Water-gas shift reaction at medium temperature:a new approach in the industrial H2 production

In recent years, fuel market and environmental policies have reinforced the key role of hydrogen, increasing dramatically the demand and the research on its application in fuel-cell systems. Hydrogen is not an energy source, but an energy carrier, therefore, it must be produced: currently, the stream reforming of hydrocarbons is the largest and most economical way to produce H₂. fundamental in the steam reforming plant, is the water-gas shift reaction, a catalytic step to convert CO and obtain additional H2 by reaction with steam, which is moderately exothermic and favoured at low temperature. Currently, the water-gas shift reaction is performed using two units: a) 1^st unit using Fe₂O_3/Cr₂O₃, known as high temperature shift catalyst, that operates at high inlet temperature(370-400℃); b) 2^nd unit operating at the lowest possible inlet temperature(200℃), buy using a CuO/ZnO/Al₂O₃ catalyst, low temperature shift catalyst. In this way, it is possible to achieve exit CO concentrations as low as 0.1-0.3%. For the shift catalyst, the criticality comes on lower temperature activity and thermal stability. Thus, it may be advantageous to combine these two steps in only one unit, operating at medium temperature shift, with an exit temperature of about 320℃. The shift from high temperature shift to medium temperature shift processes offers further advantages, low steam to carbon ratios can be achieved, without sintering and F-T side reactions.

In this preliminary study, some Cu-based commercial catalysts has been studied as a function of different parameters(pressure, temperature, steam to dry gas ratio and contact time), not only to check the catalytic activity at middle temperature, but also to define best operating conditions for a good CO conversion, operating at low steam to dry gas ratio and contact time values, thus ensuring a lower overhead of the plant.

 A critical point in the water-gas shift unit is the catalyst deactivation, caused by a sintering accelerated by poisoning, because in the presence of sulphur and chlorine, low-melting-point compounds are formed on the catalyst surface. Thus, in a second step a deactivation study has been performed to predict the volume required to operate for a desired time. The temperature profile along the reactor gave precious information on the catalyst activity: the top of the catalytic bed is quickly deactivated, acting as a guard layer, in which poisons are preferentially adsorbed.

A design tool has been programmed with visual basic for applications to simulate the water-gas shift reactor. The fitting of temperature  regarding the origin load in a medium temperature shift was very good in all the ranges of time-on-stream below 3 years of aging. Using the simulation tool with the deactivation law determined from the plant data, it was possible to link the volume of active catalyst (excluding the catalytic guard, assumed to be fully deactivated) to the approach to equilibrium. The catalyst volume was calculated to achieve 90% of maximum conversion after 3 years of time-on-stream; however, these calculated values on the profiles measured along the medium temperature shift reactor. Moreover, this tool cannot be extrapolated to cases with different specification.