Nanoparticle Emissions from a Heavy-Duty Engine Running on

Environ. Sci. Technol. 43, 9501–9506

Nanoparticle Emissions from a Heavy-Duty Engine Running on Alternative Diesel Fuels ¨ , † A N N E L E V I R T A N E N , * ,† JUHA HEIKKILA ¨ NKKO ¨ ,† JORMA KESKINEN,† TOPI RO ¨ IVI AAKKO-SAKSA,‡ AND PA TIMO MURTONEN‡ Tampere University of Technology, Physics Department, Aerosol Physics Laboratory, P.O. Box 692, FI-33101 Tampere, Finland, and VTT Technical Research Centre of Finland, P.O. Box 1000, FI-02044 VTT, Finland

Received May 15, 2009. Revised manuscript received August 12, 2009. Accepted November 1, 2009.

We have studied the effect of three different fuels (fossil diesel fuel (EN590); rapeseed methyl ester (RME); and synthetic gas-to-liquid (GTL)) on heavy-duty diesel engine emissions. Our main focus was on nanoparticle emissions of the engine. Our results show that the particle emissions from a modern diesel engine run with EN590, GTL, or RME consisted of two partly nonvolatile modes that were clearly separated in particle size. The concentration and geometric mean diameter of nonvolatile nucleation mode cores measured with RME were substantially greater than with the other fuels. The soot particle concentration and soot particle size were lowest with RME. With EN590 and GTL, a similar engine load dependence of the nonvolatile nucleation mode particle size and concentration imply a similar formation mechanism of the particles. For RME, the nonvolatile core particle size was larger and the concentration dependence on engine load was clearly different from that of EN590 and GTL. This indicates that the formation mechanism of the core particles is different for RME. This can be explained by differences in the fuel characteristics.

Introduction The diesel exhaust submicrometer particle size distribution typically constitutes two particulate modes, named often as a soot mode (or an accumulation mode) and a nucleation mode (1). The mean size of the soot agglomerates is typically 40-80 nm and they consist of (spherical) primary soot particles with diameter from 5 to 20 nm. The nucleation mode particles consist of semivolatile organic and sulfur compounds (1). The compounds nucleate during the exhaust dilution processes as the gas exits the tailpipe, to form liquid nanoparticles typically smaller than 50 nm diameter. In addition, according to the results of Ro¨nkko¨ et al. (2) and Sakurai et al. (3) the nucleation mode particles emitted from a diesel vehicle which was not equipped with a particle filter included a nanosized nonvolatile core. The mean size of the core particles was less than 5-7 nm and they may grow up to 20 nm by condensation of hydrocarbons and sulphuric acid. * Corresrponding author tel: +358 3 3115 2676; fax: +358 3 3115 2600; e-mail: [email protected]. † Tampere University of Technology. ‡ VTT Technical Research Centre of Finland. 10.1021/es9013807

 2009 American Chemical Society

Published on Web 11/16/2009

Biofuels have been offered as one solution to decrease CO2 emissions and consequently the consumption of biofuels is increasing all the time. Neat vegetable oils or animal fats are not suitable for high-speed diesel engines, and thus a transesterification process is required to produce fatty acid methyl esters (FAME). When using rapeseed oil as feedstock, the product is rapeseed methyl ester, RME (4, 5). FAME has good ignition properties, low sulfur content, no aromatics, and good lubricity. However, FAME has many drawbacks when compared to diesel fuel. FAME may contain impurities, e.g., triglycerides, glycerol, alcohols, sodium, or potassium from the production process (5, 6). FAME biodiesel generally reduces CO, HC, and PM emissions, but increases NOx emissions (7-10). FAME reduces soot, but increases soluble organic fraction (SOF) of particulate matter, which is probably caused by distillation characteristics and high viscosity of FAME. In some cases, the increased SOF can lead to high PM emissions with FAME, especially at partial loads or at cold temperatures (11-14). It has also been reported that FAME diesel may increase nucleation mode particle concentration (13). However, these trends are not clear as the engine, load conditions, diesel fuel used for comparison, and properties of FAME affect the results. FAME biodiesel originates mainly from edible feedstock, and thus other options to replace diesel fuel are explored. Synthetic fuels offer an alternative route to convert a variety of feedstocks to liquid fuels. When the feedstock is natural gas, the product is GTL (gas-to-liquid). Conventional diesel fuel contains a number of hydrocarbons, aromatics, napthens, and paraffins, whereas GTL is paraffinic fuel with high cetane number, low sulfur, nitrogen, and aromatics content. According to Alleman et al. (15) excellent fuel properties of paraffinic synthetic fuels result in significant emission reductions and good engine performance. Generally, substantial reductions in, e.g., CO, HC, NOx, and PM emissions are observed. High cetane number decreases HC and CO emissions, while these fuel properties have little effect on NOx emissions (16, 17). In this paper we report the effect of two different fuels (RME and GTL) on diesel engine emissions compared to fossil diesel fuel (sulfur free EN590). Our main focus was on nanoparticle emissions of the engine. Nanoparticles may have increased ability to cause adverse health effects. It is well-known that the nanoparticles often no longer display the same reactivity as the bulk compounds. It has been suggested that the increased biological activity of nanoparticles is the consequence of their large surface area per unit mass (e.g., (18)). The organic (19) and metallic components (20) also appear to play a role in oxidative and proinflammatory effects and thereby affect pathogenicity (21). Thus the diesel nanoparticles may have significant health impact on humans and it is highly relevant to know the effect of fuels on nanoparticle emissions and characteristics. A number of papers report the existence of nanosize nonvolatile core particles in nucleation mode in the case of conventional fossil diesel fuel (2, 22-24). According to the authors’ knowledge, this is the first study focusing on nonvolatile nanoparticle emissions of alternative diesel fuels.

Experimental Section Engine, Fuels, Lubricants, and Driving Modes. Measurements were done at an emission laboratory of VTT in Espoo on an engine dynamometer. The engine used in measurements was a medium-duty diesel engine Cummins ISBe4 160B (model year 2006, displacement 4.5 dm3, common-rail fuel injection, mileage 250 h, max. power 118 kW@ 2500 VOL. 43, NO. 24, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

9501

TABLE 1. PM and Gaseous Emissions with Different Fuels fuel

HC (mg/kWh) CO (g/kWh) NOx (g/kWh) CO2 (g/kWh) PM (mg/kWh) N (total) (no./kWh) N (nonvolatile) (no./kWh)

25% load EN590 GTL RME change: GTL/EN590 change: RME/EN590

323.8 223.9 187.0 -30.9% -42.3%

5.96 5.17 6.07 -13.3% 2.0%

11.2 10.6 12.2 -5.7% 8.2%

814 779 830 -4.3% 1.9%

40.8 25.4 26.1 -37.8% -36.0%

6.1 × 1014 5.5 × 1014 2.0 × 1015 -9% 230%

3.7 × 1014 3.2 × 1014 1.6 × 1015 -14% 343%


80.6 59.5 27.5 -26.1% -65.8%

0.57 0.38 0.49 -33.6% -14.0%

12.1 11.4 12.3 -5.7% 1.7%

633 611 643 -3.5% 1.6%

11.8 9.0 10.4 -24.0% -12.1%

4.4 × 1014 3.5 × 1014 9.8 × 1014 -20% 121%

1.7 × 1014 1.8 × 1014 7.8 × 1014 8% 364%


56.4 47.0 20.7 -16.6% -63.3%

0.61 0.45 0.47 -26.9% -24.1%

9.3 9.0 9.4 -3.4% 0.3%

636 612 645 -3.8% 1.4%

15.9 9.7 9.1 -38.8% -43.0%

2.2 × 1014 2.2 × 1014 8.9 × 1014 -2% 296%

1.2 × 1014 1.4 × 1014 7.5 × 1014 19% 516%


45.6 39.3 18.1 -13.9% -60.3%

0.32 0.27 0.27 -17.9% -16.5%

8.2 7.7 9.1 -6.9% 10.3%

628 610 638 -2.9% 1.5%

23.0 12.5 10.6 -45.5% -53.8%

3.1 × 1014 2.1 × 1014 6.7 × 1014 -32% 115%

1.4 × 1014 1.3 × 1014 5.4 × 1014 -9% 280%

min-1, max. torque 600 Nm @ 1500 min-1). The engine meets Euro IV emission standards when a selective catalytic reduction (SCR) aftertreatment is used. In this study the engine was used without the SCR. The engine was tested in a steady-state test bench in an eddy-current dynamometer by Zo¨llner (Zo¨llner B-300 AD, 260 kW). Three different fuels were used as 100% blends. The fuels used were: regular sulfur-free (

Nanoparticle Emissions from a Heavy-Duty Engine Running on

Recommend Documents