PNNL-led international study finds ultrafine aerosols have outsize impact on storm clouds, precipitation
Gasoline, diesel and older hybrids may be banned in parts of East London during parts of work week

CMU study finds secondary organic aerosol production from PFI and GDI vehicles has been “substantially” reduced by tightening NMOG standards

Secondary organic aerosol (SOA) is a major component of fine particulate matter—which causes adverse health effects—even in urban environments; vehicles may be a dominant source of SOA in urban areas, although there is still an ongoing debate over SOA formation from on-road sources.

Now, in a new study published in the ACS journal Environmental Science & Technology, researchers at Carnegie Mellon University have found that secondary organic aerosol (SOA) production from both port fuel injection (PFI) and gasoline direct injection (GDI) vehicles has been substantially reduced through the tightening of NMOG emissions. Further, their data suggest that this trend will continue with the implementation of the new federal Tier 3 and California LEV III standards.

Further, they found no difference in SOA production between PFI and GDI vehicles certified to the same standard. As a result, they suggested that the ongoing, significant shift from PFI to GDI vehicles in the US vehicle fleet should not alter SOA production.

However, they also cautioned, SOA production also depends on atmospheric conditions, specifically the NOx regime; therefore only tightening NMOG emissions standards may not be enough to reduce urban SOA levels.

Although past studies have provided substantial insight, important gaps remain in our understanding of SOA formation from gasoline vehicle exhaust due to changing engine technology and emissions certification standards. California and the federal government are both phasing in new, more stringent regulations (LEV III and Tier 3, respectively). These standards meet or exceed the most stringent existing regulations, the California super ultra-low emission vehicle (SULEV) standard. In addition, largely driven by increases in the corporate average fuel economy standards, a dramatic change in gasoline vehicle engine technology is occurring in the United States. Historically, the US fleet has been dominated by vehicles equipped with port-fuel injection engines (PFI vehicles), but the market share of vehicles equipped with gasoline direct injection engines (GDI vehicles) has increased dramatically over the past decade, reaching ~50% of new gasoline vehicles sold in the United States in 2016.

… In this study, we investigated SOA formation from a fleet of 16 on-road gasoline vehicles using a Potential Aerosol Mass OFR (PAM reactor, hereafter) during chassis dynamometer testing. The test fleet consisted of both PFI and GDI vehicles certified to a range of emissions standards from federal Tier 0 to California super-ultra low emission vehicles (SULEV). We investigate the effects of GDI technology, tightening of emissions standards, and cold-start versus hot-stabilized operations on SOA formation. Finally, we evaluate the use of an OFR as screening tool for SOA production from on-road gasoline vehicle exhaust.

—Zhao et al.

Of the 16, 10 vehicles were equipped with gasoline direct injection engines (GDI vehicles) and six with port fuel injection engines (PFI vehicles) certified to a wide range of emissions standards.

For the study, dilute tailpipe exhaust from the gasoline vehicles was photo-oxidized using a PAM reactor during chassis dynamometer testing at the California Air Resources Board’s (CARB) Haagen-Smit Laboratory. The tailpipe emissions depend on the details of engine design, engine calibration, aftertreatment system, vehicle age, and maintenance history.

All vehicles were tested using the cold-start Unified Cycle (UC), which is widely used for emissions testing. Prior to testing, each vehicle was preconditioned with an overnight soak and without evaporative canister purge. Each vehicle was refueled at the CARB Haagen-Smit Laboratory with the same commercial gasoline fuel that met the California summertime fuel standard. Major fuel components included 49% paraffins, 25% aromatics, 14% olefins, and 10% ethanol (wt%).

We measured similar SOA production from GDI and PFI vehicles certified to the same emissions standard; less SOA production from vehicles certified to stricter emissions standards; and, after accounting for differences in gas-particle partitioning, similar effective SOA yields across different engine technologies and certification standards. Therefore the ongoing, dramatic shift from PFI to GDI vehicles in the United States should not alter the contribution of gasoline vehicles to ambient SOA and the natural replacement of older vehicles with newer ones certified to stricter emissions standards should reduce atmospheric SOA levels. Compared to hot operations, cold-start exhaust had lower effective SOA yields, but still contributed more SOA overall because of substantially higher organic gas emissions. We demonstrate that the PAM reactor can be used as a screening tool for vehicle SOA production by carefully accounting for the effects of the large variations in emission rates.

—Zhao et al.


  • Yunliang Zhao, Andrew T. Lambe, Rawad Saleh, Georges Saliba, and Allen L. Robinson (2018) “Secondary Organic Aerosol Production from Gasoline Vehicle Exhaust: Effects of Engine Technology, Cold Start, and Emission Certification Standard” Environmental Science & Technology doi: 10.1021/acs.est.7b05045



Damn if you do, damned if you don't. Gas engines have been upgraded to meet the ever demanding fuel mileage requirements which in some includes the idea of direct injection, similar to diesel engines. But, direct injection causes an increase in smog which must be suppressed to meet smog standards. This in turn negatively impacts fuel mileage. Some people believe that ICEVs are at the point where they can't meet the mileage standards and still meet the smog standards.


GDI does not necessarily give more NMOG than PFI. During cold starts at low ambient temperature GDI might even be advantageous. Other driving patterns might give opposite results. On average, about equal, as they state. The BIG environmental problem with GDI is particle emissions and this can only be solved by using GPF.


It should be noted that a previous study by CMU showed that increasing the ambient NMOG:NOx ratio increased SOA production non-linearly; increasing NMOG:NOx ratio from 4:1 to 10:1 increased SOA yield by a factor of 8.

Thus, decreasing NMOG emissions while leaving NOx emissions as is would decrease the NMOG:NOx ratio and result in much lower SOA production.


GDI's particulate issue really falls on how it is tuned/operated. GDI provides an extreme advantage of low end torque compared to PFI, but it does so by dumping in fuel when a user demands a lot of power.

Driving habits and system tuning really play a role more so than how the fuel is delivered, overall I would say that GDI is a significant improvement over port. If we make our cars less responsive, and stop driving with our foot to the floor from stops, our economy and air quality would improve greatly.

I don't think a particulate filter would be necessary, it would cause lots of issues that the general public wouldn't know how to handle. Just imagine a soccer mom being derated and forced to regen on the way to school or some other function.
That and it could reduce fuel economy, and lead to more emissions.

The comments to this entry are closed.