A micromechanical model of metal powder compaction is developed as a generalization of the previous model proposed by Mayer et al. (2020). The model is applicable for an arbitrary axisymmetric deformed state including the uniaxial compression and the uniform tri-axial one. It describes the stage of compression of initially spherical particles, the percolation transition to the system of isolated pores and the stage of the collapse of pores. At the first stage, the metal particles are approximated by spheres with indented sectors with two independent indentation depths: One along the dedicated axis and another one along the perpendicular directions. The model of porous medium is formulated in finite deformations and the dislocation-driven plasticity model (Krasnikov and Mayer, 2015) is used to describe the pore collapse at the last stage. MD simulations of uniform tri-axial and uniaxial compression of iron, copper and magnesium nanoparticles of 6, 12 and 18 nanometers in diameter are performed for a number of temperatures. Activity of dislocations, which form dense structures, is the main driver of the plastic deformation of nanoparticles during compaction, while the presence of twins is revealed for copper and magnesium. MD data is used for identification of the model parameters and verification of the model. Bayesian algorithm is implemented for parameter identification, which reveals itself as an efficient tool and allows us to fit the model with the same set of parameters to MD data for all investigated temperatures, particle diameters and modes of compression. The developed and parameterized micromechanical model is embedded into the continuum model of shock wave propagation in metal powder as a constitutive model of the substance. Both partial and complete compaction by the shock wave is investigated by an example of iron. The presence of elastic precursor is revealed for the case of nanopowders because of high yield strength of nanoparticle material achieved due to the strain hardening during the preliminary relaxation under the action of surface tension. At the same time, the precursor is negligible for submicro – and microparticles. Velocities of both the precursor and the plastic shock wave are considerably lower than the sound speed of the bulk metal. This is because the average elastic modulus of powder during compaction is much smaller than that for a compact material.

金属粉末压实的微机械模型是作为 Mayer 等人提出的先前模型的推广而开发的。(2020)。该模型适用于任意轴对称变形状态，包括单轴压缩和均匀三轴变形。它描述了初始球形颗粒的压缩阶段、渗透过渡到孤立孔隙系统和孔隙坍塌阶段。在第一阶段，金属颗粒被球体近似，球体具有两个独立的压痕深度：一个沿专用轴，另一个沿垂直方向。多孔介质模型在有限变形中建立，位错驱动塑性模型（Krasnikov and Mayer，2015）用于描述最后阶段的孔隙坍塌。在多个温度下对直径为 6、12 和 18 纳米的铁、铜和镁纳米粒子进行均匀三轴和单轴压缩的 MD 模拟。形成致密结构的位错活动是纳米颗粒在压实过程中塑性变形的主要驱动力，而铜和镁则揭示了孪晶的存在。MD数据用于模型参数的识别和模型的验证。贝叶斯算法用于参数识别，这表明它本身是一种有效的工具，并允许我们将具有相同参数集的模型拟合到所有研究温度、颗粒直径和压缩模式的 MD 数据。开发和参数化的微机械模型作为物质的本构模型嵌入到金属粉末中冲击波传播的连续模型中。以铁为例研究了冲击波的部分和完全压实。对于纳米粉末，由于在表面张力作用下的初步松弛过程中应变硬化实现了纳米颗粒材料的高屈服强度，因此揭示了弹性前体的存在。同时，对于亚微粒子和微粒子来说，前体可以忽略不计。前体和塑性冲击波的速度都远低于大块金属的声速。这是因为粉末在压制过程中的平均弹性模量远小于压制材料的平均弹性模量。