Posco and LanzaTech to collaborate on conversion of flue gas to ethanol and key value-added products
Report: Toyota to bring Home EV chargers to market in 2012; expects to sell 50K PHEVs worldwide that year

Study of emissions from CNG- and gasoline-fueled autorickshaws suggests measurements and data, not assumptions, should drive policy

Fuel-based particulate matter emission factors (A: PM2.5 mass, B: organic carbon fraction (OC/PM), C: elemental carbon fraction (EC/PM)), global warming commitment (D: GWC-Kyoto, E: GWC-All), and F: fuel consumption. Credit: ACS, Reynolds et al. Click to enlarge.

Chassis dynamometer emission testing conducted on 30 in-use Indian auto-rickshaws—two-stroke and four-stroke CNG-fueled (CNG-2S and CNG-4S) and four-stroke gasoline-fueled (PET-4S)—found that global warming commitment (GWC) associated with emissions from CNG-2S was more than twice that from CNG-4S or PET-4S, due mostly to CH4 emissions. The mean fuel-based PM2.5 emission factor (mean (95% confidence interval)) for CNG-2S (14.2 g kg-1 (6.2-26.7)) was almost 30 times higher than for CNG-4S (0.5 g kg-1 (0.3-0.9)) and 12 times higher than for PET-4S (1.2 g kg-1 (0.8-1.7)).

The study by the team from the University of British Columbia, published in the ACS journal Environmental Science & Technology, suggests that “comprehensive measurements and data should drive policy interventions rather than assumptions about the impacts of clean fuels”.

In 1998, responding to very poor air quality, the Indian Supreme Court ordered all public transportation vehicles in Delhi (including three-wheeled passenger carriers or auto-rickshaws), to operate on “clean fuel”. By 2003, all buses, taxis and auto-rickshaws in Delhi were retrofitted to run on CNG; only new vehicles with CNG engines are now available. (Of the four-stroke vehicles, a subset also has functioning backup gasoline fuel systems.) Other jurisdictions in India and other countries have switched or are considering switching to CNG-fueled vehicles, citing environmental considerations, the authors note.

This study measured a comprehensive set of emissions relevant to AQ [air quality] (PM2.5 and its major components organic carbon (OC) and elemental carbon (EC), CO, THC, NOx) and climate change (CO2, CH4, OC, EC, CO, and NMHC) from a sample of in-use auto-rickshaws representing the on-road fleet in Delhi. Auto-rickshaws from Delhi were chosen for the study because (a) they operate on CNG, (b) they are “for-hire” vehicles, and are hence readily recruited, and (c) they all have the same basic chassis design and function (manufactured by one company, but available with different engine types and a wide range of model years). In Delhi, there are more than 55,000 CNG auto-rickshaws for hire, providing flexible mobility that is faster and more reliable than city buses while costing less than conventional taxis. There are an estimated 3 million auto-rickshaws in India, around two-thirds of the world total.

The overall aims of this study were to (a) determine EFs [emission factors] from light-duty CNG-fueled engines for use in inventories of health- and climate-relevant emissions, (b) compare emissions from two-stroke and four-stroke CNG engines, (c) assess the extent to which emissions from in-use CNG vehicle vary and exceed Bharat Norms for new vehicles, and (d) explore the influence of fuel choice and vehicle age on emissions from a subset of the vehicles.

—Reynolds et al.

Among the many other findings of the study were:

  • Most emission species had very wide intervehicle variability within groups, confirming the value of multiple tests for developing fleet EFs.

  • Despite the wide variation within vehicle types, substantial differences in EFs were still found between vehicle categories, for example mean CH4 EFs from CNG- 2S (310 g kg-1 (281-344)) is significantly different from CNG-4S (50 g kg-1 (38-62)).

  • When the four-stroke auto-rickshaws were separated into “new” and “old” groups and their emissions compared by fuel type, no significant difference was observed for any exhaust constituents except for NOx from the gasoline-fueled vehicles.

  • Fuel consumption for PET-4S (3.8 kg 100 km-1) was in fact almost double that of CNG-4S group (2.1 kg 100 km-1). The authors noted that older vehicles may run fuel-rich either due to poor fuel system condition or because operators and mechanics purposefully tune them thus to improve engine power and combustion reliability at the expense of fuel consumption.

  • Replacing the remaining 2-stroke auto-rickshaws in Delhi would have nearly the same effect as switching from gasoline to CNG in 10 times as many four-stroke vehicles.

Our findings suggest that CNG fuel should be limited to use in four-stroke engines to realize potential health benefits (low PM2.5 and THC, less CO than PET-4S) and climate benefits (lower fuel consumption, less GWC than CNG-2S). Although the two-stroke engines examined in this study were all CNG-fueled, other studies suggest that two-stroke engines fueled with gasoline or LPG (another popular “clean fuel” in India) have similarly high PM2.5, CO and THC EFs. These studies indicate that a switch from gasoline or LPG to CNG in two-stroke engines may bring a sizable PM emissions reduction (of around a half), however emissions are still many times higher than the “acceptable” levels of CNG-4S vehicles. The average CNG-2S vehicle emitted nearly 3 orders of magnitude (700 times) more PM than the new CNG-4S test vehicle.

...regulators should also consider following the example of Dhaka, Bangladesh, who banned use of two-stroke vehicles completely. This study has shown the need for comprehensive test programs that develop EFs based on multiple vehicles: accurate health- and climate-relevant EFs are essential inputs for vehicle emission inventories or models, and an important step toward implementing appropriate policies.

—Reynolds et al.


  • Conor C.O. Reynolds, Andrew P. Grieshop, Milind Kandlikar (2011) Climate and Health Relevant Emissions from in-Use Indian Three-Wheelers Fueled by Natural Gas and Gasoline. Environmental Science & Technology Article ASAP doi: 10.1021/es102430p



The written article does not always correspond with data on charts with a lot less differences between various engine and fuel types.

The comments to this entry are closed.