Study finds engines emit exhaust nanoparticles even when not fueled during engine braking
19 January 2014
A new study finds that as much as 20–30% of the number of vehicle engine exhaust particles larger than 3 nm may be formed during engine braking conditions—i.e., during decelerations and downhill driving while the engine is not fueled. However, the authors note, these particles have not been taken into account in emission regulations and in the assessment of associated health risks.
The study by researchers in Finland and Greece, published in the ACS journal Environmental Science & Technology, suggests that both the characteristics of these particles and the mechanism by which they form seem to differ significantly from those of soot and nucleation particles. The study also indicates that the particles were non-volatile, formed before the catalyst, and originating from engine oil. Results thus indicate that the emissions of engine braking particles can be reduced using exhaust particle filtration systems.
… changing driving conditions also affect particle emissions. In emissions legislation, this has been taken into account by including transient driving cycles in emission standards. As a result, the number of exhaust studies focusing on emissions at transient conditions has increased. However, specific data for detailed driving conditions have remained limited because the regulated quantity is the particulate mass or number integrated over the test cycle.
In general, it has been observed that during transient cycles, the particle size distributions have two modes, and the highest particle emissions occur under heavy acceleration. Transient driving conditions can affect both the nucleation mode particles and the soot particles. Increased emissions of nucleation particles, affected by fuel sulfur content and exhaust dilution and thus proposed to be formed by sulfuric compounds, has been observed at deceleration as well. As at steady state tests, the particle emissions during transient cycles has been observed to be affected by vehicle technologies like exhaust after-treatment and fuel type.
This work focuses on the observation that engines emit a high number of exhaust particles at engine braking conditions—that is, during decelerations and downhill driving. In practice, “engine braking” refers to the driver’s lifting of the gas (accelerator) pedal and keeping the gear on, and it is typically used to enhance fuel economy or simply to limit or decrease a vehicle’s speed during downhill driving or deceleration. With direct injection engines, the fuel consumption during engine braking is zero.
—Rönkkö et al.
The clear benefit of engine braking is its reduction of fuel consumption; for the engines with direct fuel injection technology, fuel consumption during engine braking is zero. The use of engine braking also decreases brake wear.
The inertia of the vehicle maintains the rotation of a crankshaft during engine braking, and thus, the movement of pistons—without fuel combustion. Intake air still goes to the cylinder, where it is compressed to elevated pressures, its temperature raises, and it is exhausted to the atmosphere.
In the study, the team from Tampere University of Technology, Metropolia University of Applied Sciences, and the Finnish Meteorological Institute (all in Finland) and Aristotle University Thessaloniki in Greece, report on three experiments: one with a heavy-duty diesel truck at on-road conditions, and two conducted with gasoline direct injection (GDI) passenger cars on the chassis dynamometer.
Among the findings were:
On-road exhaust particle studies with the heavy-duty diesel truck showed that—with only few exceptions—the highest concentration peaks (frequently exceeding the maximum measuring range of the particle counter after passive dilution) were observed during downhill driving. The peaks were repetitive, appearing throughout the downhill sections of the driving sequence.
The results suggested that the high particle concentrations during downhill driving (engine braking) were dominated by particles sized below 7 nm.
Engine braking particles were also observed in the experiments with modern GDI vehicles when following the European statutory driving cycle. Repetitively high concentrations of particles were observed during engine braking in decelerations, when the sampling was conducted directly downstream of the exhaust manifold.
The mean diameter of these particles varied from less than 10 nm to slightly larger than 20 nm, depending on the deceleration pattern. When sampled downstream of the complete exhaust line including the three-way catalyst, the concentration of particles was lower, and the mean particle size larger, probably due to the strongly size-dependent diffusional particle losses, the researchers suggested.
Analysis indicated that engine braking particles, emitted during transient driving cycles, were associated with engine oil.
The particles’ small size and non-volatility, coupled with the observation that these particles contain lube-oil-derived metals zinc, phosphorus, and calcium, are suggestive of health risks at least similar to those of exhaust particles observed before.
Because of their origin, in principle, no correlation between the emissions of these engine braking particles and CO2 is expected. It is therefore possible that emission factors, typically determined by measuring particles and carbon dioxide concentrations simultaneously in an urban environment and using these quantities to calculate particles per unit of fuel combusted, are at least partly failed. This is, however, dependent for example on the location of urban air measurement station in relation to traffic routes. Also, our chassis dynamometer studies indicate that the high particle concentrations that are in reality caused by deceleration and engine braking are in some studies erroneously associated with idling.
Our study indicates that these [nano]particles are emitted in short time bursts,causing extremely high and instantaneous increases of particle concentrations. In addition to downhill driving, these particles may be emitted while navigating locales with a frequent presence of people, such as when decelerating at crossroads, traffic lights with pedestrian crossings, or bus stops. Particularly in urban environments, this can lead to significant human exposure of these particle emissions.
…our study indicates that the particles were non-volatile, formed before the catalyst, and originating from engine oil. Thus, the results indicate that the emissions of engine braking particles can be reduced using exhaust particle filtration systems. Additionally, another potential but less direct way to reduce the emissions of these particles may be the modification of engine oil characteristics, such as metal content or viscosity. From an emissions control perspective, the problem remains for those older vehicle types that are not equipped with particle filters and possibly for new vehicle technologies for which particle filters have not been mandated, such as gasoline direct injection (GDI) vehicles.
—Rönkkö et al.
Resources
Topi Rönkkö, Liisa Pirjola, Leonidas Ntziachristos, Juha Heikkilä, Panu Karjalainen, Risto Hillamo, and Jorma Keskinen (2014) “Vehicle Engines Produce Exhaust Nanoparticles Even When Not Fueled,” Environmental Science & Technology doi: 10.1021/es405687m
Wonder how Toyota's (and other HEVs and PHEVs) fare during frequent engine stops & starts?
Posted by: HarveyD | 19 January 2014 at 10:01 AM
We now know that all combustion engines require high effiency particulate filters, not just the diesels (with their unrigorous standards). Public health demands the rapid implementation of a new standard to eliminate this attack upon us all.
Posted by: JC | 19 January 2014 at 11:17 AM
A good reason to use maximum front wheel regen braking as primary strategy. More charge recoup and reduced emission risk.
Posted by: Arnold | 19 January 2014 at 04:34 PM
This is good enough justifications to progressively phase out all ICEs, CPPs, wood burning stoves and other burning machines?.
Posted by: HarveyD | 19 January 2014 at 05:20 PM
Good point, Harvey. Toyota's HSD is designed such that no engine braking is used in normal driving. When you lift your foot off the gas pedal of a HSD-equipped car (HEV), the engine will iddle for a short time and will be shut off and be stopped completely.
The Prius' engines are so good, such that no oil consumption is observed between oil change interval of 6000 miles for the Prius Gen II and 10,000 miles for the Gen III. My 100,000-mi Prius Gen II still shows zero oil consumption after 6,000 miles between oil changes. So, in good engines w/out oil consumption, the point of this article is moot.
PHEV's will soon change all this. The Audi PHEV with 8.8 kWh battery is capable of 31 e miles. If charged twice daily, the vehicle can do 62 miles without fuel consumption. The BYD PHEV can do 44 e miles, and if charged twice daily, can do 88 e miles daily, and can be charged 4,000 times before battery needing replacement! Even the Ford Fusion PHEV can do 21 e miles. If charged twice daily, it can do 42 e miles, and this is good enough for most people, while utilizing relative small battery pack of 8 kWh that allows a fixed amount of Lithium to be spread over 10x many cars than a single BEV with 85-kWh battery pack.
So, federal law should mandate standard 120-V charging sockets at most businesses and public buildings, for public good in combating pollution. THis is similar to the law banning smoking in most public places.
Posted by: Roger Pham | 19 January 2014 at 05:37 PM
So, what can we do about it? Crush all the ICE cars? Not likely. It's another one of those defined risks, we know how to fix; but, the politics and special interest economics prevent us from fixing. It's like coal pollution, fracking and the Tea Party.
Posted by: Lad | 19 January 2014 at 07:03 PM
I am not surprised. It is known that sudden decreases in load can generate “bursts” of increased oil consumption. When this oil is burned, we get solid particles, i.e. mostly oil ash. These particles are in the nano range and are mostly not taken into account with the 23 nm cut-point in current regulations. There are two solutions for ICE vehicles:
1. Diesel with DPF. This is already implemented on most vehicle/engines. DPFs do not really get enough credit for that they decrease PM/PN emissions during practically all driving conditions, i.e. also during “off-cycle” conditions. It has been shown earlier (e.g. within the ACES project reported on this site) that DPFs effectively filtrate ash from engine exhaust. In fact, DPFs have to be sized and designed to take ash accumulation into account.
2. Gasoline engines with GPF. There is sufficient experimental data (some studies have been posted on this site) to show that this solution works. The cost is not prohibitive. “Work-around” solutions, such as double injection systems (DI & IDI), do not help to reduce engine brake particles. Now, the legislators must take the latest data into consideration and change future emission legislation correspondingly.
Re oil consumption: The amount of oil consumed in the driving condition cited in the paper is small and is not visible on the oil stick. In fact, oil dilution by diesel and gasoline might even increase the level in the oil pan on some engines, yet allow enough oil consumption to form the particles discussed in the paper. Likewise, it should be noted that the weight of <10 nm particles does not contribute in a measurable way to the total particle mass. It is somewhat amusing to find that some engines are, in fact, credited for oil dilution with fuel. Oil must be changed although the level shown by the oil stick does not imply this. If the oil is not changed, oil dilution will lead to excessive engine wear.
Re HEVs/PHEVs: Start/stop instead of idling and “sailing” (with engine off) during decelerations helps but the real solution is DPF/GPF.
Posted by: Peter_XX | 20 January 2014 at 01:57 AM
I am not surprised. It is known that sudden decreases in load can generate “bursts” of increased oil consumption. When this oil is burned, we get solid particles, i.e. mostly oil ash. These particles are in the nano range and are mostly not taken into account with the 23 nm cut-point in current regulations. There are two solutions for ICE vehicles:
1. Diesel with DPF. This is already implemented on most vehicle/engines. DPFs do not really get enough credit for that they decrease PM/PN emissions during practically all driving conditions, i.e. also during “off-cycle” conditions. It has been shown earlier (e.g. within the ACES project reported on this site) that DPFs effectively filtrate ash from engine exhaust. In fact, DPFs have to be sized and designed to take ash accumulation into account.
2. Gasoline engines with GPF. There is sufficient experimental data (some studies have been posted on this site) to show that this solution works. The cost is not prohibitive. “Work-around” solutions, such as double injection systems (DI & IDI), do not help to reduce engine brake particles. Now, the legislators must take the latest data into consideration and change future emission legislation correspondingly.
Re oil consumption: The amount of oil consumed in the driving condition cited in the paper is small and is not visible on the oil stick. In fact, oil dilution by diesel and gasoline might even increase the level in the oil pan on some engines, yet allow enough oil consumption to form the particles discussed in the paper. Likewise, it should be noted that the weight of <10 nm particles does not contribute in a measurable way to the total particle mass. It is somewhat amusing to find that some engines are, in fact, credited for oil dilution with fuel. Oil must be changed although the level shown by the oil stick does not imply this. If the oil is not changed, oil dilution will lead to excessive engine wear.
Re HEVs/PHEVs: Start/stop instead of idling and “sailing” (with engine off) during decelerations helps but the real solution is DPF/GPF.
Posted by: Peter_XX | 20 January 2014 at 02:05 AM
We know that this condition relates to restricted air intakes and specifically in throttled. Gasoline direct injection is one solution to this.
Many performance engines seek greater 'tolerances to increase cylinder lubrication and to exhaust valves.
Modern state of the art designs use low loss pairing of cylinder finish matrix in combination with optimised piston and ring designs and ashless low viscosity high end oils.
As always primary targeting or prevention does not exclude secondary post treatment. Rather applying state of the art methods is often more cost effective.
Posted by: Arnold | 21 January 2014 at 01:49 PM
@Arnold
Gasoline direct injection is no solution to this. The phenomenon is due to changing pressures in the cylinder that occasionally lift the piston rings, and thus increase oil loss. This happens also in diesel engines, although I do not have data at hand to quantify how much in comparison to throttled otto engines. You get changes in cylinder pressure also in engines with less (or no) throttling, such as GDI engines. In fact, most of new GDI engines run stoichiometric and are no different than conventional MPI engines in this respect. You should know by now that GDI significantly increase PM and PN emissions compared to conventional MPI. In my opinion, all GDI engines should have GPF to be clean under any operating condition. Since a GPF would also reduce nanoparticles from oil combustion, this would be an additional benefit.
Posted by: Peter_XX | 22 January 2014 at 12:34 AM
Observation of oil burning and visible smoke confirms my statement.
Downhill throttle closed,followed by visible smoke and blocked (restricted)air filters as a cause of excessive oil consumption are the relevant diagnostic tests.
Restricted intakes reduce (increase negative) pressure in combustion side and encourage oil to bypass piston assembly.
In so far as GDI unthrottled engines (I have no knowledge of other production T'less gasoline) and attention to cylinder piston factors have major roles in the topic discussed, and I didn't suggest G/D/PF can't work, but they (Everything)can fail for numerous reasons.
The best solution IMO is e evolution towards decarbonising the global economy not one more reactive bandaiding "solution.
Posted by: Arnold | 22 January 2014 at 03:35 PM