You can fit a kinetic model (like a Langmuir -Hinshelwood type) to the experimental data. You can use the approximation of differential reactor that can be easily calculated because integration of the plug-flow design equation, but at low conversions, only. With the kinetic constants at different temperatures and Arrhenius analysis, the apparent Ea can be calculate.

Right, but please make sure that the data are from kinetic control range.

You can evaluate the mass transfer effect (to determine the conditions of kinetic control range) in our work where we used a catalytic reactor:

La-, Mn- and Fe-doped zirconia catalysts by sol-gel synthesis: TEM characterization, mass-transfer evaluation and kinetic determination in the catalytic combustion of trichloroethylene. Colloids and Surfaces A: Physicochemical and Engineering Aspects. 371, 81-90, 2010

Normally, our test were performed at very high GHSV (up to 100 000 h-1), so I think that the data are in kinetic control range

@Javier: at low converions --> in range 0-10% is OK?

can you give me more detail about the calculation:

you mean k = Ae^(-Ea/RT) ≃ CO2%

--> ln(CO2%) = constant -Ea/RT

plot ln(CO2%) vs 1/T --> Ea

I used this method but the results are not reasonable

Thank you

0 to 10% can be OK. The approximation ln(CO%) = constant-Ea/RT can be a good first try. You must to compare your Ea results with other works.