Multiscale modelling of packed bed chemical looping reforming

Singhal, Arpit; Cloete, Schalk; Quinta‐Ferreira, Rosa M.; Amini, Shahriar

doi:10.1016/j.egypro.2017.10.287

Cited by 5 publications

(7 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The model is consistent with the previous work of the authors [10][11][12][13]. The closures [8,14] represented in Equations (4)- (7) are used in conjunction with the same reaction kinetics and boundary conditions as the PR-DNS simulation.…”

Section: D Packed Bed Modelsupporting

confidence: 72%

“…In this region, the hydrogen mole fration is low, emphasizing the effect of the varying reaction rate constant as defined in Equations (10) and (11). The solid lines in Figure 3 show that the adjustment to the effective diffusivity proposed in Equation (13) with the model coefficient set to 0.2 resulted in an excellent match to the PR-DNS results.…”

Section: Comparison Of 1d Model To Pr-dns Resultssupporting

confidence: 52%

“…A growth rate of 20% is used to propagate the mesh into the domain. More detail can be obtained from previous work of the authors [8][9][10][11]13].…”

Section: Pr-dns Setupmentioning

confidence: 99%

“…Firstly, PR-DNS is used on a geometry of~100 densely packed mono-disperse spherical particles (ε = 0.355) extracted according to the methodology outlined in our previous works [8][9][10]. Secondly, the PR-DNS results are used to improve a computationally affordable 1D packed bed model [10][11][12][13] which is based on appropriate models for effectiveness factor [14] and external heat and mass transfer [8]. This method does not resolve any gradients inside the particles as is done in the models utilizing the colocation method [7,15].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Multiscale Modeling of a Packed Bed Chemical Looping Reforming (PBCLR) Reactor

et al. 2017

Self Cite

View full text Add to dashboard Cite

Packed bed reactors are broadly used in industry and are under consideration for novel reactor concepts such as packed bed chemical looping reforming (PBCLR). Mass and heat transfer limitations in and around the particles in packed bed reactors strongly affect the behavior of these units. This study employs a multiscale modeling methodology to simulate a PBCLR reactor. Specifically, small-scale particle-resolved direct numerical simulation is utilized to improve large-scale mass transfer models for use in an industrial scale 1D model. Existing intra-particle mass transfer models perform well for simple first order reactions, but several model enhancements were required to model the more complex steam methane reforming reaction system. Three specific aspects required enhanced modeling: the generation of additional gas volume by the reforming reactions, the lack of clear reaction orders in the equilibrium reactions, and the diffusion of multiple reactant species into the particle. Large-scale simulations of the PBCLR reactor with the enhanced 1D model showed that the highly reactive Ni-based catalyst/oxygen carrier employed allows for the use of large particle sizes and high gas flowrates, offering potential for process intensification.

show abstract

Section: D Packed Bed Modelsupporting

confidence: 72%

Section: Comparison Of 1d Model To Pr-dns Resultssupporting

confidence: 52%

“…A growth rate of 20% is used to propagate the mesh into the domain. More detail can be obtained from previous work of the authors [8][9][10][11]13].…”

Section: Pr-dns Setupmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Multiscale Modeling of a Packed Bed Chemical Looping Reforming (PBCLR) Reactor

et al. 2017

Self Cite

View full text Add to dashboard Cite

show abstract

“…Only in this case, the void fraction is constant in all cells and the geometry is meshed only in one direction. More details can be found in the previous work of the authors [44] and the applications in subsequent studies [32,33]. The domain is discretized into 100 cells with the length of the geometry equal to the PR-DNS geometry as given in Table 1.…”

Section: D Packed Bed Modelmentioning

confidence: 99%

Verification of Heat and Mass Transfer Closures in Industrial Scale Packed Bed Reactor Simulations

et al. 2018

Self Cite

View full text Add to dashboard Cite

Particle-resolved direct numerical simulation (PR-DNS) is known to provide an accurate detailed insight into the local flow phenomena in static particle arrays. Most PR-DNS studies in literature do not account for reactions taking place inside the porous particles. In this study, PR-DNS is performed for catalytic reactions inside the particles using the multifluid approach where all heat and mass transfer phenomena are directly resolved both inside and outside the particles. These simulation results are then used to verify existing 1D model closures from literature over a number of different reaction parameters including different reaction orders, multiple reactions and reactants, interacting reactions, and reactions involving gas volume generation/consumption inside the particle. Results clearly showed that several modifications to existing 1D model closures are required to reproduce PR-DNS results. The resulting enhanced 1D model was then used to accurately simulate steam methane reforming, which includes all of the aforementioned reaction complexities. The effect of multiple reactants was found to be the most influential in this case.

show abstract