(169b) Linking CFD Simulation to Continuum Models Using Correlation Equations for Mass and Heat Transfer in Chemical Engineering

The reaction mechanism and the surface properties are still not fully understood, despite the technological advances. Therefore, the exact process of the reaction and the interaction of the catalytic surface have been a focus of academic research in recent years. A multi-scale hierarchical approach has been successfully applied to a complex industrial process, where a heterogeneous catalytic system is modelled from atomistic to macro level. However, this approach is unlikely to become a mainstream method for solving industrial processes, because it requires a lot of computational resources.

Computational fluid dynamics (CFD) is a powerful tool that can be used to tackle problems in chemical engineering, especially those involving fluid flow, heat and mass transfer, and chemical reactions. CFD can provide detailed information about the local phenomena that occur in complex systems, such as temperature, pressure, velocity, concentration, and reaction rate. CFD can also help to design and optimize chemical processes and equipment, such as reactors, separators, mixers, and heat exchangers. One of the challenges of using CFD in chemical engineering is to link the results of CFD simulation to the continuum models that are commonly used in engineering practice. These models assume of homogeneous properties and behaviour of the fluid, and they use correlation equations (CE) to describe the mass and heat transfer coefficients. CE are empirical formulas that relate dimensionless numbers and coefficients to predict the mass and heat transfer rates in a system. CE are useful because they simplify the complex three-dimensional phenomena to one dimension, and they can be used in much easier models, such as plug flow reactor (PFR) or continuous stirred tank reactor (CSTR).

However, CE have some limitations and drawbacks. First, they have to be determined experimentally for each specific system, which can be very expensive and time-consuming, especially for industrial scale reactors. Second, they are only valid for a certain range of the operating conditions of the system, such as Reynolds number, Prandtl number, Schmidt number, etc. Third, they do not capture the local variations and fluctuations of the properties and behaviour of the fluid, which can be important for some systems.

Therefore, a novel technique has been proposed to link CFD simulation to continuum models using CE. The idea is to use CFD simulation to calculate the local phenomena that are needed to determine CE, such as Nusselt number, Sherwood number, Stanton number, etc. Then, these values are averaged over the cross-section or volume of the system to obtain the effective CE that can be used in simple continuum models. This way, CFD simulation can provide more accurate and reliable CE that can account for the effects of geometry, flow pattern, turbulence, reaction kinetics, etc. Moreover, this technique can reduce the need for experimental data and extend the validity range of CE. With this technique, CFD simulation can be linked to pseudo-homogeneous and heterogeneous continuum models that can describe the mass and heat transfer rates in a system. This can enable fast and targeted optimization of industrial processes by changing the operating conditions or design parameters of the system. Some examples of industrial processes that could benefit from this technique are gas-liquid reactors, spray dryers, nanofluid heat exchangers, etc.