2020 官网升级中!现在您访问官网的浏览器设备分辨率宽度低于1280px请使用高分辨率宽度访问。
Source: Materials Science and Engineering
Metal additive manufacturing, also known as metal 3D printing, is a key technology with the potential to bring significant changes to the manufacturing industry. It can achieve rapid, accurate, and flexible metal component manufacturing through three-dimensional digital models, greatly improving manufacturing efficiency and profoundly affecting fields such as aerospace, automotive, energy, chemical, and pharmaceutical. The key physical phenomenon involved in this technology is rapid alloy solidification of metals. Unlike low-speed solidification related to traditional manufacturing technologies, the solid-liquid interface of metals is in extreme conditions far from equilibrium during rapid solidification, including extremely fast solid-liquid interface movement speed and significant temperature gradient. Under such extreme solidification conditions, non equilibrium effects such as solute trapping and solute drag can greatly affect the microstructure of metal materials after solidification, thereby affecting their mechanical properties. However, theoretical research on the formation mechanism of microstructure of metals under rapid solidification conditions is extremely lacking, which greatly restricts people’s ability to control the microstructure of materials, thereby limiting the further development of metal additive manufacturing. In such a situation, it is particularly important to develop a theoretical calculation model suitable for rapid solidification of metals.
Recently, the authoritative international journal of physics, Physical Review Letters, published a Phase field models proposed by researchers from Northeastern University and Colorado Institute of Mining, which can be used to predict the microstructure of alloy solidification far from equilibrium. This model quantifies the non equilibrium effects at the solid-liquid interface, including solute capture and solute resistance, under solidification conditions related to metal additive manufacturing. The author used this model for computational simulation and discovered a dynamic instability at the top of the dendritic structure driven by solute trapping when the interface velocity approaches the absolute stability limit. The simulation simultaneously restored the widely observed banded microstructure in the experiment, and revealed how this dynamic instability triggers the transition between dendritic structure and non microscopic aggregation structure. The predicted banded microstructure interval is consistent with the observed results in the solidification experiment of Al Cu thin films. This work was jointly completed by Kaihua Ji (lead author) of Northeastern University, Elaheh Dorari, Amy J. Clarke, an outstanding professor of Colorado School of Mining, and Alain Karma (corresponding author), an outstanding professor of Northeastern University.
Paper Link：https://journals.aps.org/prl/abs … sRevLett.130.026203
In this paper, the Phase field models is first used to quantitatively reduce the non-equilibrium effect in the solidification process, and a new idea is proposed to compensate for the non physical effect caused by the increase of the interface width by increasing the solute diffusivity within the interface range. The author has demonstrated that an increased solid-liquid interface width (five nanometers) can still quantitatively reduce non equilibrium effects at the physical interface scale (about one nanometers), which improves the computational efficiency of the model by three orders of magnitude, making it possible to simulate quantitative phase fields in two-dimensional or three-dimensional environments. After proving the convergence of the model, the author used it for numerical calculations and discovered a new dynamic instability at the top of the dendritic structure. Under low speed solidification conditions, the microstructure of metal materials is usually dendrites. When the solidification rate increases to near the absolute stability limit, the dendrites begin to oscillate and become unstable. The final mechanism that breaks the stability of dendrites is what the author calls a “boosting tip instability”. As the solidification rate continues to increase, this dynamic instability triggers the transition between dendritic solidification and microsegregation free solidification.
Figure 1 (a) Burgeoning tip instability when the interface velocity approaches the absolute stability limit. (b) The oscillation of interface velocity and solute concentration on the liquid phase side.
In rapid solidification experiments, the alternating light and dark banded microstructure parallel to the solid-liquid interface has been widely observed, but its formation mechanism is not very clear. The dark band has a dendritic structure, while the bright band has no microscopic aggregation structure. The author quantitatively reduced this phenomenon for the first time in computational simulation, and showed that the proportion of bright bands increased with the increase of solidification rate. This is consistent with experimental observations. The interface temperature and velocity in the computational simulation exhibit periodic oscillations, leading to the formation of banded microstructures.
Figure 2 (a) – (d) Banded microstructure at different interface velocities. (e) Periodic oscillation of interface velocity and temperature.
The author further takes into account the latent heat diffusion during the solidification process and elucidates its importance for accurately predicting the spacing of banded microstructures. In the phase field simulation considering the latent heat diffusion effect, the predicted banded microstructure spacing is consistent with the observed results in the rapidly solidified Al Cu thin film experiment. The results further prove the accuracy of the Phase field models.
Figure 3 (a) Banded microstructure observed in the solidification experiment of Al Cu thin film. (b) Phase field simulation considering latent heat diffusion effects.
In addition, the author proposes a fully variational Phase field models. Unlike the traditional non variational Phase field models used to simulate metal solidification, the fully variational phase field model can be more easily extended to complex alloy systems, such as multiphase and multi-element alloy systems. Therefore, the Phase field models has extremely broad application prospects.
