A pseudo-local heat transfer approach in a low tube to particle diameter ratio packed bed catalytic reactor: Oxidative dehydrogenation of ethane as a case study

Avoiding false runaway prediction by properly describing axial conductive heat transfer in a low dt/dp industrial-scale packed bed reactor

“…Then, the Pseudo-local Internal Heat Transfer Approach (PL-IHTA), accounting for the impact of keff,z, is applied to describe experimental temperature profiles obtained from previous works. A bench and industrial-scale packed bed operated in non-adiabatic and non-isothermal mode are considered for this purpose [7]. The third scenario represents the reactor for the oxidative dehydrogenation of ethane as a case study to examine the impact of keff,z under reaction conditions.…”

“…It is worth noting that the experimental set-up for the non-adiabatic and non-isothermal operation (left side of Figure 1) is the same as the one used in previous works from our research group [6,7]. At each axial positions the temperature is measured at twelve radial positions, i.e., moving the thermocouples over the entire diameter of the packed bed, while the tube is heated by an electric resistance.…”

“…This is critical for accurately predicting hot spot and runaway situations. Previous publications from our research group [3,[7][8][9] aimed at proposing a general methodology to describe radial heat transfer elucidating its impact on temperature profile simulations for a wall-cooled packed bed reactor with a low dt/dp ratio. In these publications, as well as in others in the literature [10][11][12][13][14], it was identified that the role of axial heat transfer by conduction in a packed bed with a low dt/dp ratio had not been appropriately addressed.…”

“…However, the impact of keff,z on temperature profile simulations under industrial operating conditions, i.e., at high Rep, (> 700) [2,10,[18][19][20] and large L/dp ratios (>50) [21,22], is neglected. Thus, although the impact of the axial conductivity on the modeling of packed bed reactors with a high dt/dp ratio has been well-established in the literature [23,24], its impact on temperature profile predictions when modeling wall-cooled packed bed reactors with a low dt/dp ratio is yet to be elucidated, more particularly when highly exothermic reactions occur and when fluid dynamics and radial heat transfer are more important [7,9,25]. In several studies, the axial conductivity has been disregarded from the reactor model [10,11,[26][27][28][29], while in a few studies that have considered it, keff,z is calculated using literature correlations that were developed under conditions that differ significantly from the industrially relevant ones, i.e., Rep < 700 and dt/dp > 8 [3,25,[30][31][32][33][34].…”

“…Next, the impact of the effective axial thermal conductivity in a non-isothermal bench-scale and industrial-scale packed bed in absence of reaction is assessed by comparison of the experimental temperature profile with the simulated ones using different correlations for keff,z, i.e., (i) the most widely used one developed by Yagi et al [12], (ii) the generic one developed by Votruba et al at Rep up to 1000 for a packed bed with 3.85 < dt/dp < 55.77 [17], and (iii) the specific one developed in this work. The Pseudo-local Internal Heat Transfer Approach, previously developed in our research group [7], is used to account for the radial temperature gradients which are particularly pronounced in the wall zone. Finally, the impact of the axial conductivity, considering the three different correlations, is assessed under reaction conditions taking the Oxidative Dehydrogenation of Ethane (ODH-C2) over a V-based multi-metallic catalyst as the case study.…”