Abstract

Titanium dioxide (TiO₂) is used extensively in products from pigments and sunscreens to optical components. The sol–gel synthesis of TiO₂ is controlled by an intricate set of interactive parameters of which optimization is an important issue. A set of 290 experimental conditions was studied in detail to model and optimize yield of TiO₂ by means of statistical and machine learning methodologies. Out of the methodologies studied, polynomial regression and optimized random forest models showed best predictive capability achieving coefficient of determination (R²) of 0.9522 and 0.9314, respectively, in comparison to linear regression. Feature importance analysis identified precursor concentration and hydrolysis ratio (water-to-precursor ratio) to play key role by having predominant influence, with secondary influence being aging time and pH. The paper highlights the value of data-based methodologies for synthesis design guidance, improved reproducibility, and expedited advances in materials chemistry.

References

• Chen, X., & Selloni, A. (2014). Introduction: titanium dioxide (TiO2) nanomaterials. Chemical reviews, 114(19), 9281-9282.
• Roduner, E. (2006). Size matters: why nanomaterials are different. Chemical society reviews, 35(7), 583-592.
• Chen, X., & Mao, S. S. (2007). Titanium dioxide nanomaterials: synthesis, properties, modifications, and applications. Chemical reviews, 107(7), 2891-2959.
• Liu, L., & Chen, X. (2014). Titanium dioxide nanomaterials: self-structural modifications. Chemical reviews, 114(19), 9890-9918.
• Meuwly, M. (2021). Machine learning for chemical reactions. Chemical Reviews, 121(16), 10218-10239.
• Adrah, F. A., Mottey, B. E., & Nyavor, H. (2024). The landscape of artificial intelligence applications in health information systems. Int. J. Comput. Appl., 975, 8887.
• Agboklu, M., Adrah, F. A., & Agbenyo, P. M. (2024). Hope Nyavor. From bits to atoms: machine learning and nanotechnology for cancer therapy. Journal of Nanotechnology Research, 6, 16-26.
• Nayebzadeh, H., Rohani, A., Sistani, A., Hassanpour, A., & Gardy, J. (2022). Modelling and optimisation of the sol-gel conditions for synthesis of semi-hexagonal titania-based nano-catalyst for esterification reaction. Catalysts, 12(2), 239.
• Alayou, S., Mengesha, M., & Tizazu, G. (2025). Application of machine learning models for predicting zinc oxide nanoparticle size. Measurement, 117785.
• Demirci, S., Şahin, D. Ö., & Demirci, S. (2025). Design of the amorphous/crystalline TiO2 nanocomposites via machine learning for photocatalytic applications. Materials Science in Semiconductor Processing, 192, 109460.
• Karatchevtseva, I., Cassidy, D. J., Zhang, Z., Triani, G., Finnie, K. S., Cram, S. L., ... & Bartlett, J. R. (2008). Crystallization of TiO2 powders and thin films prepared from modified titanium alkoxide precursors. Journal of the American Ceramic Society, 91(6), 2015-2023
• Hanaor, D. A., Chironi, I., Karatchevtseva, I., Triani, G., & Sorrell, C. C. (2012). Single and mixed phase TiO2 powders prepared by excess hydrolysis of titanium alkoxide. Advances in Applied Ceramics, 111(3), 149-158.
• Guo, W., Lin, Z., Wang, X., & Song, G. (2003). Sonochemical synthesis of nanocrystalline TiO2 by hydrolysis of titanium alkoxides. Microelectronic Engineering, 66(1-4), 95-101
• Karino, S., & Hojo, J. (2010). Synthesis and characterization of TiO2-coated SiO2 particles by hydrolysis of titanium alkoxide in alcohol solvents. Journal of the Ceramic Society of Japan, 118(1379), 591-596.
• Jasbi, N. E., & Dorranian, D. (2016). Effect of aging on the properties of TiO2 nanoparticle. Journal of Theoretical and Applied Physics, 10(3), 157-161.
• Hsiang, H. I., & Lin, S. C. (2004). Effects of aging on the phase transformation and sintering properties of TiO2 gels. Materials Science and Engineering: A, 380(1-2), 67-72.
• Aida, I. S., & Sreekantan, S. (2011). Effect of pH on TiO2 nanoparticles via sol-gel method. Advanced Materials Research, 173, 184-189.
• Yalcin, M. (2022). The effect of pH on the physical and structural properties of TiO2 nanoparticles. Journal of Crystal Growth, 585, 126603.
• Chen, Y. F., Lee, C. Y., Yeng, M. Y., & Chiu, H. T. (2003). The effect of calcination temperature on the crystallinity of TiO2 nanopowders. Journal of crystal growth, 247(3-4), 363-370
• Hu, L., Yoko, T., Kozuka, H., & Sakka, S. (1992). Effects of solvent on properties of sol—gel-derived TiO2 coating films. Thin solid films, 219(1-2), 18-23.
• Rao, Y., Antalek, B., Minter, J., Mourey, T., Blanton, T., Slater, G., ... & Fornalik, J. (2009). Organic solvent-dispersed TiO2 nanoparticle characterization. Langmuir, 25(21), 12713-12720.
• Kojima, T., & Sugimoto, T. (2008). Formation mechanism of amorphous TiO2 spheres in organic solvents 3. Effects of water, temperature, and solvent composition. The Journal of Physical Chemistry C, 112(47), 18445-18454.
• Mohammadi, S., Harvey, A., & Boodhoo, K. V. (2014). Synthesis of TiO2 nanoparticles in a spinning disc reactor. Chemical Engineering Journal, 258, 171-184.
• Adrah, F. A., Denu, M. K., & Buadu, M. A. E. (2023). Nanotechnology applications in healthcare with emphasis on sustainable COVID-19 management. J Nanotechnol Res, 5, 6-13.