Crystals

To evaluate the influence of different surface conditioning protocols—sandblasting (SB), tribochemical silica coating (TBC), CO₂ laser irradiation, and plasma-enhanced chemical vapor deposition (PECVD-Si coating for 49 min) on surface roughness (Ra), surface morphology, and composite-to-zirconia shear bond strength (SBS). Eighty 3Y-TZP plates were randomly allocated into four groups (n = 20) based on surface conditioning protocol: Group 1 (SB), Group 2 (CO₂ laser), Group 3 (TBC), and Group 4 (PECVD-Si coating for 49 min). From each group, five specimens underwent Ra assessment using a contact profilometer, and five specimens were examined for surface morphology via scanning electron microscopy (SEM). The remaining ten specimens received resin composite buildup, followed by artificial aging. Subsequently, SBS testing was performed using a universal testing machine, and failure modes were evaluated under a stereomicroscope. Statistical analysis was conducted using one-way ANOVA with post hoc Tukey test and chi-square for fracture assessment(α = 0.05). Group 1 (SB) demonstrated the lowest Ra (0.844 ± 0.063 µm) and SBS (12.21 ± 4.6 MPa), whereas Group 4 (PECVD-Si coating for 49 min) exhibited the highest Ra (1.388 ± 0.098 µm) and SBS (30.48 ± 2.5 MPa). Intergroup comparison revealed no statistically significant differences between Groups 2 and 3 for both Ra and SBS values (p > 0.05). However, Groups 1 and 4 differed significantly in both parameters (p < 0.05). PECVD-based silica coating for 49 min demonstrated superior surface conditioning efficacy for 3Y-TZP, yielding significantly higher Ra and SBS values compared to sandblasting, tribochemical silica coating, and CO₂ laser irradiation.

Microwave sintering enabled the efficient fabrication of bulk Mn₃Cu_0.5Ge_0.5N_0.9C_0.1 NTE materials in 3–5 h, versus 2 to 8 days for conventional methods. The microwave approach demonstrated high efficiency and energy savings. By adjusting temperature and dwell time, the NTE operating range can be shifted to lower temperatures. Under the optimized condition of 800 °C for 4 h, the resulting bulk material achieved an NTE coefficient of −20.56 × 10⁻⁶ K⁻¹ over a temperature interval ΔT of 88 K (from 159 K to 247 K), along with favorable densification and high hardness. The demonstrated processing efficiency, microstructural control, and tunable NTE properties establish a solid foundation for potential industrial scale-up.

High-precision prediction of the crystal diameter during the growth of electronic-grade silicon single crystals is a critical step for the fabrication of high-quality single crystals. However, the process features high-temperature operation, strong nonlinearities, significant time-delay dynamics, and external disturbances, which limit the accuracy of conventional mechanism-based models. In this study, mechanism-based models denote physics-informed heat-transfer and geometric models that relate heater power and pulling rate to diameter evolution. To address this challenge, this paper proposes a hybrid deep learning model combining a convolutional neural network (CNN), a bidirectional long short-term memory network (BiLSTM), and self-attention to improve diameter prediction during the shoulder-formation and constant-diameter stages. The proposed model leverages the CNN to extract localized spatial features from multi-source sensor data, employs the BiLSTM to capture temporal dependencies inherent to the crystal growth process, and utilizes the self-attention mechanism to dynamically highlight critical feature information, thereby substantially enhancing the model’s capacity to represent complex industrial operating conditions. Experiments on operational production data collected from an industrial Czochralski (Cz) furnace, model TDR-180, demonstrate improved prediction accuracy and robustness over mechanism-based and single data-driven baselines, supporting practical process control and production optimization.