Self-optimising reactor systems combined with computational strategies to understand and control inorganic particles.

目的：该项目的目的是使用自优化流动系统与计算策略相结合，以便更好地了解硫酸铁的晶体生长，并提供与间歇式反应器相比更好的控制。小型连续流系统将用于工艺条件筛选。机器学习将用于利用高通量筛选数据来优化流程。机械模型将用于获得预测工具。方法：使用XRD， DLS， SAXS， Raman和G3形状和尺寸分析仪等技术对七水硫酸铁进行表征。这些数据将作为今后分析的参考。首先研究纯七水硫酸铁体系。七水硫酸铁在50℃时溶于水，冷却后结晶。通过向系统中加入不同浓度的硫酸来重复结晶过程。通过在FeSO4-H2SO4-H2O体系中引入镁、锰和铜，研究杂质对硫酸铁晶体形成的影响。与数据驱动算法配对的连续流动反应器将用于硫酸铁的结晶和反应条件的优化。这将与在线表征技术相结合，如在线XRD， DLS， SAXS， Raman和UV-Vis。获得的数据将用于使用人口平衡方程与CFD配对创建机制模型。这些模型将被用作大型反应堆系统的预测工具和相图的创建。潜在影响：该项目将使用混合实验建模方法来理解和控制硫酸铁结晶，这将更好地理解不同工艺参数（如温度、pH值或溶剂）对关键颗粒特性（如尺寸、形状、多晶态、晶体结构和组成）的影响。更好地控制结晶也将有助于过滤和干燥过程形成具有所需性能的产品。此外，自优化连续结晶器将允许实现所需产品的高通量。更好地了解晶体生长的能力将有利于精细化学工业。该项目将支持多个阶段的生产，从在特定反应条件下生成颗粒特性的相图到基于模型的大规模工艺开发。预期成果：更好地了解七水硫酸铁的晶体生长。一种新的机制模型，可以作为大规模系统的预测工具。一种新型连续流反应器设计。

Aims: The aim of this project is to use self-optimising flow systems combined with computational strategies to allow better understanding of crystal growth of iron sulphate and to provide better control in comparison to batch reactors. Small scale continuous flow systems will be used to process condition screening. Machine learning will be used to utilise high-throughput screening data to optimise processes. Mechanistic models will be used to obtain predictive tools. Methodology:Iron sulphate heptahydrate will be characterised using techniques such as XRD, DLS, SAXS, Raman and G3 shape and size analyser. This data will be used as a reference in future analysis. The pure iron sulphate heptahydrate system will be investigated first. Iron sulphate heptahydrate will be dissolved in water at 50c and crystallised by cooling. The crystallisation will be repeated by introducing various concentrations of sulphuric acid into the system. The effect of impurities on iron sulphate crystal formation will then be investigated by introducing magnesium, manganese and copper into the FeSO4-H2SO4-H2O system. Continuous flow reactors paired with data driven algorithms will be used for the crystallisation of iron sulphate and optimisation of the reaction conditions. This will be paired with online characterisation techniques such as online XRD, DLS, SAXS, Raman and UV-Vis. The data obtained will be used to create mechanistic models using population balance equations paired with CFD. These models will be used as a predictive tool for large scale reactor systems and for the creation of phase maps. Potential Impact:This project will use a hybrid experimental-modelling approach to understand and control iron sulphate crystallisation, which will provide a better understanding of the influence of different process parameters such as temperature, pH or solvents on crucial particle properties such as size, shape, polymorphs, crystal structure and composition. Better control in crystallisation would also allow products with desired properties to be formed aiding both filtering and drying processes. Additionally, self-optimising continuous crystallisers will allow high throughput of the desired product to be achieved. The ability to understand the crystal growth better would be beneficial for fine chemical industries. This project will support manufacturing at multiple stages, ranging from the creation of phase maps of the resulting particle properties at specified reaction conditions to model based large-scale process development. Expected Deliverables: A better understanding of crystal growth of iron sulphate heptahydrate. A new mechanistic model, which would act as a predictive tool for large-scale systems. A new continuous flow reactor design.