Arizona now permits sale of isobutanol-blended gasoline for on-road vehicles
Santa Clara Valley orders 47 New Flyer Xcelsior 60' diesel-hybrid buses

Swiss team concludes that particulate filters should be mandatory for GDI engines

Based on a three-year study of toxic and environmentally relevant pollutants from gasoline direct injection (GDI) engines, Swiss researchers have concluded that some GDI engines emit just as many soot particles as unfiltered diesel cars did in the past. Further, the GDI particles carry numerous carcinogenic substances. Based on this current data, they recommend that particulate filters be mandatory for GDI engines.

In the spring 2014, the GasOMeP project (Gasoline Vehicle Emission Control for Organic, Metallic and Particulate Non-Legislative Pollutants) got underway. The Paul Scherrer Institute (PSI), Bern University of Applied Sciences, the University of Applied Sciences and Arts Northwestern Switzerland, several industrial partners and Empa were all involved. The project was funded by the ETH Domain’s Competence Center for Energy and Mobility (CCEM) and coordinated by Empa chemist Norbert Heeb, who has made a name for himself in the last 25 years by analyzing diesel emissions and studying filter systems.

Figure 4 genotoxic PAHs pattern WLTC cold v3-png
Empa researchers Norbert Heeb and Maria Munoz discovered large quantities of benzo(a)pyrene (red bar)—a combustion product responsible for cutting short the lives of cigarette smokers—in the exhaust gas emitted by GDI engines. Dibenzo(ah)anthracene (pink) is also carcinogenic. The carcinogenic potential in one cubic meter of exhaust gas from gasoline direct injectors is up to 1,700 times higher than the EU limit for clean air. By contrast, diesel cars with particle filters exceeded the limit only 45-fold. Click to enlarge.

The results of the GasOMeP project were presented during a conference held at the Empa Academy in late March.

For the study, the team selected seven direct-injection gasoline cars, including a Mitsubishi Carisma (2001 model, exhaust emission standard Euro 3). The other vehicles were all built between 2010 (VW Golf, Euro 4) and 2016 (Citroën C4, Euro 6b). By way of comparison, a current Peugeot 4008 (2013, Euro 5b) with a diesel engine and a particle filter was also included. All the vehicles were tested based on the WLTP cycle (Worldwide Light-Duty Vehicles Test Pro-cedure), which will be mandatory for newly licensed models as of September 2017.

The results were sobering: every tested gasoline car emitted ten to 100 times more fine soot particles than the diesel Peugeot. Under the microscope, the particles from the gasoline engines were similar in size to the soot particles that had given diesel a bad name: primary particles measuring ten to 20 nanometers in size, which congregate into particle agglomerates measuring 80 to 100 nanometers before leaving the exhaust.

Once inhaled, these particles remain in the body forever.

—Norbert Heeb

The evidence shows that they can penetrate the membrane of human alveoli in the lungs and thus get into the bloodstream. However, the particles are not the only problem. Heeb notes that liquid or solid chemical toxins from the combustion process, including polycyclic aromatic compounds, accumulate on the surface of the particles, which can then smuggle these substances into the bloodstream like a Trojan horse.

Maria Munoz, a colleague of Heeb’s from Empa’s Advanced Analytical Technologies lab, took a closer look at the exhaust emissions from the vehicles tested in the GasOMeP project and discovered the combustion product benzo(a)pyrene, a known carcinogenic substance also found in cigarette smoke.

The World Health Organization (WHO) considers even the tiniest dose of benzo(a)pyrene harmful. The EU settled on an air limit of one nanogram per cubic meter. Levels in exhaust emissions were found to be as much as 1,700 times above this limit. Or to put it another way, one cubic meter of exhaust gas transforms up to 1,700 cubic meters of clean air into a mixture deemed carcinogenic according to the EU standard.

Once again, the diesel vehicle with particle filter fared much better: in the test, the Peugeot emitted only 45 nanograms of carcinogenic substances—6 times less than the best one of the analyzed gasoline cars.




I just avoided diesel and bought a car with a 1.2 turbo charged direct injection engine.
It looks like that is just as bad or worse.


YES, what have I been telling you for the last couple of years.


Don't buy these costly new technologies, they are put in place by al gore and arnold swarchzenneger to hike prices and emit more overall pollution. Real good technologies are actually hide from commercialisation. Thus do like me , only buy the absolute cheapest car and keep it a very long time. Nissan released some good technologies and this crooked climate change journalist lie about the mpg numbers, here is the link.



VW has already started fitting particulate filters to their gasoline engine cars (2017).



Hi laszioi:

I wasn't aware of that, so thanks for the information.

I did know about the particulate emissions from GDI, but forgot about it/did not prioritise it, and perhaps in the back of my mind since the filter is only around 50 Euros or so had thought that they might have fixed it.

If the option to retrofit is available, then I will certainly do so.


Difficult to know who should be trusted?


The filters don't add much to the cost of the car, if you have to add fluid now and then so what?


People have a hard time as it is simply doing maintenance to their vehicle. Having them use a gasoline car that needs to be maintained like a diesel is asking for trouble.

Most of the particulate can be mitigated by tuning of GDI. The filters and the sensors to operate the DPF system successfully are not cheap. DPF on gasoline equipment will quickly lead to the demise of the ICE, an increase of the average age of the cars on the road, and general distain from the public.

Upfront it might add $3000-4000 to the cost of the car. Manufacturers might just cut the losses and go back to port injection.

There is no good, cheap way to put particulate filters on small cars. Big trucks are much easier to bury the costs on. Also you have to drive like a mad man to bring the filters up to light off temperatures on normal short drives, meaning most people who granny their cars will spend tons of money trying to keep their cars from de-rating on them.


Also, please note in the article/flier when it says "What we blow out when we floor the throttle" WOT is open loop and usually not regulated as it accounts for a very small fraction of driving.

WOT on any ICE car will lead to detrimental emissions, the car can't operate safely and cleanly easily with those demands. WOT richens the mixture on gasoline ICEs increasing potential for soot, its open loop and dumping fuel in for power, the mixture is richer to keep the temperature down and to save the emissions equipment(when possible).

Brian P

People are confusing particulate filters with NOx control measures (SCR). Particulate filters don't require any additives. SCR is the one that requires the special fluid and dosing system.

It's currently thought that the higher exhaust temperature from a gasoline engine during cruising conditions will keep the particulate filter hot enough (unlike diesel applications), and the frequent short periods of coasting (which shuts off the fuel completely but need not shut off air supply) will provide sufficient oxygen for passive regeneration without needing any active method of regeneration, aside from perhaps holding the throttle open a bit during coasting to pump air through (this is a trivial matter for drive-by-wire, which everything has nowadays). If that proves to be the case, the filter is unlikely to require any maintenance until the engine gets old enough to start consuming lubricating oil ... then, look out.


So it is a ceramic filter, can't cost that much.


unlikely to require any maintenance until the engine gets old enough to start consuming lubricating oil ... then, look out.
Wheather the O2 sensor, catalytic ageing poor maintenance of other components and lubricants (A big stink in E.U. over disabling of SCR AD blue on trucks as a money saving excersise) the results are the same.

All components are subject to failure for numerous reasons as any mechanic (technician) knows.

There are anecdotal accounts of imported second hand cars from (in this example Japan) that have reached the end of economical life in that country with relatively low KLM's where the evidence suggests that the engines oils have never been changed.
Owners making economic decisions, knowing the cars will be scrap value most often neglect appropriate maintenance.

New cars are not uniformally compliant to any standard simply because they are 'new'. The number of faults on new vehicles has been shown over decades to be high(er in the first year) than average yet they are often exempt from even basic licensing requirements for (here .au) 5 years.

How many districts have appropriate testing?
I can say for sure - NONE. Some districts do random sampling. Just because the technology exists does not mean it meets the objective..

Evidence globally is proof that until last year testing authorities did not even have equipment capable of higher resolution diagnosis.

I doubt given the volumes of and technical realities of I.C.E. engine cars that any can hope to meet their social obligation.

Even if the magic formula could be achieved, we still must face facts that there are almost no vehicles in today's fleet capable of resolving the matter. We are constantly asked to take assurances on 'faith'.


The plain fact is ICEs are inefficient, costly, complicated polluting and obsolete. We have a clean form-factor that, why are we allowing the car companies to keep on, keeping on? If Trump and the Republicans were worth their salt, they would be assisting the car makers to transition to EVs and assisting the oil companies to re-invest in renewable energy...but alas, instead they encourage the companies to continue with fossil fuels and their nasty side effects...why, when the answer is there waiting to be implemented...makes no sense and makes them look like what they are...greedy dasters.


Engines will be with us for decades, to pretend everything will be EV real soon now is a fantasy.


More efficient GDI engines require higher octane. Higher octane fuel will effect higher BC emissions as the petrol octane booster chemistry is the culprit. This problem only exists per petrol industry demands upon limitation of the supply to their product. They refuse ethanol mid blends as they make a premium on high octane fuels. Unbiased agencies are finding ethanol will eliminate the problem. We don't need PM filters we need new fuel. UAI has done research into this and finds a lot of collusion between the regulators and petrol. This stuff is right out there in public domain and no politician will touch it as the petrol industry is keeping all fat and happy. No draining of the swamp or rocking the boat is desired, thank you very much. The suspects as usual are fat and happy the way things are and will only implement improvements that help their friends even at taxpayer cost.


Ethanol/gasoline blends also generate particles. So does E85, particularly at cold starts at low ambient temperatures. This was shown almost a decade ago. Thus, these engines also need filters, just as GDIs running on gasoline.

Besides soot, engines also produce ash from the lube oil. This you get also with gaseous fuels, e.g. CNG and hydrogen (they also need lubrication). Lube oil ash has severe health effects and thus, a filter is needed in this case as well.


If you look carefully at the graph, you see that only one engine in the graph is environmentally acceptable.


Not so, ethanol makes a tremendous improvement in PMs. Only when tested with bogus EPA test fuel designed by the petrol does ethanol look bad. The test fuel is not even close to average consumer fuel. The test fuel is designed to make ethanol fuel look bad upon test results with gasoline. Again, this info is within the public domain. Independent (non government) labs prove the accusation. The problem is the EPA doesn't care or aren't concerned. They are independent and beyond any evaluation. They do what they want and what they want is to gain political power and wealth.

Yes, ethanol zone of emissions is higher upon cold startup, but one must understand that ethanol fuel has zero engineering dedicated to minimize pollution as compared to gasoline or diesel fuel. If ever that would happen, ethanol has maximum ability to minimize pollution as a comparison. The chemistry of ethanol is simple and consistent. Since the fuel boasts of liquid oxygen half the problem is solved. Note that petrol additives for octane boost are unhealthy and bad for the environment let alone cost more as compared to ethanol. But what is petrol to do if ethanol takes over the high octane market? They will lose a lucrative refinery product in which they have no other high profit use for. They don't want to crack the fuel additive to plain gasoline for the sake of the environment. Much easier to utilize politics to force the transportation industry to engineer a lower polluting way to use the stuff. So what if consumers pay more. Consumers don't pick up the take for crony capitalism.


In Sweden, it was shown already in 2008 that harmful emission components at low ambient temperatures were higher from E85 than from gasoline. See link. This work has also been cited in so many other scientific publications that it is well-known in the scientific community (e.g. search for Westerholm & E85). I attached the link to a home page where the main report can be downloaded for free. There are also several peer-reviewed papers from this work. The excerpt below from the summary is quite clear. Moreover, if we compare to E85 to diesel cars, we also find that these harmful emission compounds are much, much higher for E85. This conclusion is valid at any temperature but, of course, more pronounced at low temperatures.

“…increased ethanol content in the ethanol/petrol fuel blends increases emissions of polycyclic aromatic hydrocarbons (PAH) at decreasing (ambient) temperatures, i.e. engine cold start emissions.”


Some points-

1. Running E85 within an engine optimised for gasoline not the best condition to minimise E85 emission nor maximise the fuel's efficiency. At that, the emissions of E85 are a mixed bag, with cold start higher and normal operation and acceleration much less. However, the emissions in general are less toxic ie. no carcinogenic benzene. BTW their is much science dedicated to the task to minimize the environmental concerns of petrol and some of that is set up to diss ethanol. It's accurate science, just how they set it up will result in phony conclusions. UAI has taken much of this "science" to task with independant lab results. A lot of money out their for scientist to prove wanted results. Science is not as objective and of pure virtue as the public is led to believe. There is a cottage industry afoot for R&D paper generation to keep the gov't free money flowing. Same for petrol marketing purposes as well.

2. E85 vs diesel? That can't be a good comparison. A diesel engineered and tweaked to minimize pollution with pollution control equipment doing the same. As compared to what? We have no E85 engine. Cummings per CA EPA did engineer a E85 test vehicle in which a diesel engine was utilized within spark ignition conversion. They found the engine optimized for E85 fuel beat mpg of a comparative gasoline engine option and best the torque of diesel 2x. Yes, the E85 engine only needed half the displacement to equal diesel engine torque. The project review noted a turbo cooler could have bumped up the E85 efficiency and that E100 had much higher efficiency than what would be normally expect. Oh, Calf. air emissions were easily met with the normal (cheap) catalytic auto system as opposed to the very expensive diesel system. The E85 being half the size would also be half the weight of the normal heavy diesel and much cheaper.

3. Also, were not really suggesting E85 a good solution for replacing gasoline GDI fuel. What automotive wants is high octane. As you are a reader of GCC, you already know a commision is at work as we speak to prequalify fuels for the task. Apparently, they are leaving no stone unturned and believe a bimix or trimix of complementary fuels the solution to high efficiency and low emissions. Alcohols look to play a major role in curing the problem. Most know the petrol's high octane adjuncts very nasty and unhealthy.


You should read some reports/papers from the study I mentioned before commenting. Of course, the study could only test available production vehicles but there is significant discussion about future development potential of E85. For sure, there has been a lot of research on ethanol fuel in advanced engines during the last years but basically nothing has been implemented on production vehicles. These are just “simple” conversions. Thus, for the moment, E85 and ethanol blends do not provide any real emission benefit compared to gasoline; or diesel if you prefer to use diesel as reference in the comparison. And, as seen in the Swiss study, diesel was much better than GDI.

I do not bother to comment in detail on your contribution but let me just mention that there is, in fact, benzene in the exhaust from E85. As said, read the report. You will find data on benzene there. Benzene was lower for E85 at high ambient temperature but there was only a small difference between fuels at low temperature. You should also note that many PAHs are much more carcinogenic than benzene and these were higher for E85. In fact, the mentioned study evaluated the total carcinogenic impact of all the compounds measured. E85 did not score well in this comparison. In contrast, the CNG/biogas car was much better than gasoline.


I don't buy it. I'm not going to read the 9 yr old study. Diesel is the problem and in a far 2rd place is DI gasoline. No up to date analysis of the problem is claiming E85 the problem. U.S. cities are thinking in terms of banning diesel from their streets. Chicago has already regulated the sale of E15 fuel to minimize pollution. Your info reminds me of the the "science" that claimed Brazil's pollution was a direct result of high ethanol blends. Later the analysis looked at a period of history of high annual ethanol sales and low annual sales. Yup harmful pollution went up with more petrol fuel and down with more ethanol within congested cities. The problem of ethanol has always been what base stock the fuel is blended to. Meaning you can chemically engineer a horrible base stock and that is exactly the test fuel utilized within EPA. They let petrol design the test fuel at a incredible price tag.


It is not who is right, it is how do we fix it. Ceramic filters on GDI seems like a small price to pay if it reduces heart and lung disease.

The comments to this entry are closed.