Two severity indices for scale-up of steam cracking coils

A direct experimental scale-up method is presented for steam cracking of hydrocarbon feedstocks. It is based on the "severity" concept but uses two severity indices instead of one to unambiguously characterize the product yields for a given feedstock. Reaction path analysis suggests that the combination of the ethylene/ethane yield ratio and the methane yield characterizes the complete product spectrum. Simulation results for n-butane cracking in a pilot plant reactor, a Lummus SRT-1 reactor, and a 4-2-1 split coil reactor confirm that the same product spectrum is obtained for identical values of both indices in the different reactor geometries although the process conditions differ strongly. The soundness of the approach is further corroborated by experimental results obtained from a pilot plant reactor and a small scale Uno-Quattro coil. This approach is only valid when similar feedstocks are used. A rule of thumb is that the highest PIONA weight fraction for a feedstock cannot deviate by more than 5% of the corresponding value for the reference feedstock. Stronger deviations in feedstock composition lead to larger differences between the product spectra.