【作者】 伍川；

【导师】 时钧； 黄培；

【作者基本信息】 南京工业大学 ， 化学工程， 2002， 博士

【摘要】 溶液结晶是由液相制备与分离固体产品的重要手段。多数传统工业与结晶过程都有着十分密切的关系。近年来，新材料、信息电子、生物化工、新型能源材料及环境科学等高新技术产品的制备、分离和提纯也越来越有赖于结晶技术的支撑。尤其是20世纪80年代的超分子化学的理论和实践证明，固体物质的性质不仅取决其化学组成，还取决于其超分子结构，即表现出所谓的“构效关系”，化学组成相同的固体，超分子结构改变后将呈现迥异的性质或功能。结晶过程已成为国际工业界和学术界近期关注的焦点之一。 结晶是一个多相流的复杂过程，受制于平衡热力学和过程动力学。热力学理论和方法已足以获得准确的相平衡关系；然而即使对于简单的二元物系的结晶过程，晶核形成和晶体生长动力学的研究仍面临许多困难，通常采用经验模型表述，而动力学参数的准确性和可靠性是结晶器放大设计、过程控制与优化的关键，因此多相流信息的准确表征、结晶机理的进一步揭示及建立起与之相适应的数学模型有着十分重要的学术研究和实际应用价值。 本文对溶液结晶动力学进展及其现状作了较全面的综述。基于粒数衡算模型，以KNO₃-H₂O为模型研究体系，根据结晶过程中溶质质量守恒和McCabe定律，首先建立了晶体线性生长动力学模型，并确定出动力学参数估计所需要表征的固液两相物理参数，进而建立了实验表征方法和测定装置。 根据二元电解质溶液体系的浓度与电导率和温度之间的关系，建立一电导率在线测量装置，可在一定范围内测定溶液的浓度或过饱和度。实验得到与Hlozny等人基本一致的溶液电导率变化趋势。但由于悬浮晶粒及其在强流体场作用下对测试电极产生的冲刷破坏作用，致使电极常数改变，影响测定结果的准确性和重现性，难于实现浓度的在线测定。为此，本文基于阿基米德定律，采用密度法改进了实验装置，实现了对结晶过程中溶液浓度或过饱和度的连续在线测量，并提高了测定精度，所测得实验数据具有较好的重现性。为解决动力学研究中所涉及的固相信息实时测定问题，根据模型需在线检测结晶过程中溶液透光率，本文以光敏电阻作为光电检测元件，构建了一双光路检测系统，有效地提高了测定系统的稳定性。通过调节激光光源负载电压改变光强度及调 摘要 整检测池的宽度，消除了多重散射对透光率的影响，将透光率的变化范围控制 在满足Beer1a毗ert定律的范围内工将温度、溶液密度和透光率信号经模数转 换，并用 Microsoft Visual Basic语言编写的数据采集软件，实现了实验过程 中各被测量的自动连续采集。上述装置的建立和改进为结晶动力学模型参数的 求取或估计奠定了良好的基础。 实验测定了洲。－Hzo体系自发成核结晶过程中温度、透光率和浓度与时问 的关系，由所建晶体线性生长动力学模型计算出整个区间的生长速率。因自发 成核结晶过程中伴有明显的晶核生成，不符生长经验模型要求的粒数恒定的前 提，无法直接进行参数估计。本文利用生长经验模型表达式的对数生长速率与 对数相对过饱和度之间存在的线性关系，将自发成核结晶过程大致分为操作前 ，期的成核和后期的生长两个阶段，进一步作晶体生长参数估计，所得结果与文 献值相吻合。／鉴于自发成核结晶过程特征，本文进一步分析认为：可将经历初级成核阶 段后晶核的形成与二次成核近似相当，即自发成核结晶过程由初级成核、二次 成核和线性生长三个主要阶段组成。据此建立了伴有成核和晶体生长的动力学 模型，井采用分段非线性回归的方法，由实时浓度、透光率和相对过饱和度对 二次成核和晶体生长动力学进行了参数估计，所得成核和生长参数均与文献结 果较好地吻合；参数验证的结果表明：由实时浓度和透光率的变化并利用分段 回归的方法，可以就自发成核的结晶过程定性区分出二次成核和晶体生长两个 阶段。 本文进一步针对添加晶种的规。－H。0体系溶液间歇结晶过程，在所建晶体 线性生长速率模型基础上，得到二次成核及生长的浓度预测模型，进而由溶液 模拟浓度与实测浓度之差AC考察添加晶种负荷对抑制成核作用的影响。实验 与计算结果表明：增加晶种添加量，AC随之减小，即成核过程越不显著；当 AC为0时，体系中成核作用被完全抑制，此时溶液浓度的降低完全是由于溶 质在晶种表面的生长。同时，晶种的加入对体系过饱和度同样具有抑制作用， 过饱和度变化范围随着晶种负荷的增加而减小。 在自发成核和添加晶种结晶过程实验研究和相关模型分析的基础上，本文 基于结晶过程中晶核数目和晶体粒度的变化对晶体粒度分布(“①二阶和三阶矩 量影响程度的不同，提出并定义了一无因次判据K＼ 由文献报道的添加晶种 二互更多还原

【Abstract】 Crystallization from solution is an important unit operation process in the field of chemical engineering, which could be used to separate and purify the higher purity solid products with lower costs, and has been applied in most traditional industries. Recently, crystallization operations are being employed to obtain high value-added products such as new materials, biochemical products and novel energy sources. It has been demonstrated by the theory of super molecule that the properties and/or functions of solid product are not only depended on its chemical composition, but also on its molecule structure, that is called as "constitutive relationship". Currently, the crystallization processes is being one of the most interesting research focuses.Crystallization is a sophisticated process with multiphase flows, which is related to its equilibrium thermodynamics and kinetics. With the thermodynamic theories and techniques, it is relatively easy to determine the phase equilibrium data with enough accuracy, however there still exist much more difficulties in crystallization kinetics study even for a simple binary system. That is the reason that nucleation and crystal growth rate are generally represented in form of the empirical expressions. The crystallization kinetic is important for crystallizer design, process control and optimization, and it is strongly depended upon the accurate characterization of process information concerning with multiphase flows and the further disclose of its mechanisms with suitable mathematical models.The progress and current status of the kinetic study on crystallization from solution was briefly reviewed in this thesis. Considering the mass balance law for solute, and combining with the populance balance model and the McCabe’s L law, a linear crystal growth rate model was presented for potassium nitrate aqueous system. The test variables of both the liquid and solid for kinetic parameter estimations were given on this model, and the experimental apparatus were also set up.In order to determine the concentration or supersaturation in crystallization process, an online test apparatus was set up based on the relationship between the concentration and the electroconductivity at a temperature for binary electrolyte systems, which was also used in similar fields to measure the electrolyte concentration. The results showed that the accuracy and the repeatability for online concentrationacquisition could not be assured with this approach even though the obtained electroconductivity curves were essentially consistent with those of Hlozny’s, because the electrode constant will be changed in the crystallization operation. Substituted by a densimeter based on the Archimedes Laws, the online concentration measurement was further modified, and the quality of the measured data was improved with satisfactory accuracy and repeatability. Nevertheless, information obtained from the continuous phase (concentration or supersaturation measurements) was still not enough to completely define the system. In order to significantly reduce the limits of uncertainty in kinetic parameter estimations, it was also necessary to use information on the solid phase. Therefore, in addition to the measurement of dissolved solid concentrations obtained through online densimeter, transmittance measurements should be performed to provide online data about the second moment of CSD (Crystal Size Distribution). In this thesis, a light sensitive resistor served as the light/electricity conversion component was used to measure the transmittance. In order to efficiently improve the stability of the apparatus, a dual-optical detection circuit was adopted. To eliminate the multiple scattering effects on transmittance measurements, the range of the transmittance was limited to 70-100% by adjusting the width of the light flow cell and changing the intensity of the incident light through regulating the load voltage of the laser light source. Moreover, the data acquisition software was programmed in Microsoft Visual Bas更多还原