Empirical interatomic potentials remain widely used due to their high computational efficiency. Yet their development and validation protocols typically emphasize properties that have limited sensitivity to the overall shape of the potential energy landscape (PEL) and can give rise to poor transferability to other properties. Here, we consider the configurational and vibrational entropy of liquid silicon (Si) predicted by seven widely utilized empirical potentials with a view of using these energy landscape parameters to assess potentials more broadly. Both vibrational and configurational entropy contributions are found to be remarkably sensitive to the form of the potential, revealing substantial differences in basin curvature, basin density, and overall PEL topology that are not captured by conventional fitting targets. With a view toward using this information for future potential development, we also show that the Debye-Waller parameter u2 provides a practical experimental proxy for configurational entropy through the localization model of relaxation.

The 8-inch silicon carbide (SiC) crystal grown by physical vapor transport (PVT) method is becoming the mainstream in the semiconductor market. Induction heating is widely used in the SiC crystal growth and theoretically it can create a temperature distribution that is beneficial for the formation of a convex crystallization interface, whereas experimental observations often reveal a concave interface. In this study, the problems of powder recrystallization and its influence on crystallization interface shape for 8-inch SiC crystal growth by PVT method are studied by experiments and numerical simulations. It's found that the low temperature at the top central region of the SiC powder leads to recrystallization of the sublimated powder and formation of a blockage, which results in a concave crystallization interface shape. Two techniques including temperature control and powder height control are proposed to influence the recrystallization and the interface shape. Results show that the powder recrystallization can be completely avoided by reducing the powder height, whereas the interface shape presents significant deformation at the crystal periphery. Therefore, the effect of powder recrystallization is not always negative, and it can be used to obtain a slight convex interface shape for high-quality SiC crystal growth.

Based on previously developed LES-DPM model, this study focused on the effects of feeding port configurations on the melting and migration characteristics of silicon particles during continuous Czochralski monocrystalline silicon growth. And the melt temperature distribution, particularly temperature fluctuations in the melt near the crystallization interface were also analyzed. The results show that symmetrical arrangement of multiple feeding ports to disperse the silicon particles can effectively prevent the formation of local low-temperature areas around the feeding ports. However, this arrangement also enlarges the influence area of thermal disturbance near the melt free surface. And excessive dispersion of feeding ports not only fails to significantly shorten particle melting time but also increases the complexity of feeding device. In addition, reducing the distance between feeding port and crucible wall can effectively promote rapid particles melting. Meanwhile, more concentrated particle movement helps improve the uniformity of temperature distribution near the triple point, thereby enhancing the stability of crystal growth. However, this also leads to increased impacts and contacts between silicon particles and crucible wall, shortening the crucible's service life.

Sub‐1 nm high‐entropy materials (SNHEMs) have garnered significant attention for their exceptional catalytic activity and stability in electrocatalysis and energy storage. Compared with conventional nanomaterials, SNHEMs effectively mitigate aggregation, and they offer higher exposure of active sites than traditional high‐entropy materials (HEMs). These advantages arise from their ultrahigh specific surface area, abundant active sites, and tailorable electronic/band structures. As SNHEMs emerge as a cutting‐edge field for advanced energy materials, a comprehensive review of this class of materials is imperative. Herein, we summarize the primary controllable synthesis strategies of SNHEMs; systematically discuss recent advances in pivotal electrocatalytic reactions, such as the hydrogen evolution reaction (HER), oxygen evolution reaction (OER), and oxygen reduction reaction (ORR); and elaborate on their applications in energy storage devices, including sodium‐ion batteries (SIBs), lithium‐oxygen batteries (LOBs), and zinc‐air batteries (ZABs). Finally, we highlight the current research challenge and outline promising directions for future development.

The distribution of phosphorus dopant in silicon has an important influence on the resistivity of silicon crystals. A two-dimensional transient global model is established in this paper to investigate the transport of phosphorus dopant during the iso-diameter growth of 200 mm Czochralski monocrystalline silicon. We focused on the dynamic consumption process of phosphorus dopant in the silicon melt during crystal growth, and analyzed the effects of segregation and evaporation behavior, as well as the turbulent viscosity on the transport of phosphorus dopant. The results reveal that the concentration of phosphorus dopant at crystal-melt interface exhibits a trend in which the concentration is elevated at the center compared to the edge. Furthermore, the phosphorus dopant concentration in the melt progressively increases with crystal growth, resulting in a higher content in the tail of monocrystalline silicon compared to the head. By comparing with the experimental data, it is confirmed that when calculating the transport of phosphorus dopant in the melt, the melt turbulent viscosity and the evaporation behavior of phosphorus needs to be considered. At the same time, we also determined the evaporation coefficient for accurately calculating phosphorus dopant distribution.

This study employs density functional theory and quantum chemical calculations to investigate the gas-phase nucleation and growth mechanisms during chemical vapor deposition of silicon carbide. Based on thermodynamic stability evaluations of large clusters under various configurations, the lowest-energy ground-state structure was determined. The Gibbs free energy (ΔG) calculations of pure silicon clusters (Sin), single-carbon silicon clusters (SinC), and double-carbon silicon clusters (SinC2) were conducted at different temperatures. The findings reveal that silicon atoms promote cluster growth. The special 2D-to-3D configurational transition attenuates the reaction’s spontaneity. During the initial nucleation stage, the system tends to form SinC; however, as the size increases, it evolves into the more stable SinC2. This study reveals gas-phase cluster formation at the atomic scale, providing a theoretical foundation for suppressing detrimental gas-phase nucleation.

To understand and address the thermal stress and cracking issues in β-Ga₂O₃ crystal growth by the edge-defined film-fed growth (EFG) method, a global three-dimensional anisotropic thermo-mechanical coupling model is established to investigate the influence mechanisms of anisotropic heat transfer, thermal expansion, and elastic behavior on the thermal field and thermal stress distributions. Results indicate that the anisotropic heat transfer within β-Ga₂O₃ crystal has limited effects on the temperature distributions and crystallization interface shapes for different growth orientations. The anisotropic thermal expansion and thermoelastic behavior are the key factors contributing to thermal stress differences in crystals with different growth orientations. Specifically, for crystals grown along [010] with the (001) principal plane and for those grown along [001] with the (010) principal plane, the concentrations of von Mises stress and first principal stress are higher than those for the [010]-(100) and [100]-(010) crystal orientations. Analysis from the perspective of resolved normal stress reveals that when growing with the (010) principal plane, crystals generally experience large tensile resolved normal stress on the (100) cleavage plane. Conversely, crystals grown along [010] with (100) principal plane demonstrate relatively small tensile resolved normal stress. This research elucidates the cracking mechanisms in β-Ga₂O₃ growth across specific orientations and lays a key theoretical foundation for optimizing growth processes and reducing defect risks that is extensible to the growth of various anisotropic brittle crystals.

Sub-1 nm materials (SNMs) feature large specific surface areas and near-100% atomic exposures, thus offering abundant exposed active sites and properties well suited for applications in optics, electricity, catalysis, etc. In particular, polyoxometalate (POM)-based SNMs hold great promise for energy storage/conversion applications because of their nanoscale dimensions and synergetic effect of multiple components. Despite the abundance of reported POM-based SNMs, systematic reviews of their synthesis and energy storage/conversion applications are scarce. This review summarizes progress in the synthesis of POM-based SNMs and their applications in energy storage (e.g., Li-ion, Na-ion, Liu2013O2, Znu2013O2, and Liu2013S batteries) and conversion (e.g., photothermal conversion, solar steam generation, and seawater desalination). Moreover, we discuss the related challenges and future perspectives, providing new insights into the rational design and development of efficient POM-based SNMs suitable for energy storage and conversion.

Energy saving in Czochralski (CZ) silicon growth is of great significance for the cost reduction of photovoltaic power generation. As a critical component of the CZ furnace, the heat shield directly influences the heat flow transport within the CZ furnace and subsequently affects the heating power. Therefore, a systematic study of the heat transfer characteristics of the heat shield on heat flow transport and heating power in the CZ furnace is essential for achieving low-power-consumption growth of CZ silicon. In this work, based on the 2D global numerical model, the effects of radiative and conductive heat transfer of the heat shield on heat flow and heating power in the CZ furnace were investigated. The results indicate that the decrease of emissivity and the increase of thickness at the bottom of the heat shield can effectively reduce both the heating power and crystallization interface deflection, which is attributed to the suppression of radiative heat transfer from the furnace to the heat shield and conductive heat flux through the heat shield. Comparatively, the decrease of heat shield emissivity has a more pronounced energy-saving effect. An optimized heat shield configuration, featuring an emissivity of 0.1 and a thickness of 260 mm, can reduce the heating power by 8.82 % and the crystallization interface deflection by 13.84 %. This study provides a practical reference for reducing heating power in CZ silicon growth.

Based on quantum chemical theory, the gas-phase chemical cross-reaction pathways in metal–organic chemical vapor deposition of InAlN films have been investigated. First, calculations of the oligomerization reactions of Al and In amides were performed to investigate the relative difficulty of generating several dimers and trimers within this system. The elimination reaction pathways were then investigated. The final stable products of the polymers were analyzed from both thermodynamic and kinetic perspectives, and the propensities of the reaction pathways were assessed from the viewpoint of the reaction rates. The results indicated that in the gas-phase reaction pathways of InAlN grown by metal–organic chemical vapor deposition, the equilibrium temperature for the dimerization reactions lies between 960 and 1061 K, and above this temperature range, dimers cannot form. Trimer formation is unlikely during the InAlN growth process. At higher reaction temperatures, the oligomers tend to first undergo intramolecular elimination, losing two CH4 molecules to form intermediate products, which then react with NH3 via intermolecular elimination to yield stable products. At lower reaction temperatures, the tendency is for continuous intermolecular elimination with NH4 to generate stable, methyl-free products.

The 1D sub-1 nm nanowires (SNWs) usually possess a highly promising avenue in catalysis and energy storage fields owing to their high aspect ratios facilitating charge transport, and near 100% surface atomic exposure offering abundant active sites. However, current research is mainly focused on metal oxide SNWs, whereas the synthesis of sulfide SNWs has been rarely reported. Herein, a new kind of Bi2S3-phosphomolybdic acid (Bi2S3-PMA) SNWs induced by polyoxometalate clusters has been successfully prepared. Molecular dynamics simulation demonstrates that PMA and Bi2S3 co-assemble into stable 1D SNWs via non-covalent interactions. Benefiting from the unique sub-1 nm structure and the synergetic effects of Bi2S3 and PMA, the SNWs exhibit enhanced light absorption ability, well-matched energy band structure, and efficient separation/transfer capability of photo-generated carriers. As the cathode catalyst in light-assisted Li-CO2 batteries (LCBs), the Bi2S3-PMA-based LCBs deliver a low overpotential of 0.22 V, superior cycling stability for 300 h at 0.01 mA cm-2 and 150 h at 0.05 mA cm-2. Meanwhile, the battery also realizes an exceptionally long-term lifetime of 4000 h under no light. Density functional theory calculations disclose that the presence of electron-rich PMA promotes the adsorption of LiCO2 and Li2CO3 on SNWs, which further boosts battery efficiency.

Induction heating is commonly used in the top-seeded solution growth (TSSG) of SiC crystals to provide heating power and at the same time generates Lorentz force in the solution, namely the thermal effect and the magnetic effect. Therefore, studying the thermal-magnetic effects is critical to minimize the system power consumption and improve the crystal growth simultaneously. In this study, a global numerical model was established to calculate the induction heating, heat transfer, solution flow and carbon transport in the SiC crystal growth by TSSG method. The combined thermal-magnetic effects were systematically studied to find out optimal heating parameters that can simultaneously utilize both effects. The results indicate that excessively high or low frequencies increase the total heating power consumption and reduce the induction heating efficiency. The optimal frequency range is 1-2 kHz in this study, for which the minimum total power consumption is 34.4 kW and the maximum heating efficiency is 82.0%. The Lorentz force in the solution changes significantly with the increase of frequency, and thus influences the crystal growth parameters. It's found that the growth rate is relatively high and uniform at 1-2 kHz. Therefore, the thermal-magnetic effects can be utilized simultaneously to optimize the SiC crystal growth.

During the preparation of silicon carbide (SiC) thin films by chemical vapor deposition (CVD), the Soret effect induced by a large temperature gradient influences the deposition rate and uniformity; its sensitivity to process parameters remains unclear. A computational fluid dynamics model coupling detailed gas-phase and surface reaction kinetics was developed and validated for a cold/warm wall vertical CVD reactor. Comparing simulations with and without the thermal diffusion term reveals the dual role—suppressing deposition rate while degrading film uniformity. The thermal diffusion contributions to deposition rate (TDC_GR) and uniformity (TDC_GU) are introduced as quantitative metrics, and simulations evaluated the effects of inlet–substrate temperature difference (ΔT), reactor pressure (p), substrate rotation speed (ω), and carrier gas flow rate (Q) on the Soret effect, clarifying optimal conditions. Results show ΔT dominates. At ΔT = 1700 K, TDC_GR = −56.39% and TDC_GU = 5.29%. Pressure affected TDC_GR negligibly but significantly reduced TDC_GU by enhancing gas-phase mixing; increasing p from 7500 to 12,500 Pa led to a decrease in TDC_GU from 4.19% to 1.93%. Optimal parameters (ΔT = 1400 K, p = 12,500 Pa, ω = 800 rpm, Q = 50 slm) achieved a deposition rate of 10.71 μm/h and non-uniformity of 0.45%. These findings provide theoretical guidance for precise SiC-CVD process control in cold- or hot-wall vertical reactor architectures.

During drilling operations, formation gas readily invades the wellbore, potentially causing a kick. This study investigated the effects of wellbore size, drilling fluid density, bottomhole pressure difference, and reservoir thickness on gas migration and the gas–liquid two-phase flow during kick development in large-diameter wellbores. A three-dimensional numerical model of a large wellbore was constructed using the VOF multiphase flow model and the SST k–ω turbulence model. Transient simulations were performed to analyze gas–liquid flow patterns under various conditions. Key parameters studied included wellbore size, drilling fluid density, bottomhole pressure difference, and reservoir thickness. Gas–liquid volume fraction contour plots helped interpret the influence of different parameters. Results show that, under otherwise identical conditions, smaller wellbore size, lower drilling fluid density, negative pressure difference (underbalance), and thicker reservoir lead to faster gas migration rates. These conditions also create more complex gas–liquid mixture flow which significantly increase kick risk. Conversely, larger wellbore size, higher density of drilling fluid, positive pressure difference, and thinner reservoir reduce the gas migration velocity. These conditions effectively control gas influx. This research quantifies the impact of the four key factors on annular kick flow in large-diameter wells. It establishes the VOF-SST numerical model. The findings provide theoretical support for optimizing well control parameters in deep wells and preventing blowout accidents.

Numerical simulation is an effective approach for improving the edge-defined film-fed growth (EFG) of β-Ga2O3 crystals by studying the heat transfer, melt flow, and crystallization interface shape. However, the nonaxisymmetric structure of the EFG furnace, the anisotropy of the β-Ga2O3 crystal, the infrared radiation absorption by free carriers, and the tracking of three-dimensional (3D) crystallization interface shape bring challenges in modeling and simulation. In this study, a global 3D nonaxisymmetric numerical model was established for the growth of ribbon β-Ga2O3 crystals. The anisotropic thermal conduction coupled with the thermal radiation inside the crystal was considered, and the dynamic mesh technique was developed to track the severely deformed crystallization interface. Subsequently, the numerical model was used to study the influences of thermal radiation absorption by free carriers on temperature distribution and crystallization interface shape. The results indicate that the thermal radiation absorption inside the crystal directly affects the shape of the crystallization interface and the stability of crystal growth. Strong radiation absorption leads to a concave crystallization interface, whereas the concavity of the interface shape exhibits a decreasing trend with the increase of crystal height, which is beyond expectation. All these phenomena are related to the heat transfer and temperature distribution in the EFG furnace, and the 3D numerical modeling and simulation are helpful in deeply understanding the reasons behind the phenomena and improving the β-Ga2O3 crystal growth.

In the process of monocrystalline silicon growth by Czochralski method, water-cooled jacket has gradually become an important component widely used in the industry to increase the crystal pulling rate. Although increasing the emissivity of the crystal and the heat shield side of the water-cooled jacket can further increase the pulling rate, it will change the flow, heat transfer, melt-crystal interface deformation and oxygen transport, which affects the stable growth and the oxygen content of the silicon crystal. In this study, a global 2D numerical model was established to study the effect of the emissivity of the water-cooled jacket on flow, heat transfer, the deflection and oxygen content of melt-crystal interface. The results show that increasing the emissivity of the crystal side of the water-cooled jacket and reducing the emissivity of the heat shield side can achieve simultaneous control of the deflection and oxygen content of melt-crystal interface. This study provides a theoretical reference for optimizing the water-cooled jacket to achieve a stable low-oxygen growth of silicon crystal.

Oxygen is one of the critical impurities in silicon crystal. Excessive oxygen leads to increased defects and reduced efficiency of solar cells. Therefore, reducing oxygen concentration is a key issue in silicon crystal. In this study, a global 2D model for the continuous-feeding Czochralski silicon crystal growth was established, considering the complex shapes of the corner and bottom of inner crucible, instead of using the simplification of cylindrical partition. Based on the model, numerical simulations were performed to investigate the effects of inner crucible diameter, corner shape, and bottom shape on flow, heat transfer and oxygen transport. The results indicate that the structure of inner crucible significantly affects the oxygen transport. With the decrease of inner crucible diameter, the oxygen at crystallization interface increases; with the increase of curvature radius of inner crucible corner, the oxygen at crystallization interface decreases; with the increase of curvature radius of inner crucible bottom, the oxygen at crystallization interface increases. The findings can provide guidance for the optimization of inner crucible structure for reducing oxygen concentration.

Reducing cost remains a key technological focus in the process of photovoltaic Czochralski (CZ) silicon growth. To date, numerous studies have optimized various components of the CZ furnace, such as insulations, heat shield, and heaters, to achieve energy savings. However, research on the heater electrodes, a major high-heat-losing component, remains limited since the traditional 2D model fails to account for them. In this study, a 3D global model was developed considering the heater electrodes. Based on the model, the flow and heat transfer were simulated, and the results of reference case was verified by comparing with experimental data. Furthermore, the effect of heater electrode diameter on heat transfer paths and heat losses distribution were studied. The findings of this study provide a theoretical foundation for further reducing heat losses in the CZ silicon growth process.

We introduce Baichuan-Omni-1.5, an omni-modal model that not only has omni-modal understanding capabilities but also provides end-to-end audio generation capabilities. To achieve fluent and high-quality interaction across modalities without compromising the capabilities of any modality, we prioritized optimizing three key aspects. First, we establish a comprehensive data cleaning and synthesis pipeline for multimodal data, obtaining about 500B high-quality data (text, audio, and vision). Second, an audio-tokenizer (Baichuan-Audio-Tokenizer) has been designed to capture both semantic and acoustic information from audio, enabling seamless integration and enhanced compatibility with MLLM. Lastly, we designed a multi-stage training strategy that progressively integrates multimodal alignment and multitask fine-tuning, ensuring effective synergy across all modalities. Baichuan-Omni-1.5 leads contemporary models (including GPT4o-mini and MiniCPM-o 2.6) in terms of comprehensive omni-modal capabilities. Notably, it achieves results comparable to leading models such as Qwen2-VL-72B across various multimodal medical benchmarks.