A comparative study of the elemental composition of the exhaust emissions of cars powered by liquefied petroleum gas and unleaded petrol

Elements emitted from the exhausts of new Ford Falcon Forte cars powered by unleaded petrol (ULP) and liquefied petroleum gas (LPG) were measured on a chassis dynamometer. The measurements were carried out in February, June and August 2001, and at two steady state driving conditions (60 and 80 km h(-1)). Thirty seven elements were quantified in the exhaust samples by inductively coupled plasma mass spectrometry (ICPMS). The total emission factors of the elements from the exhausts of ULP cars were higher than those of LPG cars at both engine speeds even though high variability in the exhaust emissions from different cars was noted. The effect of the operating conditions such as mileage of the cars, engine speed, fuel and lubricating oil compositions on the emissions was studied. To investigate the effects of these conditions, multivariate data analysis methods were employed including exploratory principal component analysis (PCA), and the multi-criteria decision making methods (MCDM), preference ranking organization method for enrichment evaluation (PROMETHEE) and geometrical analysis for interactive aid (GAIA), for ranking the cars on the basis of the emission factors of the elements. PCA biplot of the complete data matrix showed a clear discrimination of the February, June and August emission test results. In addition, (i) platinum group elements (PGE) emissions were separated from each other in the three different clusters viz. Pt with February, Pd with June and Rh with August; (ii) the motor oil related elements, Zn and P, were particularly associated with the June and August tests (these vectors were also grouped with V, Al and Cu); and (iii) highest emissions of most major elements were associated with the August test after the cars have recorded their highest mileage. Extensive analysis with the aid of the MCDM ranking methods demonstrated clearly that cars powered by LPG outperform those powered by ULP. In general, cars tested in June perform better than those tested in August, which suggested that mileage was the key criterion of car performance on the basis of elemental emission factors. (c) 2006 Elsevier Ltd. All rights reserved.