Hydrogen purification by selective methanation of CO in CO/CO2/H2

Normally, the hydrogen-rich feed gas to PEM Fuel Cells has a relatively high concentration of CO, which is poisonous to the anode catalyst. CO can be removed by selective oxidation or by methanation. For methanation, it is desired to minimize the use of hydrogen by selectively removing the CO without “loosing” hydrogen by methanating CO2.

Studies have shown that CO has an inhibiting effect on the methanation of CO2. This inhibiting effect is thought to influence the selective methanation of CO. The aim of this project has been to remove CO from a process gas containing H2, CO and CO2 by selective catalytic methanation. We have examined the inhibiting effect of CO on the methanation of CO2 with Ni/alumina catalyst pellets of different sizes in an isothermal fixed bed reactor.

Experimental results indicate that the combined methanation of CO and CO2 can be described by two net reactions: Methanation of CO and the reverse water gas shift reaction: (1) CO + 3H2 -> CH4 + H2O (2) CO2 + H2 -> CO + H2O In experiments with only CO2 and hydrogen in the feed gas there is a significant production of CO and methane.

The concentration of CO in the reactor rapidly reaches a level, where it inhibits (or slows down) the CO2 conversion and the conversion of CO2 is effectively the same rate as the conversion of CO to methane. This is in agreement with an overall reaction path of CO2 -> CO -> methane, which can be described by (1) and (2).

The modeling efforts were divided into two parts. First, a model for a single catalyst pellet was derived and solved. This was incorporated in a model for the entire fixed bed reactor, which was used to get steady state reaction rates by fitting the experimental data by a least squares method. The figure shows an example of the detailed analysis of concentration profiles down through the reactor and inside the catalyst pellets/particles.

The small particles, which have a rather high effectiveness factor with respect to methanation of CO, have a high CO selectivity, whereas the larger pellets have very low selectivity even at high CO inlet concentrations. Negative selectivity, i.e. a net production of CO, was observed for pellets at low CO inlet concentrations at both 250°C and 350°C.

Negative selectivity was also observed at high reactor temperatures for pellets even at relatively high inlet concentrations of CO. The fundamental understanding of the combined effects of reaction kinetics and pore diffusion is crucial for interpreting the experimental data. We have found that the selectivity decreases by increasing the reactor temperature or catalyst particle size and when the CO inlet concentration is reduced.

As a result, the selectivity drops significantly in an integral reactor operating at high CO-conversion. The lower limit of CO concentration in the outlet is determined by the quasi-equilibrium between CO removal and CO production from CO2.