Airborne carbonyls in large office buildings in the Midwestern United States have been studied both indoors and near the air-intakes environment using HPLC and GCMS methods. Air sampling for carbonyls was conducted using 2.4-dinitrophenylhydrazine coated silica gel cartridges and stainless steel Tenax TAB tubes. 1,3-dinitrobenzene, 2.4-dinitrophenol and 2.4-dinitroaniline were tentatively identified as degradation products of 2.4-dinitrophenylhydrazine by HPLC and GCMS analysis. Performance of the Tenax TAB-GCMS method under field sampling conditions was studied and Tenax TAB data was validated, to estimate the lowest possible concentration of carbonyls in the environment, using a correction term derived from error propagation analysis. Analytical recovery, analytes' breakthrough percentage, and limitations of the DNPH-cartridge sampling are presented along with GCMS ion fragmentation data of the tentatively identified product(s). Carbonyls including formaldehyde (6.8 i 8.8 ppbv)! acetaldehyde (2.8 i 3.2 ppbv), acetone (4.7 i 8.9 ppbv), methyl ethyl ketone (0.4 i 0.7 ppbv), 4-methyl-2-pentanone (0.5 1.7 ppbv). n-hexanal (0.2 i 0.6 ppbv) and n-heptanal (0.2 i 1.8 ppbv) were present in indoor air. Outdoor concentrations were: formaldehyde, 2.1 f 3.1 ppbv; acetaldehyde, 2.2 i 2.4 ppbv; acetone, 1.4 i 3.1 ppbv; methyl ethyl ketone, 0.6 i 0.4 ppbv; 4-methyI-2-pentanone. 0.2 i 0.1 ppbv; n-hexanal. 0.2 i 0.5 ppbv; and n-heptanal (0.2 i 0.4 ppbv). Total carbonyl concentrations (&,,) were 15.6 i 13.2 ppbv and 6.9 i 5.1 ppbv in indoor and outdoor air respectively. Average formaldehyde levels (7ppbv) in non-complaint buildings were lower than the recommended concentration levels in indoors of residential environment ( 100 ppbv). (ng . L-') 24.45 (molecular weight) ppbv =