A study of carbonaceous aerosol was initiated in Nanchang, a city in eastern China, for the first time. Daily and diurnal (daytime and nighttime) PM(2.5) (particulate matter with aerodynamic diameter <= 2.5 mu m) samples were collected at an outdoor site and in three different indoor environments (common office, special printing and copying office, and student dormitory) in a campus of Nanchang University during summer 2009 (5-20 June). Daily PM(10) (particulate matter with aerodynamic diameter <= 10 mu m) samples were collected only at the outdoor site, whereas PM(2.5) samples were collected at both indoor and outdoor sites. Loaded PM(2.5) and PK(10) samples were analyzed for organic and elemental carbon (OC, EC) by thermal/optical reflectance following the Interagency Monitoring of Protected Visual Environments-Advanced (IMPROVE-A) protocol. Ambient mass concentrations of PM(10) and PM(2.5) in Nanchang were compared with the air quality standards in China and the United States, and revealed high air pollution levels in Nanchang. PM(2.5) accounted for about 70% of PM(10), but the ratio of OC and EC in PM(2.5) to that in PK(10) was higher than 80%, which indicated that OC and EC were mainly distributed in the fine particles. The variations of carbonaceous aerosol between daytime and nighttime indicated that OC was released and formed more rapidly in daytime than in nighttime. OC/EC ratios were used to quantify secondary organic carbon (SOC). The differences in SOC and SOC/OC between daytime and nighttime were useful in interpreting the secondary formation mechanism. The results of (1) OC and EC contributions to PM(2.5) at indoor sites and the outdoor site; (2) indoor-outdoor correlation of OC and EC; (3) OC-EC correlation; and (4) relative contributions of indoor and outdoor sources to indoor carbonaceous aerosol indicated that OC indoor sources existed in indoor sites, with the highest OC emissions in 12 (the special printing and copying office), and that indoor EC originated from outdoor sources. The distributions of eight carbon fractions in emissions from the printer and copier showed obviously high OC1 (> 20%) and OC2 (similar to 30%), and obviously low EC1-OP (a pyrolyzed carbon fraction) (< 10%), when compared with other sources.