scispace - formally typeset
Search or ask a question

Showing papers by "Sankaran Sundaresan published in 2000"


Journal ArticleDOI
TL;DR: In this paper, a 3D simulation of two-phase flow using, say, a million nodes, involves the solution of over nine million nonlinear equations at each time step; the size of the problem becomes much larger when energy and species bal- ances are included.
Abstract: ew things are more central to chemical engineering than multiphase flow chemical reactors; they are used in industry to produce a variety of chemicals, where economy of scale remains the driving factor. The engineering issues are classical, and the modern trend towards miniaturization will have little impact on these systems. Multiphase reactors for classic applications such as fluid cat- alytic cracking are still evolving. New processes such as slurry bubble column reactors for gas conversion are under development. Reliable multiphase reactor models that can be used with confi- dence for improving existing processes and scale-up of new Coarse-grid simulation of macro-scale flow structures Macro-scale coherent structures in multiphase flow, particularly those involving vortical motion, can be captured only through three-dimensional (3-D) transient simulations. A 3-D simulation of two-phase flow using, say, a million nodes, involves the solution of over nine million nonlinear equations at each time step; the size of the problem becomes much larger when energy and species bal- ances are included. Simulations of such magnitude are not com- mon in process design today, but will become routine in the not- too-distant future. As we shall see, even with this many nodes, the grid structure in simulations of commercial-scale reactors remains coarse; so, the model equations used to simulate commercial reac- tor performance must represent averages at this coarse-grid scale. The drag and stresses influencing the hydrodynamics and effective dispersion coefficients and reaction rates appearing in the energy and species balance equations are affected significantly by the sub- grid scale fluctuations and are grid-size dependent. Yet, this depen- dence is just beginning to be appreciated. Experiments, both labo- ratory and computational, and theory that will lead to accurate scale-dependent closures for these quantities represent important frontiers in this class of problems. The coarse-grid simulation of multiphase flow is conceptually similar to large eddy simulation of single-phase turbulent flow, where one accounts for the effects of unresolved eddies through sub-grid models (Fox, 1996). In single-phase turbulent flow, the flow of energy associated with fluctuations is predominantly from large scale to smaller ones. In the class of multiphase flows consid- ered here, meso-scale structures arise as a result of local instabilities and grow into larger and larger scales, so macro-scale shear is not a requirement for creating and sustaining a chaotic state of flow (unlike in single-phase flow). Because of this difference, we cannot simply adopt the ideas developed in single-phase flow. In this arti- cle, we examine what is known generically about flow structures at different scales in gas-solid, gas-liquid, and gas-liquid-solid sys- tems, and point out where challenges lie in building the hydrody- namic components of the next generation of reactor models.

220 citations


Book ChapterDOI
TL;DR: In this paper, the authors investigated the effect of surface acidity on the performance of VPO and supported vanadia catalysts, and found that the enhancement of acidity had a beneficial effect on both the rate of n-butane oxidation and selectivity to maleic anhydride.
Abstract: The present study addresses the nature of the promoter effect in the bulk VPO and supported vanadia catalysts. No correlation was found between the electronegativity of the promoter or oxide support cation and the catalytic properties of these two catalytic systems. The enhancement of surface acidity had a beneficial effect on both the rate of n-butane oxidation and selectivity to maleic anhydride over the bulk VPO and supported vanadia catalysts. These findings suggested that the activation of n-butane on both the bulk and supported vanadia catalysts may require both a redox and an acid site. The results of the present study further demonstrate that the supported vanadia catalysts represent a suitable model for bulk VPO catalysts.

15 citations