Neutron imaging to inform multi-scale simulations of catalytic reactors

Large-scale simulations of chemical reactions and processes are invaluable for optimising product yields, lowering energetic and financial costs, and ensuring the safe operation of a catalytic process. As such simulations, such as Computational Fluid Dynamics (CFD), are vital for developing more sustainable chemical processes. In this proposal we will use neutron imaging to give unique experimental insights into the diffusion of propane within a packed bed of a SAPO-34 catalyst to create unique multiscale reactor models.
Optimising catalytic industrial processes, and accelerating scale up of emerging technologies, relies heavily on in silico predictions and simulations, the precision of which depends on the accuracy of computational models. In particular, computational fluid dynamics (CFD) predicts variations in local flows and temperatures, within a catalytic fixed bed reactor, over a wide length scale (mm to m), while simultaneously kinetic models predict product and reagent concentrations. Systems involving impenetrable particles, considering surface reactions only, are facile to model. Porous catalysts require an extra level of complexity, as both bulk mass-transfer and pore-diffusion must be considered. Very few techniques can give this level of detail, making it an ideal method for validating, testing, and training CFD simulations for greater accuracy.

化学反应与过程的大规模模拟对于优化产品收率、降低能量与财务成本、保障催化过程安全运行具有不可替代的价值。诸如计算流体力学（Computational Fluid Dynamics, CFD）这类模拟技术，对开发更可持续的化工过程至关重要。在本研究方案中，我们将采用中子成像技术，获取丙烷在SAPO-34催化剂填充床内扩散过程的独特实验观测数据，以此构建独具特色的多尺度反应器模型。
优化催化工业过程、加速新兴技术的规模化放大，高度依赖计算机模拟（in silico）预测与仿真，其精度直接取决于计算模型的准确性。具体而言，计算流体力学（CFD）可在宽长度尺度（毫米至米级）内预测催化固定床反应器内的局部流场与温度分布，而动力学模型则可同时预测产物与试剂的浓度分布。仅考虑表面反应的不可渗透颗粒体系建模相对简便，但多孔催化剂的建模复杂度更高，因为必须同时兼顾主体传质与孔道扩散两个过程。目前几乎没有其他技术能够提供如此细致的观测细节，因此中子成像是验证、测试并提升CFD模拟精度的理想方法。