Recent progress on the simulation technology of magnetic fluid
Recent progress on the simulation technology of magnetic fluid
复制标题
DOI：
10.1360/n972018-01068
复制
发表时间：
2019-04
影响因子：
--
通讯作者：
Lei Pei;X. Gong;Shouhu Xuan
中科院分区：
文献类型：
--
作者：
Lei Pei;X. Gong;Shouhu Xuan
磁流体是由磁性颗粒、非磁性基体和添加剂组成的新型磁流变（MR）智能材料。施加外部磁场后，磁性颗粒将由于磁偶极力而彼此相互作用。磁流体的粘度和屈服应力可以在毫秒内增加几个数量级，这被称为磁流变效应。磁性流体的可控和可逆特性使其广泛应用于药物靶向输送、磁热疗法、商用阻尼器和抛光等领域。为了理解磁性流体的力学行为，研究人员开发了几种针对不同流动条件的理论模型。然而，由于计算量大，理论模型仅适用于一些特殊问题，例如二维和轴对称问题。近年来，随着计算机性能的发展，模拟已成为研究磁流体磁流变机理的重要方法。本文综述了磁流体理论与模拟的最新进展。首先介绍了剪切模式、挤压模式和阀模式下磁流体的理论模型。其次，对现有的磁流体模拟方法，如分子动力学、颗粒级动力学模拟、有限元法等进行了阐述。讨论了这些方法的有效性及其优缺点。然后从新型磁性流体力学性能模拟、常规磁性流体复杂力学行为模拟和生物磁性流体模拟3个方面总结了磁性流体模拟的研究进展。最后，提出了磁流体模拟的一些未来趋势。今后工作应重点关注以下3个主题。首先，需要考虑各种微观相互作用的综合理论模型。为了制备磁流变效应高、分散性好、密度低的磁性颗粒，实验中通常采用表面包覆、改性和添加剂等方法。不幸的是，在模拟中很少考虑非磁性元件对磁流变效应的影响。其次，目前的模拟无法同时获得复杂流动中磁流体的宏观力学性能和微观结构。有限元方法和计算流体力学适用于复杂的宏观问题，但不能同时获得微观结构。介观模拟方法无法表现出颗粒的大规模聚集，从而导致与实验相比存在偏差。建立多尺度模拟方法、提高模拟精度已成为迫切要求。结合不同空间尺度的模拟方法，利用机器学习降低时间成本已成为一种可能的方法。第三，建立多物理场耦合仿真方法。近年来，磁性流体的磁场控制电特性引起了研究人员的兴趣。磁性流体具有这种新颖的可控特性，可广泛应用于电池和传感器中。利用机械、电学和磁学模型共同研究磁性流体的其他物理性质将是未来的趋势。这些成果都将有助于高性能磁流体的发展，进一步扩大磁流体的应用范围。
Magnetic fluid is a novel magnetorheological (MR) intelligent material consisting of magnetic particles, non-magnetic matrix, and additive agents. After applying the external magnetic field, magnetic particles will interact with each other due to the magnetic dipolar forces. The viscosity and yield stress of magnetic fluid could increase several orders of magnitude in milliseconds, which is called the MR effect. The controllable and reversible property makes magnetic fluid widely applied in drug targeting delivery, magnetic thermal therapy, commercial dampers, and polishing etc. In order to comprehend the mechanical behaviors of magnetic fluid, researchers developed several theoretical models for different flow conditions. However, due to the extensive calculation, theoretical models are only applicable for some special problems, such as 2-dimensional and axial symmetry. In recent years, with the development of computer performance, simulation has become an important method to investigate the MR mechanism of magnetic fluid. This paper reviews the recent progress in the theory and simulation of magnetic fluid. Firstly, theoretical models of magnetic fluid under shear mode, squeeze mode, and valve mode are introduced. Secondly, the existing simulation methods for magnetic fluid, such as molecular dynamics, particle-level dynamic simulation, and finite element method, are illustrated. The validity of the methods and their merits and drawbacks are discussed. Then, the research progress of the simulation of magnetic fluid is summarized from 3 aspects: Simulations of mechanical properties of novel magnetic fluid, simulations of complex mechanical behaviors of conventional magnetic fluid, and simulations of biomagnetic fluid. Finally, some future trends of simulation of magnetic fluid are proposed. The following 3 topics should be emphasized in the future work. First, a comprehensive theoretical model considering a variety of microscopic interactions is required. In order to prepare magnetic particles with high MR effect, excellent dispersibility, and low density, surface coating, modification, and additive agents are usually applied in experiments. Unfortunately, the influence of non-magnetic components on the MR effect is seldom considered in simulations. Second, current simulations could not simultaneously obtain the macroscopic mechanical properties and microstructures of magnetic fluid in complex flow. Finite element method and computational fluid dynamics are applicable for complex macroscopic problems but can not obtain the microstructures at the same time. Mesoscopic simulation methods can not exhibit large-scale aggregations of particles, which leads to the deviation compared with experiments. To establish multi-scale simulation methods and improve the accuracy of simulations have become an urgent requirement. Combining simulation methods with different spatial scales and reducing the time cost by using machine learning have become a possible approach. Third, multi-physics coupling simulation methods should be established. The magnetic field controlled electrical properties of magnetic fluid have attracted researchers interest in recent years. Magnetic fluid with this novel controllable property could be widely applied in battery and sensors. Investigations on other physical properties of magnetic fluid by using mechanical, electrical, and magnetic model together will be a future trend. These achievements will all contribute to the development of high-performance magnetic fluids and further enlarge the range of applications of magnetic fluid.