolysis of 4 titanium alkoxides [30]. Upon low-temperature heating, these thin films convert into amorphous 5 TiO₂, while completing hydrolysis and removing residual alcohols from the film. During high-6 temperature annealing, the amorphous thin films transform into highly oriented, dense epitaxial 7 thin films through a well-studied solid-state epitaxy (SSE) process [38,39]. 8   Figures 7 and 8 compare the surface morphology of each thin film, the substrate, and the 9 film annealed at 250°C. As clearly observed in the SEM and AFM 3D images, all epitaxial film 10 surfaces exhibit a similar morphology, consisting of apparently identical grains. In contrast, the 11 SEM images show that the films annealed at 250°C have a flat, smooth surface similar to that of 12 the substrate. It is evident that surface roughness due to grain formation emerges after annealing. 13 This grain structure can be attributed to film shrinkage caused by the pyrolysis of residual organic 14 materials, material rearrangement during densification, and the natural surface roughening that 15 occurs during solid-state epitaxy [35]. Notably, fractures and bulging due to thermal stress were 16 not observed, emphasizing that the film is continuous and firmly attached to the substrate. 17 Furthermore, the cross-sectional SEM image (Figure 7(g)) confirms the continuity and thickness 18 of the anatase film and the firm bond between thin film and substrate. Also, the minimal variation 19 of surface roughness values of each thin film further confirms, the overall morphologies remain 20 consistent across all samples, without significant differences. 21 All the observations align with the typical behavior of SSE, wherein an initially amorphous 22 thin film transforms into a crystalline, epitaxially aligned structure upon thermal treatment. This 23 10  process explains the disappearance of secondary crystal phases, the enhancement of preferred 1 orientation, shifting the peak positions and the temperature-dependent increase in XRD peak 2 intensity. These behaviors result from a complex interplay of factors, including lattice mismatch, 3 differences in thermal expansion coefficients (LAO thermal expansion coefficient: 12.6 × 10⁻⁶ °C⁻¹, 4 anatase in-plane thermal expansion coefficient: 4.47×10⁻⁶ °C⁻¹ [40,41]), thermally induced atomic 5 mobility and rearrangement, and defect formation. Due to the complexity of these interactions, 6 isolating the individual contribution of each factor is challenging. 7 From a thermodynamic perspective, single crystalline structures are more stable than 8 amorphous ones. At elevated temperatures, atoms acquire sufficient energy to overcome the 9 activation barrier and rearrange from a disordered to an ordered crystalline state. This leads to 10 enhanced atomic alignment with the substrate lattice—i.e., epitaxial growth—and results in 11 increased film density. During SSE, this crystallization typically initiates at the film–substrate 12 interface and propagates toward the film surface. The first few unit cell layers become well-aligned 13 with the substrate, generating significant strain. However, as the crystallization progresses, this 14 strain is gradually relaxed. Since anatase has larger in-plane lattice parameters than both LAO and 15 STO substrates, strain relaxation introduces increased spacing in the upper unit cells [39,42–44]. 16 Additionally, higher temperatures increase atomic vibrations and thermal expansion in 17 both the film and substrate, further contributing to expansion and potential misalignment in the 18 upper layers. Consequently, the likelihood of defect formation—such as interstitials or propagating 19 dislocations—also increases with temperature. Upon cooling to room temperature, the film and 20 substrate contract at different rates due to their distinct thermal expansion coefficients. This 21 differential contraction, combined with pre-existing high-temperature defects, results in in-plane 22 compressive stress as the lattice attempts to contract. This residual strain can influence both the 23 11  position and sharpness of XRD peaks, ultimately impacting the measured crystallinity and 1 structural quality of the film. On the other hand, at high temperatures, surface roughness may 2 increase due to pyrolysis, density increase, and stress relaxation [39,43,44]. 3 Two main solvent-based approaches, distinguished by their growth mechanisms, can be 4 identified for fabricating epitaxial anatase thin films: sol-gel synthesis via solid-state epitaxy (SSE) 5 and solvothermal/hydrothermal synthesis via liquid-phase epitaxy. In both methods, titanium 6 alkoxides are commonly used as the precursor due to their high reactivity. To control the hydrolysis 7 and condensation reactions, additives such as acids, ligands or chelating agents, polymers, and 8 other solvents are typically introduced [19,20,22,23]. These result in complex mixtures that are 9 generally colloidal solutions consisting of partially hydrolyzed species. 10 In sol-gel-based synthesis, the colloidal solution is spread over the substrate, forming a 11 relatively less compact and often amorphous thin film. During annealing, this film transforms into 12 an epitaxial layer. However, due to its initial low density, the resulting surface may be less smooth, 13 and there is greater freedom for crystal orientation to deviate. Moreover, the non-uniformity of the 14 film can lead to weaker regions, making it more susceptible to cracking. In contrast, 15 solvothermal/hydrothermal synthesis can introduce impurities into the crystal lattice due to trapped 16 additives from the solvent system. For example, fluorine was found incorporated into the lattice in 17 a solvothermally grown film, as reported by Ono et al [23]. In both approaches, the presence of 18 these additives can compromise the chemical purity and structural quality of the final film. 19 However, the on-site controlled hydrolysis method offers distinct advantages. This 20 technique enables better control over unwanted impurities, as the reaction is regulated directly at 21 the substrate surface. It also produces a compact amorphous film that minimizes orientation 22 freedom during the epitaxial conversion process. Due to the uniformity and compactness of this 23 12  film, there are fewer weak regions, significantly reducing the risk of cracking. These factors 1 contribute to the formation of high-quality epitaxial anatase films using this method. 2 For this study, a simple setup similar to that described in the previous work and a standard 3 furnace were used to synthesize high-quality single crystalline thin films. As a result, the initial 4 equipment cost is minimal compared to advanced fabrication techniques, and the scalability of the 5 process is high. Unlike vacuum-based methods, which consume significant energy to maintain 6 vacuum conditions and heat large assemblies, this approach requires energy primarily to heat only 7 the substrate and thin films. The coating process is rapid, and the annealing can be performed in 8 batches, making the overall method both energy-efficient and timesaving. Furthermore, in contrast 9 to hydrothermal and solvothermal syntheses, this technique uses only a small amount of chemicals, 10 resulting in minimal chemical waste. Therefore, this method simplifies the process of synthesizing 11 high-quality epitaxial anatase films, which has traditionally been challenging to achieve. Such 12 high-quality single-crystalline films offer enhanced charge transport due to reduced recombination 13 losses and fewer trapping sites. As a result, these thin films can significantly improve the efficiency 14 of photovoltaic devices, as well as the sensitivity and responsivity of the sensors. Moreover, the 15 low defect density enhances their optical transparency and electrical conductivity, which are 16 crucial for applications in transparent electronics and optical computing, where precise control 17 over light propagation and minimal scattering are essential. Additionally, the (001) orientation of 18 anatase significantly boosts photocatalytic efficiency, making these films highly suitable for 19 environmental remediation and water splitting applications.  20 Additionally, the on-site controlled hydrolysis method involves several adjustable 21 parameters such as evaporation temperature, precursor concentration, number of coatings and spin-22 coating rate that influence the thickness of the resulting amorphous compact film. By carefully 23 13  tuning these parameters, the thickness of the initial amorphous film can be precisely controlled[30]. 1 Since this amorphous film is later converted into an epitaxial film through solid-state epitaxy, the 2 final thickness of the epitaxial film is largely determined by the initial thickness of the amorphous 3 layer, making the epitaxial film thickness controllable with similar precision. However, surface 4 roughening during the epitaxial transformation can introduce slight variations in thickness, which 5 should be taken into consideration in applications where precise film thickness is critical. 6 On the other hand, this fabrication method requires high-temperature annealing, which is 7 a disadvantage that may limit its applicability in certain areas. Moreover, factors such as the effect 8 of annealing temperature, film thickness, and dopability still require further investigation. 9 4. Conclusions 10 In this study, the fabrication of highly oriented epitaxial anatase thin films was successfully 11 achieved through on-site controlled hydrolysis. As previously reported, the epitaxial relationship 12 between the anatase film and the LaAlO₃ (LAO) substrate follows LaAlO₃ [001] || anatase [001] 13 for all thin films synthesized using different solvent mixtures. The epitaxial quality and surface 14 morphology of the thin films do not exhibit significant changes with variations in the solvent 15 mixture but are primarily influenced by the annealing temperature. Therefore, the effect of the 16 solvent or Ti precursor on epitaxial quality or morphology is negligible. Overall, this method offers 17 a fast, simple, low-cost, environmentally friendly, and efficient approach to fabricating high 18 quality 001 oriented anatase epitaxial thin films. Therefore, this method could accelerate the 19 development of next-generation energy, sensing, and electronic devices with improved 20 performance, stability, and environmental sustainability.  21 14  Acknowledgement 1 All the staff members of Instrumental Analysis Center, Hamamatsu campus, Shizuoka University. 2 Part of this work was conducted under the Cooperative Research Project Program of Research 3 Institute of Electronics, Shizuoka University. Part of this research was also supported by the 4 Collaborative Research Project of Laboratory for Materials and Structures, Institute of Innovative 5 Research, Institute of Science Tokyo. 6 Funding 7 Part of this work was supported by a Grant-in-Aid for Scientific Research from the Ministry of 8 Education, Culture, Sports, Science and Technology (No. 22H01770). 9 Disclosure statement 10 No potential conflict of interest was reported by the authors. 11 Data Availability Statement 12 The data that support the findings of this study are available from the corresponding author, 13 Naoki Wakiya, upon reasonable request. 14 References. 15 [1] Luttrell T, Halpegamage S, Tao J, et al. Why is anatase a better photocatalyst than rutile? - Model 16 studies on epitaxial TiO2 films. Sci Rep. 2014;4:4043.  17 [2] Li Y, Xia B, Jiang B. Thermal-induced durable superhydrophilicity of TiO2 films with ultra-18 smooth surfaces. J Solgel Sci Technol. 2018;87(1):50–58.  19 [3] Soussi A, Ait Hssi A, Boujnah M, et al. Electronic and Optical Properties of TiO2 Thin Films: 20 Combined Experimental and Theoretical Study. J Electron Mater. 2021;50(8):4497–4510.  21 15  [4] Sathish, M., Radhika, N. & Saleh, B. Current Status, Challenges, and Future Prospects of Thin 1 Film Coating Techniques and Coating Structures. J Bio Tribo Corros. 2023; 9:35. 2 [5] Qin J, Zhang Z, Shi W, et al. The optimum titanium precursor of fabricating TiO2 compact layer 3 for perovskite solar cells. Nanoscale Res Lett. 2017;12:640.  4 [6] Nakata K, Fujishima A. TiO2 photocatalysis: Design and applications. Journal of Photochemistry 5 and Photobiology C: Photochem. Rev. 2012;13(3):169–89.  6 [7] Rzaij JM, Rzaij JM, Abass AM. Review on: TiO2 Thin Film as a Metal Oxide Gas Sensor. J. 7 Chem. Rev. 2020;2(2):114–21.  8 [8] Arconada N, Durán A, Suárez S, et al. Synthesis and photocatalytic properties of dense and 9 porous TiO2-anatase thin films prepared by sol–gel. Appl Catal B. 2009;86(1–2):1–7.  10 [9] Yamada N, Hitosugi T, Kasai J, et al. Transparent conducting Nb-doped anatase TiO2 (TNO) thin 11 films sputtered from various oxide targets. Thin Solid Films. 2010;518(11):3101–3104. 12 [10] Sato Y, Akizuki H, Kamiyama T, et al. Transparent conductive Nb-doped TiO2 films 13 deposited by direct-current magnetron sputtering using a TiO2−x target. Thin Solid Films. 14 2008;516(17):5758–5762. 15 [11] Sabyrov K, Burrows ND, Penn RL. Size-dependent anatase to rutile phase transformation 16 and particle growth. Chem. Mater. 2013:25(8):1408-1415. 17 [12] Hanaor DAH, Sorrell CC. Review of the anatase to rutile phase transformation. J. Mater. 18 Sci. 2010;46(4):855–874.  19 [13] Tompsett GA, Bowmaker GA, Cooney RP, et-el. The Raman Spectrum of Brookite, TiO2 20 (Pbca,2=8). J. Raman Spectrosc. 1995;26(1):57-62. 21 [14] Jung HS, Lee JK, Lee J, et al. Mobility enhanced photoactivity in sol-gel grown epitaxial 22 anatase TiO2 films. Langmuir. 2008;24(6):2695–2698. 23 [15] Krupski K, Sanchez AM, Krupski A, et al. Optimisation of anatase TiO2 thin film growth 24 on LaAlO3(001) using pulsed laser deposition. Appl Surf Sci. 2016;388:684–90.  25 16  [16] Wang YH, Rahman KH, Wu CC, at al. A review on the pathways of the improved 1 structural characteristics and photocatalytic performance of titanium dioxide (TiO2) thin films 2 fabricated by the magnetron-sputtering technique. Catalysts. 2020;10(6):598. 3 [17] McDaniel MD, Posadas A, Wang T, et al. Growth and characterization of epitaxial 4 anatase TiO2 (001) on SrTiO2-buffered Si(001) using atomic layer deposition. Thin Solid Films. 5 2012;520(21):6525–6530.  6 [18] Barlaz DE, Seebauer EG. Solid phase epitaxial regrowth of (001) anatase titanium 7 dioxide. J. Vac. Sci. Technol. A. 2016; 34(2):20603. 8 [19] Jung HS, Lee JK, Lee J, et al. Strain relaxation in sol-gel grown epitaxial anatase thin 9 films. J. Phys. Chem. C. 2008;112(11):4205–4208.  10 [20] Oropeza FE, Zhang KHL, Regoutz A, et al. Growth of epitaxial anatase nano islands on 11 SrTiO3 (001) by dip coating. Cryst. Growth Des. 2013;13(4):1438–1444 12 [21] Zhang Z. A low-temperature and low-cost method to produce high-quality epitaxial 13 anatase TiO2 thin films. J. Mater. Res. 2005;20(2):292–294. 14 [22] Ono K, Shimomura M. Solvothermal Epitaxy of an Anatase TiO2 Thin Film on SrTiO2 15 (001) and the Effect of Precursor Composition on Its Growth. Cryst Growth Des. 16 2024;24(4):1798–1805.  17 [23] Ono K, Kimura K, Kato T, et al. Epitaxial growth of a homogeneous anatase TiO2 thin 18 film on LaAlO3 (001) using a solvothermal method with anticorrosive ligands. Chem. Eng. J. 19 2023;451(3):138893.  20 [24] Xu H, Feng X, Luan C, et al. Preparation and characterization of single crystalline 21 anatase TiO2 epitaxial films on LaAlO3(001) substrates by metal organic chemical vapor 22 deposition. Scr Mater. 2016;124:76–80.  23 [25] Hong D, Liu C, Pearson J, et el. Epitaxial growth of high quality SrFeO3 films on (001) 24 oriented (LaAlO3)0.3(Sr2TaAlO6)0.7. Appl Phys Lett. 2017; 111(23):232408. 25 17  [26] Wang WL, Peng CY, Ho YT, et al. Microstructure of a-plane ZnO grown on LaAlO3 1 (001). Thin Solid Films. 2010 Mar 31;518(11):2967–2970. 2 [27] Hayward SA, Morrison FD, Redfern SAT, et al. Transformation processes in LaAlO3: 3 Neutron diffraction, dielectric, thermal, optical, and Raman studies. Phys. Rev. B. 2005;72(5): 4 054110.  5 [28] Zhao J, Angel RJ, Ross NL. The structural variation of rhombohedral LaAlO3 perovskite 6 under non-hydrostatic stress fields in a diamond-anvil cell. J. Phys.: Condens. Matter. 7 2011;23(17):175901.  8 [29] Kennedy RJ, Stampe PA. The influence of lattice mismatch and film thickness on the 9 growth of TiO2 on LaAlO3 and SrTiO3 substrates. J Cryst Growth. 2003;252(1–3):333–342.  10 [30] Wijekoon S.H.D.P, Shimomura M, Kawaguchi T, et al.  Smooth TiO2 thin film 11 fabrication by on-site controlled hydrolysis of alcohol-titanium alkoxide mixtures. Surf. 12 Interfaces. 2025;58: 105755. 13 [31] Granada-Ramirez DA, Cardona-Bedoya JA, Hernandez-Rojas U, et al. Assessment of Cr 14 doping on TiO2 thin films deposited by a wet chemical method. Ceram Int. 2023;49(18):30347–15 30354.  16 [32] Hua L, Yin Z, Cao S. Recent Advances in Synthesis and Applications of Carbon-Doped 17 TiO2 Nanomaterials. Catalysts. 2020;10(12):1431. 18 [33] Qi D, Xing M, Zhang J. Hydrophobic carbon-doped TiO2/MCF-F composite as a high 19 performance photocatalyst. J. Phys. Chem. C. 2025;118(14):7329–7336.  20 [34] Wu Y, Zhang J, Xiao L, et al. Properties of carbon and iron modified TiO2 photocatalyst 21 synthesized at low temperature and photodegradation of acid orange 7 under visible light. Appl 22 Surf Sci. 2010;256(13):4260–4268.  23 [35] Playford HY. Variations in the local structure of nano-sized anatase TiO2. J. Solid State 24 Chem. 2020; 288:121414. 25 18  [36] Izumisawa F, Ishii Y, Kimura M, et al. Symmetry change in LaNiO3 films caused by 1 epitaxial strain from LaAlO3, SrTiO3, and DyScO3 pseudocubic (001) surfaces. J. Appl. Phys. 2 2024;136(7):75303.  3 [37] Jung HS, Lee JK, Lee J, et al. Mobility enhanced photoactivity in sol-gel grown epitaxial 4 anatase TiO2 films. Langmuir. 2008 Mar 18;24(6):2695–2698. 5 [38] Ohring M. The Material science of thin films. San Diego: Academic Pres; 1992. Chapter 6 7, Epitaxy; p.307-353 . 7 [39] Olson GL, Roth JA. Kinetics of solid phase crystallization in amorphous silicon. 8 Mater. Sci. Rep. 1988 May 3: 1-78 9 [40] Lanthanum Aluminate - SHINKOSHA Crystals for a bright future. Shizuoka. [Internet]. 10 [cited 2025 May 3]. Available from: 11 https://www.shinkosha.com/english/product/epi_substrate/lao/ 12 [41] Hummer DR, Heaney PJ, Post JE. Thermal expansion of anatase and rutile between 300 13 and 575 K using synchrotron powder X-ray diffraction. Powder Diffr. 2007 Dec;22(4):352–7.  14 [42] Patel AR, Rao K V, Shivakumar GK. Role of Deposition Conditions, Impurities and 15 Structural Defects in the Epitaxial Growth of Thin Films. Appl Phys. 1981; 24:85–8. 16 [43] Mieghem P Van, Jain SC, Nijs J, et al. Stress relaxation in laterally small, strained 17 semiconductor epilayers.  J. Appl. Phys. 1994 Jan; 75 (1): 666-668  18 [44] Dunstan DU. Review Strain and strain relaxation in semiconductors. J. Mater. Sci.: 19 Mater. Electron. 1997;8:337–375. 20  21  22   23 https://www.shinkosha.com/english/product/epi_substrate/lao/19  Tables and captions 1 Table 1 Summary of surface properties and crystalline quality of epitaxial films 2 Solvent composition FWHM of rocking curves at 700/ (degrees) FWHM of rocking curves at 800/ (degrees) Surface roughness (Rz) At 800/ (nm) TTIP-nPr 0.80 0.43 15.6 TTIP-nBu 0.86 0.46 16.3 TTBu-nPr 0.75 0.48 17.6 TTBu-nBu 0.78 0.37 13.5  3 Figures and captions 4  5  6 Figure 1. The schematic diagram of the procedure of synthesizing epitaxial anatase films 7  8  9  10 20   1 Figure 2.  XPS spectra of thin films annealed at 800 °C. (a) Survey spectra obtained for different 2 thin films made using different solvent mixtures (b) Ti 2p region of deconvoluted high resolution 3 XPS spectrum of the film made from TTIP-nPr (c) O 1s region of deconvoluted high resolution 4 XPS spectrum of the film made from TTIP-nPr . (d) C 1s region of deconvoluted high resolution 5 XPS spectrum of the film made from TTIP-nPr . 6 a b c d 534 533 532 531 530 529 528 527Intensity (a.u.)Binding energy (eV)OOHOL468 466 464 462 460 458 456Intensity (a.u.)Binding energy (eV)Ti 2P3/2Ti 2P1/2291 290 289 288 287 286 285 284 283 282Intensity (a.u.)Bindinng energy (eV)O=CO-CC-CH21   1 Figure 3. XRD 2θ/ω of thingies made by different solvent mixtures annealed at (a) 700 °C and (b) 2 800 °C **(c) reference 2θ positions of anatase and LaAlO3 3 log 10(Intensity)Two thetaa  TTIP-nPr TTIP-nBu TTBu-nPr TTBu-nBuLaAlO3 001LaAlO3 002Anatase 004b20 25 30 35 40 45 50c  Anatase LaAlO3a b 22   1 Figure 4.  XRD rocking curves of the thin film made by different solvent mixtures annealed at (a) 2 700 °C and (b) 800 °C  3  4 Figure 5 . XRD ϕ scan results for (101) TiO2 and (101) at 2ϴ =33.4, ψ = 45 for substrate and 2ϴ 5 = 25.3, ψ = 68.3 for anatase thin film made by TTBu-nBu solvent mixture and annealed at 800°C 6  7  8  9  10  11  12  13  14 Figure 6. Orientation analysis by pole figures of LAO {101} and anatase {101} of the thin film 15 prepared by TTIP-nPr and annealed at 800°C 16 LAO 2ϴ =33.4° ψ = 45° Anatase 2ϴ =25.3° ψ = 68.3° 23   1 Figure 7. SEM images of each thin of each thin film and annealed at 800°C (a) TTIP-nPr, (b) 2 TTIP-nBu, (c) TTBu-nPr, (d) TTBu-nBu (e) 250°C annealed thin film prepared by TTIP-nBu 3 solvent mixture and (f) LaALO3 substrate (g) Cross sectional view of thin film made by 4 TTIP=nPr solvent mixture ** at 15 kV tension and 8 mA probe current** 5 a b c d e f TiO2 film Substrate g 24   1  2  3  4 Figure 8.  AFM 3D images of each thin of each thin film and annealed at 800°C (a) TTIP-nPr, 5 (b) TTIP-nBu, (c) TTBu-nPr, (d) TTBu-nBu  6  7  8 a Rz = 15.6nm  d Rz = 13.5nm  b Rz = 16.3nm  c Rz = 17.6 nm  25   1 Supplementary materials 2  3  4 Figure S1. deconvoluted high resolution XPS spectra of Ti 2p and O 1s regions thin films annealed 5 at 800 °C. (a) Ti 2p region of TTIP-nPr film (b) Ti 2p region of TTBu-nPr film, (c) Ti 2p region 6 of TTBu-nBu, (d) O1s region of TTIP-nPr film, (e) O1s region of TTBu-nPr film, (f) O1s region 7 of TTBu-nBu film (g) C1s region of TTIP-nPr film (h) C1s region of TTBu-nPr film and (i) C1s 8 region of TTBu-nBu film. 9  10 468 466 464 462 460 458 456Intensity (a.u.)Binding energy (eV)Ti 2P1/2Ti 2P3/2468 466 464 462 460 458 456Intensity (a.u.)Binding energy (eV)Ti 2P3/2Ti 2P1/2468 466 464 462 460 458 456Intensity (a.u.)Binding energy (eV)Ti 2P3/2Ti 2P1/2534 533 532 531 530 529 528 527Intensity (a.u.)Binding energy (eV)OOHOL534 533 532 531 530 529 528 527Intensity (a.u.)Binding energy (eV)OOHOL534 533 532 531 530 529 528 527Intensity (a.u.)Binding energy (eV)OOHOL291 290 289 288 287 286 285 284 283 282Intensity (a.u.)Bindinng energy (eV)C-CHO-CO=C291 290 289 288 287 286 285 284 283 282Intensity (a.u.)Bindinng energy (eV)C-CHO-CO=C291 290 289 288 287 286 285 284 283 282Intensity (a.u.)Bindinng energy (eV)C-CHO-CO=Ca b c d e f i h g 26   1 Figure S2. Orientation analysis by pole figures of LAO {101} and anatase {101} of the thin film 2 prepared at annealed at 800°C (a) TTIP-nBu film (b) TTBu-nPr film, and (c) TTBu-nBu film. 3  4 Anatase 2ϴ =25.3° ψ = 68.3° LAO 2ϴ =33.4° ψ = 45° a b c 27   1 Figure S3. XRD rocking curves of each LaAlO3 substrates 800 °C 2 23.7 23.8 23.9 24.0 24.1 24.2020k40k60k80k100k120k140k160kIntensity(cps)omega (degrees) TTIP-nPr TTIP-nBu TTBu-nPr TTBu-nBu 1. Introduction 2. Experimental 2.1.  Materials and method 2.2. Characterization 3. Results and discussion 4. Conclusions Acknowledgement Funding Disclosure statement Data Availability Statement References. Supplementary materials