Heat transfer parameters in syngas units

Absolute values of the heat transfer parameters that ean be used for seale-up are dififieult to determine in bench-scale units due to the very high gas velocities and heat fluxes in industrial units. Attempts to determine them in pilot plants operating at industrial conditions but without reaction are also highly uncertain due to small driving forces. The steam reforming reaction is, however, strongly endothermic and limited by chemical equilibrium. This implies that for a new catalyst, the reaction will be close to chemical equilibrium in the major part of the tube, so variation in catalyst activities will only have a small impact on the temperature profile (refer to Section 3.3.7). If, however, heat transfer 

The actual catalyst shape influences the heat transfer coefficient through its actual size and shape. Void may enter some correlations, e.g. in [296] using equations for pressure drop (refer to Section 3.3.4), but they do not distinguish between external and internal void. CFD simulation in [160] shows a minor impact of internal void on the radial temperature profiles. 

Similar results have also been obtained for different commercial and development cylindrical type catalysts with one or more holes in a 70 mm tube, where the tube-to-particle diameter ratio ranged from 3.4 to 5, but at identical operating conditions [120]. The average heat flux was as large as 160 kW/m. It should be added that the pressure drop varied with more than a factor 2 between the individual experiments. The small overall particle size dependence on the heat transfer coefficient in Equation (3.25) where the dp is raised to the power a-1 appears thus to be correct. 

Nearly all measurements in the literature to determine heat transfer parameters are without reaction as discussed in [11] [520]. The latter reference concludes that fundamentally there is no reason for any impact of the reaction itself, but the catalyst has, of course, a major impact on the temperature within the tube and on the tube wall, since a highly active catalyst will give significantly lower temperatures compared with a catalyst with low activity (refer to Example 3.4). But this does not prove that the reaction itself has an impact on the heat transfer rate, and it is usually not included in the modelling. 

Since a major fraction of the heat added in a tubular reformer is used for reaction, heat fluxes are significantly higher in the tubular reformer with reaction compared with bench-scale measurements widiout reaction. Parameter identification in measurements without reaction is highly uncertain in a tubular reformer due to the small temperature differences, but experiments with reactions under very different operating conditions can be used to check if the reaction has any impact on the heat transfer rate. 

---