[Objectives] Diamond, with advantages of high melting point, good insulation, stable chemical properties, and high thermal conductivity, is a promising semiconductor material for high-power devices, sensors, and quantum computing. The best method for producing high-quality single crystal diamonds through microwave plasma chemical vapor deposition (MPCVD) is homoepitaxial growth, which requires expensive single crystal diamond substrates. Therefore, heteroepitaxial growth on foreign substrates is a compromise approach that has to be adopted, where the selection of a proper substrate becomes a major challenge for growing high-quality diamonds. To enhance the quality of heteroepitaxial growth, potassium tantalum niobate (KTa1-xNbxO3, referred to as KTN) crystals are proposed as the substrate. Their lattice parameter (0.3994 nm) is close to that of diamond (0.3567 nm). The structural similarity between them makes it possible to achieve an ideal result with small lattice mismatch and high crystal quality.

[Methods] Diamond thin films were grown on large-size, high-quality KTN single crystals using MPCVD. The effect of different growth times on diamond quality was explored by varying the duration of diamond growth. The surface morphologies, microstructures, and crystalline quality were analyzed using scanning electron microscope (SEM), X-ray diffraction (XRD) and Raman spectroscopy. Diamond samples with different growth times were used to degrade Rhodamine B (RhB) to assess their photocatalytic effect. The absorbance of RhB was measured to evaluate the degradation efficiency of diamond films as catalysts for the pollutant solution.

[Results ] Analysis of plasma diagnostic spectra confirmed the presence of H groups and carbon source groups, while CN groups were absent. The H group exhibited the highest spectral line intensity, higher than that of the C2 group. This indicated that the system was well-sealed and provided sufficient reactive groups for diamond growth. The ratio of peak intensities between Hα and Hβ was high, while the ratio between C2 and Hα was low, both of which were favorable for the deposition of high-quality diamonds. Raman spectroscopy revealed structural changes in diamonds through comparing Isp3/Isp2, the relative intensities of diamond-phase characteristic peaks and non-diamond-phase characteristic peaks. The ratio gradually increased with the extension of growth time, indicating that the quality of diamonds was improving. XRD analysis of diamond films showed characteristic diamond peaks. In the 6h sample, characteristic peaks appeared at 34.8° (111), 40.5° (200), 58.6° (220) and 70.0° (311), consistent with the standard card (JCPDS No. 35-0801) as the characteristic peaks of TaC, indicating the presence of TaC in the 6h sample. However, the TaC characteristic peaks were lower in the 9h sample, and only diamond characteristic peaks were observed in the 12h sample. This indicated that TaC was only a transitional layer formed in the pre-nucleation stage, and TaC was no longer generated with the extension of growth time. From the SEM images of diamond films at different deposition time, it could be seen that the grain size of the 12h sample was much larger and more evenly distributed than those of other samples. The surface of the 12h sample was mostly composed of (100) faceted grains with a high degree of flatness, and the grains growing at the grain boundaries tended to grow significantly larger. Analysis of photocatalytic activity showed that higher diamond purity and lower non-diamond phase proportion led to better photocatalytic activity, more stable photocatalytic performance, and higher reusability.

[Conclusions] A new diamond substrate, potassium tantalum niobate (KTa1-xNbxO3), was successfully used to grow high-quality diamond films using MPCVD technology. The samples were characterized by XRD, SEM and Raman spectroscopy. Results indicate the formation of a TaC transition layer when growing diamond films on the KTN substrate, which is favorable for stable diamond growth. Moreover, with increasing growth time, the diamond grain size increases, the diamond phase content increases, and the sample quality improves. Photocatalytic results show that the sample grown for 12 hours has stronger photocatalytic ability and higher photocatalytic stability, achieving a degradation efficiency of 91.9% for RhB, which is 1.6 times higher than that of the 3h sample. These findings broaden the application of diamond and provide valuable insights for substrate selection in diamond fabrication.