Nucleation Mode Particles with a Nonvolatile Core in the Exhaust of a

Aug 17, 2007 - The characteristics of the nucleation mode particles of a Euro IV heavy-duty diesel vehicle exhaust were studied. The NOx and PM emissi...
9 downloads 0 Views 219KB Size
Download PDF
Environ. Sci. Technol. 2007, 41, 6384-6389

Nucleation Mode Particles with a Nonvolatile Core in the Exhaust of a Heavy Duty Diesel Vehicle TOPI RO ¨ NKKO ¨ ,† ANNELE VIRTANEN,† J O N N A K A N N O S T O , † J O R M A K E S K I N E N , * ,† M A I J A L A P P I , ‡ A N D L I I S A P I R J O L A §,| Aerosol Physics Laboratory, Institute of Physics, Tampere University of Technology, P.O. Box 692, FI-33101 Tampere, Finland, Emission Control, VTT Technical Research Centre of Finland, P.O.Box 1000, FI-02044 VTT, Finland, Department of Technology, Helsinki Polytechnic, P.O. Box 4020, FI-00099 Helsinki, Finland, and Department of Physical Sciences, University of Helsinki, P.O. Box 64, FI-00014 Helsinki, Finland

The characteristics of the nucleation mode particles of a Euro IV heavy-duty diesel vehicle exhaust were studied. The NOx and PM emissions of the vehicle were controlled through the use of cooled EGR and high-pressure fuel injection techniques; no exhaust gas after-treatment was used. Particle measurements were performed in vehicle laboratory and on road. Nucleation mode dominated the particle number size distribution in all the tested driving conditions. According to the on-road measurements, the nucleation mode was already formed after 0.7 s residence time in the atmosphere and no significant changes were observed for longer residence times. The nucleation mode was insensitive to the fuel sulfur content, dilution air temperature, and relative humidity. An increase in the dilution ratio decreased the size of the nucleation mode particles. This behavior was observed to be linked to the total hydrocarbon concentration in the diluted sample. In volatility measurements, the nucleation mode particles were observed to have a nonvolatile core with volatile species condensed on it. The results indicate that the nucleation mode particles have a nonvolatile core formed before the dilution process. The core particles have grown because of the condensation of semivolatile material, mainly hydrocarbons, during the dilution.

Introduction Because of the hazards to health and the environment, the particle emissions of diesel vehicles have been limited by a consecutive series of more stringent regulations, such as the Euro standards. Compared to Euro III, the PM emission limit of the now effective Euro IV standard is significantly lower, and there are plans to limit the number of emitted soot particles. These changes require new development in fuels, lubricants, engine operation, and after-treatment devices. Both the engine design and after-treatment devices affect the particle size distribution of diesel exhaust (1) and physical characteristics of the exhaust particles (2). * Correspondence author phone: +358 3 3115 2676; fax: +358 3 3115 2600; e-mail: [email protected]. † Tampere University of Technology. ‡ VTT Technical Research Centre of Finland. § Helsinki Polytechnic. | University of Helsinki. 6384

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 18, 2007

The size distribution of diesel exhaust typically consists of two modes. In the field of vehicle exhaust studies it is customary to name them accumulation mode and nucleation mode (3). Accumulation mode particles consist of solid carbonaceous agglomerates with adsorbed and condensed semivolatile species. The accumulation mode in the vehicle exhaust is not sensitive to dilution conditions (4, 5). The nucleation mode consists mostly of liquid particles which are usually smaller than 50 nm in diameter. These liquid nanoparticles consist of water and semivolatile organic and sulfur compounds, and they are believed to form during dilution of the exhaust gas. Thus, their formation has been reported to be sensitive to dilution conditions such as dilution ratio, temperature and relative humidity of the dilution air (6), and the exact development of temperature and instantaneous partial pressures of vapors during dilution (7). Because of the sensitivity to sampling and dilution conditions, laboratory measurements of the nucleation mode particles may differ considerably from on-road emissions (4). On the other hand, in our previous study we reported that the trends in the formation of nucleation mode particles are similar in the laboratory and on-road (8). The exact formation mechanisms of nucleation mode particles are still unknown. Even chemical composition data of the smallest particles is missing, owing to the low mass concentration of the particles. In some previous studies the formation of nucleation mode particles has been linked to high sulfur or high hydrocarbon content in exhaust gas (1, 8-10). In these cases the particle formation depends on the driving parameters (8), fuel and lubricant oil properties (1013), and exhaust after-treatment systems (1, 13). During combustion, the sulfur of fuel and lubricant oil oxidizes mostly to sulfur dioxide. If the oxidation catalyst or catalyzed particle filter is used and the exhaust temperature is high enough, a significant fraction of sulfur dioxide can be converted to sulfur trioxide (14) which can react with water vapor molecules producing sulfuric acid (15). If the exhaust concentration of the sulfuric acid is high enough and the dilution conditions are suitable, the exhaust particle concentration can increase due to the homogeneous nucleation of water and sulfuric acid vapors (7, 16). Vaaraslahti et al. (13) reported that the volume of the nucleation mode particles depends on the available sulfur from the fuel and lubricating oil when a heavy duty diesel engine was equipped with an oxidation catalyst and a diesel particle filter. They also found that the nucleation mode was unstable as a function of time, because of a storage-release effect of sulfur compounds within the exhaust after-treatment system. Arnold et al. (15) measured the concentration of gaseous sulfuric acid in the exhaust of diesel passenger cars equipped with oxidation catalysts and particle filters. They observed that a higher sulfuric acid concentration correlates with a higher particle concentration in diluted exhaust gas. Maricq (17) reported that the nucleation mode particles are electrically neutral if the vehicle is equipped with a particle filter. When no after-treatment device is used, the hydrocarbons form the main component in nucleation mode if the fuel sulfur content (FSC) is at the level determined by the current legislation. Tobias et al. (18) found that hydrocarbon compounds from unburned fuel or oil formed most of the nanoparticle mass, whereas sulfuric acid was present at concentrations of a few percent. As a formation mechanism, they propose the nucleation of sulfuric acid and water followed by condensation growth by hydrocarbons. Sakurai et al. (19, 20) studied the volatility and composition of diesel nanoparticles and found that the nanoparticles comprise at 10.1021/es0705339 CCC: $37.00

 2007 American Chemical Society Published on Web 08/17/2007

TABLE 1. Driving Parameters of Test Conditions, Regulated Emissions, and Soluble Organic Fraction (SOF) of PM in Test Conditions condition number

speed (km/h)

engine speed (rpm)

torque (N·m)

CO (ppm)

THC (ppm)

NOx (ppm)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

20 20 30 30 40 40 40 40 40 50 50 60 60 70 70

1490 1490 1480 1480 1590 1590 1590 1590 1590 1540 1540 1850 1850 1750 1750

273 337 275 552 311 496 726 949 1059 325 876 416 917 451 1013

116 89 89 47 269 111 69 61 70 157 58 95 58 97 59

26 21 21 19 58 30 22 20 20 41 23 27 19 30 20

233 370 349 701 147 263 395 505 402 218 492 277 368 322 427

least 95% unburned lubricating oil. They also report that the particles may have a nonvolatile core consisting of soot, metal oxides, or low volatility organic compounds. In this paper we study the emissions and characteristics of nucleation mode particles in the exhaust of the modern Euro IV heavy-duty diesel vehicle. Exhaust gas recirculation (EGR) is used but no exhaust after-treatment. The measurements were performed mainly using a heavy-duty chassis dynamometer, but some particular tests were performed also on-road as chase measurements in order to ensure the results in real-world conditions.

Experimental Section Vehicle. The tested vehicle was a diesel truck equipped with exhaust gas recirculation system (EGR) (model year 2005, Euro IV emission level, displacement 11.7 dm3, mileage 5500 km, max. power 309 kW at 1900 rpm, max. torque 2100 N·m at 1100-1350 rpm). The emission standard is met by a new engine design and control, without the use of exhaust aftertreatment devices. The short exhaust pipe of the vehicle was positioned on the left side of the vehicle. Commercial diesel fuel with less than 10 ppm sulfur was used in the tests. Additional tests were also made with the same fuel but doped to a FSC of 48 ppm. A commercial lubricant oil (Shell 15W/ 40) was used, with measured sulfur content of 3370 ppm. The measurements consisted of 15 different constant speed driving conditions at low or medium load (Table 1). The driving parameters were recorded by tapping into the Controller Area Network of the tested vehicle. These parameters were used to achieve similar driving conditions in the laboratory and on-road. Measurements. The laboratory measurements of particle emissions were performed at the emission laboratory of VTT in Espoo on a heavy-duty chassis dynamometer. A porous tube type diluter (21) followed by an aging chamber was used to dilute the exhaust sample. After that, two ejector type diluters were used to dilute the sample to the measurement range of the particle instruments. The sampling system was a modified version of the partial flow sampling system (PFSS) developed in the "Particulates" research program of the EU (12). Similar diluters have been used in several vehicle or engine exhaust studies (1, 6, 8, 13, 22, 23). The values of the dilution ratios were calculated by using CO2 concentrations measured before and after the primary diluter. The temperature (T) and relative humidity (RH) of the dilution air were adjustable. Both were measured before the air entered the primary diluter. The used dilution air temperature and dilution ratio were 30 °C and 12 in primary dilution, respectively. In addition, driving conditions 5-8 were

PM (mg/kWh)

SOF (%)

884 129 51 23

94 85 82 80

measured using different dilution air parameters. After the secondary dilution by ejector diluters, the total dilution ratio was approximately 770 or higher. Particle size distributions were measured by the electrical low-pressure impactor (ELPI, Dekati, Inc.) (24) with filter stage (25), and two scanning mobility particle sizers (SMPS) (26). One SMPS was equipped with DMA 3085 and CPC 3025 (Nano-SMPS, TSI Inc.) and the other with DMA 3071 and CPC 3025 (SMPS, TSI Inc.). The measurement range of ELPI was 7 nm to 6.6 µm. NanoSMPS and SMPS had measurement ranges of 3 nm to 60 nm and 10 nm to 400 nm, respectively. A 90 s scan time was used in both instruments. At least three SMPS distribution measurements were made under each condition. Exhaust concentrations of gaseous compounds (THC, NOx, CO, and CO2) were measured by the Pierburg Ama 4000 I exhaust gas measuring system. Sampling conditions for PM were according to EU directive 1999/96/EC. The sample was taken from a full flow dilution tunnel of the HD CVS system. The SOF (soluble organic fraction) was extracted using the Soxhlet extraction method with dichloromethane (DCM). The SOF values were obtained as a weight difference between the original filter mass and the residue after the extraction. Particle volatility was studied by using a thermodenuder (Dekati, Inc.). In the thermodenuder, the diluted sample is led through a heater where the volatile compounds are evaporated. After the heater, the evaporated compounds are gradually cooled and absorbed into active charcoal. Because the total dilution ratio of the sample entering the thermodenuder is high, renucleation of vaporized compounds after thermodenuder treatment can be assumed to be negligible. However, the losses of nanoparticles are an issue and they were measured for 6-45 nm monodisperse silver particles for the temperature range 28-275 °C. The losses decreased with increasing particle diameter, ranging from 95% at 6 nm through 74% at 10 nm to 28-40% at 30 nm, depending on the temperature. All particle size distributions measured with the thermodenuder were corrected for the losses in the thermodenuder and in the Nano-DMA (27). The on-road tests were performed as chase measurements by following the vehicle with the “Sniffer” laboratory vehicle (28). The procedure and instrumentation were similar to those used in a previous study (8), except that the SMPS was now used in addition to Nano-SMPS and ELPI. The measurements were performed during 5 days with relative humidity between 38 and 85% and temperatures of 13-23 °C. All driving conditions were measured with a sampling distance 10 m between the test vehicle and the laboratory vehicle. Additionally, for driving conditions 5-8, sampling distances of 5, 15, and 20 m were used. The distance between the VOL. 41, NO. 18, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6385

FIGURE 1. Particle number size distributions measured with and without the thermodenuder in driving conditions 8 (a) and 1 (b) (torque 949 Nm and 273 Nm, respectively). Dynamometer measurement, primary dilution ratio (DR) 12, dilution air temperature 30 °C, combination of Nano-SMPS and SMPS data. Thermodenuder temperature 270 °C. outlet of the tailpipe and the rear of the vehicle was 3.2 m. Thus for 40 km/h vehicle speed the sampling distances 5, 10, 15, and 20 m corresponded to exhaust residence times 0.72.1 s in the atmosphere. At least five SMPS distribution measurements were made at each test. The background concentrations of particles and gases were measured between the tests. Dilution ratios were calculated from CO2 concentration measured on road and the raw exhaust concentrations measured in the laboratory for the same driving condition. All particle concentrations presented in the next chapter have been calculated to raw exhaust.

Results Gaseous and PM Emissions. In Table 1, the amount of gaseous regulated pollutants, PM, and particle soluble organic fraction (SOF) in the tested driving conditions are presented. The PM emissions and SOF were measured only for driving conditions 5-8. The presented values were measured with the undoped fuel. In the analyzed conditions, SOF constituted most of the particulate mass. The organic matter was either condensed on existing particles or participated in the nucleation process. It should be noted that in the driving conditions 5 and 6, the engine load is very low. Therefore, the measured PM-emission and SOF values are much higher than the values that would be obtained over any test cycle. Particle Size Distributions. In all driving conditions, the number distribution was dominated by the nucleation mode particles. Depending on dilution and driving conditions, the nucleation mode geometric mean diameter (GMD) was increased to 19-35 nm. Because of the high nucleation mode concentration, the size distribution looked unimodal until measurements with the thermodenuder showed that the size distribution consisted of the accumulation mode and the more volatile nucleation mode (see Figure 1). The relatively large particle size and high number concentration of the nucleation mode are in agreement with the high SOF values. The nucleation mode was stable in respect of time in all the tested driving conditions, indicating an absence of storagerelease effects of the semivolatile material. In on-road conditions the nucleation mode was observed to be formed already at 5 m behind the vehicle (corresponds to about 0.7 s residence time in the atmosphere). When particle concentrations were calculated to raw exhaust, no significant changes in particle size distribution were observed between 5 m and 20 m behind the test vehicle. The results are similar to those reported in our previous study (8), and they mean 6386

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 18, 2007

FIGURE 2. Exhaust particle number concentration with undoped (fuel sulfur content (FSC) < 10 ppm) and doped (FSC 48 ppm) fuels. that fast dynamical processes in the exhaust plume have taken place in time scales under 0.7 s. Further processes like coagulation and additional condensation/evaporation are too slow to cause significant changes between 0.7 and 2.1 s. Effect of Fuel Sulfur Content on Particle Emissions. The effect of sulfur was measured indirectly by changing the sulfur content of the fuel. Particle size distributions were measured with primary dilution conditions T ) 30 °C, low RH, and DR ) 12. As the number distribution was dominated by the nucleation mode, we only looked at the total particle number concentration measured with undoped and doped fuels, as shown in Figure 2. Differences in the total number between the undoped and doped fuel are small. There is no indication that sulfur addition to the fuel increased the particle number. Further, the GMD and the shape of the distribution were insensitive to the FSC. This indicates that the nucleation mode formation is not sulfur driven. Effect of Dilution Parameters on Particle Size Distribution. In driving conditions 5-8, different dilution parameters were used to clarify their effects on particle size distribution in the vehicle exhaust. The effect of primary dilution ratio was tested with dilution ratios of 12-30. The dilution air temperature was 30 °C and the relative humidity close to zero. In all the tested driving conditions, the increase in dilution ratio caused a decrease of the geometric mean diameter (GMD) of the nucleation mode, while the concentration remained unchanged. In Figures 3a and 3b the

FIGURE 3. The geometric mean diameter (GMD) of the nucleation mode (a) and the total particle number in respect of total hydrocarbon concentration (THC) in a diluted sample. Driving conditions 5-8 were used, and the temperature of the dry dilution air was 30 °C. GMD of the nucleation mode, and the number concentration of particles are presented as a function of total hydrocarbon concentration (THC) in the diluted sample. Different THC values represent different dilution ratios, as an increase of dilution ratio leads to lower concentration of hydrocarbons. Lower THC points show lower GMD of the nucleation mode, indicating decreased particle growth by condensation. On the other hand, changes in the dilution ratio did not significantly affect the particle number concentration (Figure 3b). In addition, the measured particle size distributions were insensitive to dilution air temperature and relative humidity changes between 20 °C and 82 °C and 0% and 60%, respectively. Volatility. In the laboratory, driving conditions 1-8, 10, and 14-15 were measured also with the thermodenuder. During these measurements, the temperature of the thermodenuder was adjusted to 270 °C. The temperature of the gases entering the denuder part, measured between the heater and the denuder part, was approximately 240 °C in all driving conditions. Figure 1 shows the particle size distributions measured with and without the thermodenuder in driving conditions 1 and 8. At the lowest loads, the distribution was unimodal even after the thermodenuder. We take this to mean that the number of core particles of the nucleation mode masks the less numerous soot particles (engine torque less than 340 N·m, see Figure 1b). Lower number concentration of soot particles at low engine load is typical and can be readily explained by the difference in the air-fuel ratio. Because of the cutoff size of the CPC (3 nm) and significant particle losses in the thermodenuder, it is impossible to tell the absolute number of solid cores. When the engine torque was more than 340 N·m, particle size distributions measured after the thermodenuder consisted of two modes with GMDs smaller than 9 nm and 37-47 nm, indicating that both the accumulation and the nucleation mode particles had a nonvolatile core (see Figure 1a). For driving condition 8, when a two-modal lognormal distribution was fitted on the data measured with the thermodenuder (Figure 1a), the number concentration of the nucleation mode particles was more than 70% of the concentration of nucleation mode particles measured without the thermodenuder. The results above (see Figure 1) show that a nonvolatile core existed in the nucleation mode particles when the sample was heated up to 240 °C. To study the particle volatility in more detail, the temperature of the thermodenuder was altered from room temperature up to 192 °C for driving condition 8. Figure 4 shows the change of the nucleation mode GMD as a function of the thermodenuder temperature. The GMD decreases with increased temperature up to 130 °C, where it plateaus to approximately 6 nm and shows no significant change at higher temperatures. On the road, the thermodenuder tests were made in driving conditions 6 and 7. Figure 5 shows the size distribution

FIGURE 4. The nucleation mode GMD for driving condition 8 as a function of thermodenuder temperature. DR 12, DT 30 °C, and RH close to 0%.

FIGURE 5. Size distributions measured on the road with and without thermodenuder (driving condition 6) and fitted distribution of nonvolatile core of nucleation mode particles (fitting procedure is explained in the text). The thermodenuder temperature was set to 270 °C. measured both with and without the thermodenuder in driving condition 6. The distributions are quite similar to those measured in the laboratory.

Discussion The PM emission characteristics of the present vehicle differ from the ones usually reported because of the high relative VOL. 41, NO. 18, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6387

SOF values (80-95%). Vaaraslahti et al. (13, 29) reported SOF values between 30 and 50% for low or medium load conditions when the measurements were made without exhaust after-treatment. It is instructive to compare also the absolute soluble organics emission values (as mg/kWh) to earlier technology. The SOF-% value of 80 (Table 1) for driving condition 8 corresponds to a soluble organics emission value of 18 mg/kWh. This medium load condition approximately corresponds to the mode 3 (50% load) of the ESC cycle. For this mode, Vaaraslahti et al. (2005) show soluble organics emission values of 10-15 mg/kWh for a rather low-emitting Euro II engine. At least for this load the emission of soluble organics is not extraordinarily high. It is just that the soot emission values are low. Compared to the soot emission value of approximately 30 mg/kWh shown by Vaaraslahti et al., the present technique has reduced the soot emission value by more than a factor of 6. This is achieved by the EGR and the advanced fuel injection techniques. The volatile fraction of the diesel particles has been attributed to lubricating oil (19). It seems conceivable that the advanced combustion technology does not change the behavior of the lubricant oil in the cylinder nearly as much as it changes the soot formation from fuel. Therefore, high SOF-% values are expected to be rather typical for this type of technology. The low soot concentration is accompanied by low GMD of the soot particles, ranging from 37 to 47 nm. Therefore, the condensation sink formed by the soot particles to the semivolatile species is low, and more hydrocarbon molecules are available for the growth of the nucleation mode particles. This explains the growth of the nucleation mode to mask the soot mode, as in Figure 1. The semivolatile hydrocarbon species do not readily explain the initiation of the nucleation. Usually, it is assumed that nucleation is sulfur driven. With the current low FSC values, this is only possible when oxidizing after-treatment devices are used. At high exhaust temperatures, normally met at high load, these after-treatment devices increase the concentration of sulfuric acid and cause the formation of the nucleation mode (1, 7, 8, 13, 15). The low FSC of the present study and the absence of any oxidizing devices mean that the present case cannot be explained by the binary water-sulfuric acid nucleation process (7). This view is supported by our finding that the nucleation mode is insensitive to changes in the FSC. We find nonvolatile core particles with a very small diameter in all the driving conditions studied with the thermodenuder. Earlier, Sakurai et al. (19, 20) reported that the diesel exhaust is externally mixed and particles can be divided to “less volatile” and “more volatile” fractions. In the “more volatile” particles with initial size 12, 30, and 70 nm, they observed nonvolatile species. However, they did not commit on the formation process of the nucleation mode particles or on the number fraction of the nucleation mode particles having the nonvolatile species. Is it possible that in our case all the nucleation mode particles are formed by condensation onto nonvolatile particles? Because of the losses of the smallest particles in the sampling and measurement system, it is hard to calculate the total concentration of the nonvolatile core particles. However, from the chase measurement results shown in Figure 5 it is a least possible to see the left slope of the distribution of the core particles. This makes it possible to fit a distribution to the measured values, also shown in Figure 5. It was fitted on the distribution of the nonvolatile core particles by varying the GSD and GMD of the distribution and keeping the number of the particles equal to the number of the nucleation mode particles measured without the thermodenuder. The number of the nucleation mode particles was calculated by subtracting the number of accumulation mode particles from the total particle number. 6388

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 18, 2007

The accumulation mode particle number was calculated by fitting two-modal lognormal distribution on size distribution measured with the thermodenuder, and the total particle number was calculated from the distribution measured without the thermodenuder. As we see, the fitted distribution matches well with the measured data. Thus it is plausible that all the nucleation mode particles have a nonvolatile core. Because of the small particle size, the core is probably not soot. We propose that in our case the nanoparticles grow onto solid nuclei originating from the engine. The nuclei could be composed of, for example, (oxidized) metals or pyrolyzed hydrocarbons. During cooling, the nonvolatile core of nucleation mode particles is covered by semivolatile compounds, such as hydrocarbons. These compounds evaporate in the thermodenuder, leaving the solid core to be detected. The high SOF level of the vehicle is believed to make this process so dominating. Nevertheless, it is possible that the same process explains also some previous observations of nucleation at conditions where the sulfur process seems unlikely. As an example, we have reported nucleation mode for an Euro II engine for low and medium load, without exhaust after-treatment, and using low sulfur fuel (1, 13). In addition, Kittelson et al. (30) observed nucleation mode for low FSC even at low load conditions. They only found nonvolatile core particles at idle. However, core particles smaller than 9 nm would not have been detected by the SMPS they used.

Acknowledgments This work was supported by Tekes, The Finnish Funding Agency for Technology and Innovation, and the Ministry of Transport and Communications, Finland.

Literature Cited (1) Vaaraslahti, K.; Virtanen, A.; Ristima¨ki, J.; Keskinen, J. Nucleation mode formation in heavy-duty diesel exhaust with and without a particulate filter. Environ. Sci. Technol. 2004, 38, 4884-4890. (2) Su, D. S.; Mu ¨ ller, J.-O.; Jentoft, R. E.; Rothe, D.; Jacob, E.; Schlo¨gl, R. Fullerene-like soot from EuroIV diesel engine: consequences for catalytic automotive pollution control. Topics Catal. 2004, 30/31, 241-245. (3) Kittelson, D. Engines and nanoparticles: a review. J. Aerosol Sci. 1998, 29, 575-588. (4) Kittelson, D. B.; Watts, W.; Johnson, J. Diesel aerosol sampling methodology; CRC E-43 Final Report; University of Minnesota: Minneapolis, 2002. (5) Maricq, M. M.; Chase, R. E.; Xu, N.; Laing, P. M. The effects of the catalytic converter and fuel level on motor vehicle particulate matter emissions: Light duty diesel vehicles. Environ. Sci. Technol. 2002, 36, 283-289. (6) Mathis, U.; Ristima¨ki, J.; Mohr, M.; Keskinen, J.; Ntziachristos, L.; Samaras, Z.; Mikkanen, P. Sampling conditions for the measurement of nucleation mode particles in the exhaust of a diesel vehicle. Aerosol Sci. Technol. 2004, 38, 1149-1160. (7) Lemmetty, M.; Pirjola, L.; Ma¨kela¨, J. M.; Ro¨nkko¨, T.; Keskinen, J. Computation of maximum rate of water-sulphuric acid nucleation in diesel exhaust. J. Aerosol Sci. 2006, 37, 15961604. (8) Ro¨nkko¨, T.; Virtanen, A.; Vaaraslahti, K.; Keskinen, J.; Pirjola, L.; Lappi, M. Effect of dilution conditions and driving parameters on nucleation mode particles in diesel exhaust: Laboratory and on-road study. Atmos. Environ. 2006, 40, 2893-2901. (9) Giechaskiel, B.; Ntziachristos, L.; Samaras, Z.; Scheer, V.; Casati, R.; Vogt, R. Formation potential of vehicle exhaust nucleation mode particles on-road and in the laboratory. Atmos. Environ. 2005, 39, 3191-3198. (10) Vogt, R.; Scheer, V.; Casati, R.; Benter, T. On-road measurements of particle emission in the exhaust plume of a diesel passenger car. Environ. Sci. Technol. 2003, 37, 4070-4076. (11) Khalek, I. A.; Spears, M.; Charmley, W. Particle size distribution from heavy-duty diesel engine: Steady-state and transient emission measurement using two dilution systems and two fuels. SAE Tech. Pap. Ser. 2003, 2003-01-0285. (12) Ntziachristos, L.; Giechaskiel, B.; Pistikopoulos, P.; Samaras, Z.; Mathis, U.; Mohr, M.; Ristima¨ki, J.; Keskinen, J.; Mikkanen, P.;

(13)

(14)

(15)

(16)

(17) (18)

(19)

(20)

(21)

Casati, R.; Scheer, V.; Vogt, R. Performance evaluation of a novel sampling and measurement system for exhaust particle characterization. SAE Tech. Pap. Ser. 2004, 2004-01-1439. Vaaraslahti, K.; Keskinen, J.; Giechaskiel, B.; Solla, A.; Murtonen, T.; Vesala, H. Effect of lubricant on the formation of heavy duty diesel exhaust nanoparticles. Environ. Sci. Technol. 2005, 39, 8497-8504. Giechaskiel, B.; Ntziachristos, L.; Samaras, Z.; Casati, R.; Scheer, V.; Vogt, R. Effect of speed and speed-transition on the formation of nucleation mode particles from a light duty diesel vehicle. SAE Tech. Pap. Ser. 2007, 2007-01-1110. Arnold, F.; Pirjola, L.; Aufmhoff, H.; Schuck, T.; La¨hde, T.; Ha¨meri, K. First gaseous sulphuric acid measurements in automobile exhaust: Implications for volatile nanoparticle formation. Atmos. Environ. 2006, 40, 7079-7105. Vehkama¨ki, H.; Kulmala, M.; Lehtinen, K. E. J.; Noppel, M. Modelling binary homogeneous nucleation of water-sulfuric acid vapours: parameterisation for high temperature emissions. Environ Sci. Technol. 2003, 37, 3392-3398. Maricq, M. M. On the electrical charge of motor vehicle exhaust particles. J. Aerosol Sci. 2006, 37, 858-874. Tobias, H.; Beving, D.; Ziemann, P.; Sakurai, H.; Zuk, M.; McMurry, P.; Zarling, D.; Waytulonis, R.; Kittelson, D. Chemical analysis of diesel engine nanoparticles using a Nano-DMA/ Thermal desorption particle beam mass spectrometer. Environ. Sci. Technol. 2001, 35, 2233-2243. Sakurai, H.; Tobias, H.; Park, K.; Zarling, D.; Docherty, K.; Kittelson, D.; McMurry, P.; Ziemann, P. On-line measurements of diesel nanoparticle composition and volatility. Atmos. Environ. 2003, 37, 1199-1210. Sakurai, H.; Park, K.; McMurry, P.; Zarling, D.; Kittelson, D.; Ziemann, P. Size-dependent mixing characteristics of volatile and nonvolatile components in diesel exhaust aerosols. Environ. Sci. Technol. 2003, 37, 5487-5495. Mikkanen, P.; Moisio, M.; Keskinen, J.; Ristima¨ki, J.; Marjama¨ki, M. Sampling method for particle measurements of vehicle

exhaust. SAE Tech. Pap. Ser. 2001, 2001-01-0219. (22) Virtanen, A.; Ristima¨ki, J.; Vaaraslahti, K.; Keskinen, J. Effect of engine load on diesel soot particles. Environ. Sci. Technol. 2004, 38, 2551-2556. (23) Ristima¨ki, J.; Maricq, M.; Keskinen, J.; Virtanen, A.; Maricq, M.; Aakko, P. Cold temperature PM emissions measurement: Method evaluation and application to light duty vehicles. Environ. Sci. Technol. 2005, 39, 9424-9430. (24) Keskinen, J.; Pietarinen, K.; Lehtima¨ki, M. Electrical low pressure impactor. J. Aerosol Sci. 1992, 23, 353-360. (25) Marjama¨ki, M.; Ntziachristos, L.; Virtanen, A.; Ristima¨ki, J.; Keskinen, J.; Moisio, M.; Palonen, M.; Lappi, M. Electrical filter stage for the ELPI. SAE Tech. Pap. Ser. 2002, 2002-01-0055. (26) Wang, S. C.; Flagan, R. C. Scanning electrical mobility spectrometer. Aerosol Sci. Technol. 1990, 13, 230-240. (27) Chen, D.-R.; Pui, D.; Hummes, D.; Fissan, H.; Quant, F.; Sem, G. Design and evaluation of a nanometer aerosol differential mobility analyzer (Nano-DMA). J. Aerosol Sci. 1998, 29, 497509. (28) Pirjola, L.; Parviainen, H.; Hussein, T.; Valli, A.; Ha¨meri, K.; Aalto, P.; Virtanen, A.; Keskinen, J.; Pakkanen, T. A.; Ma¨kela¨, T.; Hillamo, R. E. ”Sniffer” - a novel tool for chasing vehicles and measuring traffic pollutants. Atmos. Environ. 2004, 38, 3625-3635. (29) Vaaraslahti, K.; Ristima¨ki, J.; Virtanen, A.; Keskinen, J.; Giechaskiel, B.; Solla, A. Effect of oxidation catalysts on diesel soot particles. 4 Environmental Sci. Technol. 2006, 40, 4776-4781. (30) Kittelson, D. B.; Watts, W. F.; Johnson, J. P. On-road and laboratory evaluation of combustion aerosols-Part1: Summary of diesel engine results. J. Aerosol Sci. 2006, 37, 913-930.

Received for review March 2, 2007. Revised manuscript received June 15, 2007. Accepted July 5, 2007. ES0705339

VOL. 41, NO. 18, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6389