磁性纳米粒子粒径表征方法研究

发布时间：2018-02-24 16:04

本文关键词： 粒径测量磁纳米粒子磁化曲线曲线拟合 L-M算法 GlobalSearch算法　出处：《华中科技大学》2015年硕士论文　论文类型：学位论文

【摘要】：本文利用磁性纳米粒子的磁化曲线进行粒径分布测量研究。磁性纳米粒子的磁化曲线满足郎之万方程,包含着粒子的粒径分布信息。搭建用于获得磁性纳米粒子磁化曲线的三角波磁场装置,利用优化算法对磁化曲线拟合,得到粒径分布参数。磁性纳米粒子粒径分布的研究对其应用于浓度成像或者温度成像等生物医学领域具有重要的意义。首先,假设磁性纳米粒子的粒径服从对数正态分布,对磁化曲线进行离散化处理,得到粒径测量模型。磁流体中磁性纳米粒子的单体和聚集体处于动态平衡状态。在外磁场下,磁性纳米粒子的磁化强度由单体和聚集体共同构成。由此得到聚集体存在情况下粒子粒径测量模型。利用优化算法对磁化曲线进行拟合,得到磁性纳米粒子的粒径分布。其次,利用三角波磁场装置准确地获取磁性纳米粒子的磁化曲线。三角波磁场装置包括激励磁场产生、磁场测量和数据处理三个功能模块。其中,激励磁场产生模块通过激励波形预编辑和PID电流调节技术产生三角波激励磁场;磁场测量模块通过检测采样电阻上的电压信号实现激励磁场的测量,通过两个反向串联的探测线圈实现磁性纳米粒子磁化强度的测量;数据处理模块包括背景磁场的抑制、数据平均处理、磁化强度波形的重构、磁化曲线的获取几个方面。最后,利用优化算法对磁化曲线进行拟合,得到粒子粒径分布参数。最常用的非线性曲线拟合算法是L-M算法。L-M算法是一种局部优化算法,不能获得精确的粒子粒径分布。对优化算法进行改进,采用全局优化算法GlobalSearch。GlobalSearch算法的测量结果非常理想,即使在初始点远离最优解和样品浓度未知的情况下。利用SOR-20样品和SOR-30样品进行实验验证。
[Abstract]:In this paper, the magnetization curve of magnetic nanoparticles is used to measure the particle size distribution, and the magnetization curve of magnetic nanoparticles satisfies the Langzhiwan equation. The triangular wave magnetic field device used to obtain magnetization curve of magnetic nanoparticles was built, and the magnetization curve was fitted by optimization algorithm. The study of particle size distribution of magnetic nanoparticles is of great significance for their application in biomedical fields such as concentration imaging or temperature imaging. First, assuming that the particle size of magnetic nanoparticles is in logarithmic normal distribution, The magnetization curve is discretized and the particle size measurement model is obtained. The monomer and aggregate of magnetic nanoparticles in magnetic fluid are in dynamic equilibrium state. The magnetization of magnetic nanoparticles is composed of monomers and aggregates. The particle size measurement model in the presence of aggregates is obtained. The magnetization curve is fitted by an optimization algorithm, and the particle size distribution of magnetic nanoparticles is obtained. The magnetization curve of magnetic nanoparticles is accurately obtained by using the triangular wave magnetic field device. The triangular wave magnetic field device includes three functional modules: excitation magnetic field generation, magnetic field measurement and data processing. The excitation magnetic field generation module generates the triangular wave excitation magnetic field by the pre-editing of the excitation waveform and the PID current regulation technology, and the magnetic field measurement module realizes the measurement of the excitation magnetic field by detecting the voltage signal on the sample resistor. The magnetization of magnetic nanoparticles is measured by two reverse series detection coils. The data processing module includes the suppression of background magnetic field, the average processing of data, the reconstruction of magnetization waveform and the acquisition of magnetization curve. The parameters of particle size distribution are obtained by fitting the magnetization curve with optimization algorithm. The most commonly used nonlinear curve fitting algorithm is L-M algorithm. L-M algorithm is a local optimization algorithm. The particle size distribution can not be obtained accurately. The optimization algorithm is improved, and the global optimization algorithm GlobalSearch.GlobalSearch algorithm is used to measure the results very well. Even if the initial point is far from the optimal solution and the concentration of the sample is unknown, the SOR-20 sample and the SOR-30 sample are used for experimental verification.
【学位授予单位】：华中科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TM27;TB383.1

【参考文献】