The accumulation of shear transformations (STs) in space and time is responsible for plastic deformation in amorphous solids. Here we study the effect of finite strain rates on STs during simulations of athermal shear deformation in an atomistic model of amorphous silicon. We present a time-resolved analysis of STs by mapping the plastic events identified in the atomistic simulations on a collection of Eshelby inclusions, which are characterized in terms of number, effective volume, lifetime, and orientation. Our analysis led us to distinguish between small and large events. We find that the main effect of a lower strain rate is to allow for a larger number of small events, roughly identified by an effective volume ${\gamma }_0{V}_0<20$ Å${}^3$, while the number and characteristics of larger events are surprisingly independent of the strain rate. We show that at low strains, the decrease of the stress observed at lower strain rates is mainly due to the excess of small events, while at larger strains, when the glass approaches the yield point where a shear band forms, larger events start to play a role and organize due to their elastic interactions. This phenomenology is compared with the predictions of mesoscale elastoplastic models. The technique developed here can be used as a systematic tool to analyze plasticity during molecular dynamics simulations. It can also give valuable information to develop physically grounded mesoscale models of plasticity, providing quantitative predictions of the mechanical properties of amorphous materials.

中文翻译：

---

剪切态非晶硅瞬态塑性状态下的时间分辨剪切转变

剪切变形（ST）在空间和时间上的累积是非晶态固体塑性变形的原因。在这里，我们研究了在非晶硅原子模型中模拟无热剪切变形期间有限应变速率对ST的影响。我们通过将在原子模拟中确定的塑性事件映射到一系列Eshelby夹杂物上，对ST进行时间分辨分析，该夹杂物以数量，有效体积，寿命和方向表示。我们的分析使我们能够区分大事件和小事件。我们发现较低的应变率的主要作用是允许大量的小事件，大致由有效体积确定${\gamma }_0{V}_0<20$ 一种${}^3$，而较大事件的数量和特征出乎意料地与应变率无关。我们表明，在低应变下，在较低应变速率下观察到的应力减小主要是由于过多的小事件引起的，而在较大应变下，当玻璃接近剪切带形成的屈服点时，较大的事件开始发生由于它们之间的弹性相互作用而发挥作用和组织。将该现象学与中尺度弹塑性模型的预测进行了比较。此处开发的技术可以用作分析分子动力学模拟过程中可塑性的系统工具。它还可以提供有价值的信息，以开发基于物理的中尺度可塑性模型，从而提供对非晶态材料机械性能的定量预测。