A packed bed absorber-stripper system was used to remove selected indoor pollutants during the dehumidification of air by triethylene glycol solutions. Triethylene glycol concentrations of 90% and 95% by weight in water were used. Both random and structured packings were employed to provide the contact surface between the liquid and gas phases. A six-inch I. D. absorption column was operated between 50% and 80% of flooding conditions. The heights of a transfer unit for mass transfer for randomly packed 5/8-inch polypropylene Flexi rings and 1/2-inch ceramic Intalox saddles varied from 0.12 m to 0.17 m when dehumidifying air only. However, the height of a transfer unit was in the range of 0.31 m to 0.40 m for the cross corrugated cellulose and PVC structured packings. Heat and mass transfer coefficients were also calculated from the experimental data and were correlated with various process variables. The values predicted by these correlations were within +/- 10% of the experimental data. Pollutants used in the study included formaldehyde, toluene, 1,1,1-trichloroethane and carbon dioxide, and their concentrations in the air were controlled to 0.02 +/- 0.005 ppm, 3 +/- 0.02 ppm, 24 +/- 0.1 ppm, and 1000 +/- 5 ppm; respectively. Although nearly 100% of the toluene and 1,1,1-trichloroethane were removed by the 95% triethylene glycol solution, only 56% of the carbon dioxide and 30% of the formaldehyde could be removed from the air stream under similar conditions. As expected, the removal of these pollutants by the triethylene glycol solution was not affected by varying the relative humidity of the air.