A fundamental approach to design and decision making of integrated and in-situ catalytic adsorption-reaction processes

本提案旨在开发用于设计和决策集成和原位催化吸附-反应过程的基本工具。这些工具将是通用的，适用于催化反应分离系统的综合设计和决策。集成的原位催化吸附反应过程在生化、化学和药物反应中具有巨大的潜在重要性，特别是在这些反应由平衡驱动和/或被其他竞争反应抑制的情况下，以及随着新型催化吸附材料（如沸石）的发展。在这些系统中，集成和原位分离通过吸附产物(s) /流(s)完成所需反应的转化。因此，它们有潜力实现非常高的纯度、选择性和高价值、小批量产品的产量，这使它们特别适用于（生物）催化、酶促反应。用于设计这些系统的计算工具相当粗糙、简单，与支撑这些过程和实验工具的潜力无关。本研究面临的挑战是建立基于第一性原理的基础和强大的预测工具，以协助这些过程的设计和决策以及实验程序的分析和决策。提出的方法包括一个多尺度建模框架和一个过程设计框架。多尺度建模框架将创建与过程设计框架相关的各种参数的精确模型，这些参数来自催化剂-吸附剂和反应物-吸附物之间的基本分子水平相互作用。过程设计模型将以其基本随机介观模型的形式使用这些参数。过程设计框架将采用基于现象的过程设计上层结构的模块化表示，其中同质和异质现象将在单独的模块中开发。这种表述对于催化反应分离系统的设计和决策是通用的。采用随机和确定性方法的混合优化将被开发，以从过程设计上层结构中生成许多设计备选方案和决策。通过这项研究，一些新颖、有效和多学科的技术将发展起来，这些技术将在大范围的复杂系统中连接现象、模型和信息。在多尺度建模中，动力学蒙特卡罗（MC）模拟和小波变换加速MC模拟将分别用于微观和中观或粗粒度的参数表示。小波变换是一种广泛应用于图像分析的低损耗数据压缩工具，将首次用于动力学建模中的粗粒化。一旦这些基本和通用工具将在里程碑1中建立，模拟移动床吸附反应器（SMBR）的特殊设计框架将在里程碑2中提出。这将通过开发SMBR操作的准稳态模型并将这些模型纳入里程碑1的框架来实现。这种多学科的研究将使计算、材料科学、化学和化学工程等学科受益。它还将使从事高通量研究、实验、开发和催化剂、化学品、材料和制药产品建模的研究人员受益。

This proposal seeks to develop fundamental tools for design and decision making of integrated and in-situ catalytic adsorption-reaction processes. These tools will be generic for applying to integrated design and decision making of catalytic reaction-separation systems. The integrated and in-situ catalytic adsorption-reaction processes have tremendous potential importance in biochemical, chemical and pharmaceutical reactions, especially where these are driven by equilibrium and / or inhibited by other competing reactions, and following the advancements of novel catalytic-adsorptive materials (e.g. zeolites). In these systems integrated and in-situ separation through adsorption of product(s) / stream(s) completes conversion of desired reactions. Thus, they have potential to achieve very high purity, selectivity and yields of high value, low volume products, that make them particularly applicable to (bio)catalytic, enzymatic reactions. The computational tools for designing these systems are rather crude, simple and not relevant to underpin the potential of these processes and experimental tools. This research takes the challenge of establishing fundamental and robust predictive tools based on first principles to assist design and decision making of these processes and analysis and decision making of experimental procedures. The methodology proposed consists of a multi-scale modelling framework and a process design framework. The multi-scale modelling framework will create the precise models of the various parameters associated with the process design framework from fundamental molecular level interactions between catalyst-adsorbent and reactant-adsorbate specie. The process design models will use these parameters in the form of their fundamental stochastic mesoscopic models. The process design framework will adopt a phenomena-based modular representation of process design superstructure, where homogeneous and heterogeneous phenomena will be developed in separate modules. This representation is generic for design and decision making of catalytic reaction-separation systems. A hybrid optimisation employing stochastic and deterministic approaches will be developed to generate a number of design alternatives and decision making from process design superstructures. Several novel, effective and multi-disciplinary technologies will be evolved through this research that link phenomena, models and information across wide scales of complex systems. In multi-scale modelling, kinetic Monte Carlo (MC) simulations and wavelet transform accelerated MC simulations will be employed for microscopic and mesoscopic or coarse grained representations of parameters respectively. The wavelet transform, a loss less data compression tool widely used in image analysis, will be employed for the first time for coarse graining in kinetic modelling. Once these fundamental and generic tools will be established in milestone 1, a special design framework for simulated moving bed adsorptive reactors (SMBR) will be presented in milestone 2. This will be achieved through development of quasi steady state models of SMBR operations and inclusion of these models into the frameworks of milestone 1. This multi-disciplinary research will benefit disciplines like computation, material science, chemistry and chemical engineering. It will also benefit the researchers engaged in high throughput research, experimentation, development and modelling of catalysts, chemicals, materials and pharmaceutical products.