A dsorption by activated carbon is one among many ways to prevent air pollution. The purpose of this experiment is to test the ability of local activated carbon - made from coconut shell - to adsorp contaminant gas. The experiment was conducted under these conditions : small coloumn filled with active carbon with 4 x 8 , 6 x 12, and 8 x 16 mesh size variation, gas flow - provided by a compressed cylinder filled with 9800 ppm carbon dioxide - variations (2 and 3 L/minute), and regeneration.Cheremisinoff ( 1993) define adsorption as a reversible process which allow fluidmolecules to concentrate in a highly porous surface. Adsorption is used to concentrate or store contaminants until they can be recovered or destroyed in the most economical manner (Bethea, 1978). Adsorbate removal efficiency depend on the adsorbent size, the height of the adsorbent, and gas flowrate.The adsorbent coloumn was saturated by carbon dioxide. The C02 concentration living the coloumn is then measured by acid-base titration with phenolpthalein indicator.After 720 minutes, all adsorbe ts hich were operated in a 2 L/minute C02 flowrate was saturated, rneanwhile the other ad rbe ts being 'saturated by a 3 L/minute gas, saturate after 510 minutes running. The sa urate time differences happened because at a 3 I/minute flo rate so . g9 article can not be adsorbed due to the relatively rapid velocity. The 8 x 16 esh carbon removal efficiencies are 70,51 % for the 2 L/minute gas flowca e an 69,~3 % fo he - t;'mi te gas. Compared to the rest carbon this size shows ~he highest r o~al e 1 iency because of it's large surface area. After being saturated, all a - rbents wef regenerated 100°C for 15 and 20 minutes. The 8 x 16 mesh's weight after regeneration is the lightest of them all. In addition the amount of contaminant left in the surface pore is the smallest. Another coloumn runnings show a more faster saturated time and less efficiensies compare to the first experiments.The result of this experiment suggested that compare to the 4 x 8 and 6 x 12 mesh carbon, the 8 x 16 mesh activated carbon has the greatest removal eficiency in both flowrate operation. Reuse of those carbon after regenerated at 10D°C for 15 and 20 minute indicate a more faster saturated time and poorer removal efficiencies than the first experiment. In order to achieve higher eficiency carbon's properties like CTC, ash content, and moisture content need some adjusments. This can be done by modifying.the production's method and condition. To find out the ability of local activated carbon as-qas adsorbent more experiments should be made.