Emissions of NH 3 and nine nitrogen-containing volatile organic compounds (NVOCs) (C 1−3 -amines, C 1−3 -amides, and C 1−3 -imines) from motor vehicles powered by gasoline, diesel, and natural gas under low-speed driving conditions from roadside in situ measurements were characterized using a water-cluster chemical ionization mass spectrometer and trace gas monitors. The total emission strength of diesel trucks was the greatest followed by those of gasoline cars and natural gas cars. NH 3 emission per vehicle was found to be 2−3 orders of magnitude greater than that of all NVOCs, regardless of the type of vehicle. Although much lower than the emissions of amides or imines, emissions of amines were sufficient to produce atmospheric concentrations exceeding the threshold level for amines to enhance atmospheric nucleation by several orders of magnitude. Different engine emission reduction technologies (e.g., three-way catalytic converter vs selective catalytic reduction) can lead to different NH 3 and NVOC emission profiles. During the lifetime of a vehicle, its emission level was most likely to increase with its mileage. Source profiles of NH 3 and NVOC emissions from the three types of vehicles were also obtained from the measurements. These profiles can be a valuable contribution to the air pollution management system in terms of source apportionment, elucidating the emission contributions from a specific type of vehicle.

Alkaline gases, including NH3, C1–3-amines, C1–3-amides, and C1–3-imines, were measured in situ using a water cluster-CIMS in urban Beijing during the wintertime of 2018, with a campaign average of 2.8 ± 2.0 ppbv, 5.2 ± 4.3, 101.1 ± 94.5, and 5.2 ± 5.4 pptv, respectively. Source apportionment analysis constrained by emission profiles of in-use motor vehicles was performed using a SoFi-PMF software package, and five emission sources were identified as gasoline-powered vehicles (GV), diesel-powered vehicles (DV), septic system emission (SS), soil emission (SE), and combustion-related sources (CS). SS was the dominant NH3 source (60.0%), followed by DV (18.6%), SE (13.1%), CS (4.3%), and GV (4.0%). GV and DV were responsible for 69.9 and 85.2% of C1- and C2-amines emissions, respectively. Most of the C3-amines were emitted from nonmotor vehicular sources (SS = 61.3%; SE = 17.8%; CS = 9.1%). DV accounted for 71.9 and 34.1% of C1- and C2-amides emissions, respectively. CS was mainly comprised of amides and imines, likely originating from the pyrolysis of nitrogen-containing compounds. Our results suggested that motor vehicle exhausts can not only contribute to criteria air pollutants emission but also promote new particle formation, which has not been well recognized and considered in current regulations. Urban residential septic system was the predominant contributor to background NH3. Enhanced NH3 emissions from soil and combustion-related sources were the major cause of PM2.5 buildup during the haze events. Combustion-related sources, together with motor vehicles, were responsible for most of the observed amides and imines and may be of public health concern within the vicinity of these sources.