a) b)
SYTTY
TRANSFORMATION AND ASSESSMENT OF EXPOSURE TO ORGANIC COMPOUNDS IN COMBUSTION DERIVED FINE PARTICLES
Project leader: Taisto Raunemaa, University of Kuopio, Department
of Environmental Sciences, P.O.Box 1627, FIN-70211 Kuopio, Finlanad, tel.
+358-17-163235, e-mail Taisto.Raunemaa@uku.fi
| PUBLICATIONS |
| TIIVISTELMÄ SUOMEKSI |
Researchers:
Ari Leskinen, University of Kuopio, Department of Environmental Sciences,
P.O. Box 1627 70211 Kuopio, tel. +358-17-163 238, e-mail: Ari.Leskinen@uku.fi
Timo Ålander, University of Kuopio, Department of Environmental
Sciences, P.O. Box 1627 70211 Kuopio, tel. +358-17-163 237, e-mail: Timo.Olander@uku.fi
Maija Lehtonen (worked as researcher in 1998-2000)
Financing SYTTY organisation: Tekes
Funding from SYTTY / Total funding of project (€): 193416/225607
Person-months of work funded by SYTTY / Total person-months of work:
57/68
KEY WORDS: combustion aerosol, fine particles, organic compounds,
transformation, exposure, lung model
EXTENDED ABSTRACT
1 Introduction
In studies during the past decades correlations between health effects and fine particles have been reported e.g. by Dockery et al. (1993) and Pope et al. (1995). Fine particles deposit especially into the alveolar region of the lungs, and they often contain harmful compounds, e.g. polyaromatic hydrocarbons (PAH) and other toxic semivolatile organic compounds (SVOC) causing adverse health effects.
In several recent studies the composition of combustion aerosols from wood burning and diesel engines have been examined. In only a few of them, however, attention has been paid to the effect of environmental conditions after stack or tailpipe emission. One important factor affected by ambient conditions and influencing the organic content of particles is the change in gas/particle partitioning of SVOC, which has been investigated in wood and diesel combustion aerosols e.g. by Strommen and Kamens (1997). Thus, it is important to know, how combustion aerosol is transformed after emission when estimating lung exposure to organics in aerosol particles.
In the present work, transformation experiments with combustion aerosols were performed in a large environmental chamber. The emphasis was on the changes in the particle phase, as gaseous emissions are controlled by emission standards. The size distribution and other changes of aerosol in ageing were considered relevant to observe instead of bulk mass emission. An important step is gas to particle (gtp) conversion, which requires systematic investigations in smog chambers. The data from the environmental chamber experiments were applied to estimate the lung exposure to fine particles. This task was carried out by calculating the number and mass deposition of aerosols in the human respiratory tract using the LUDEP lung model.
2 Methods
Aerosol generation and experiments in environmental chamber
Spruce chips (water content 20 %) were combusted in a stoker burner
connected to a small-scale boiler (power output 20 kW). Flue gas excess
oxygen was 13 %. Diesel exhaust aerosol was produced by a diesel powered
vehicle (Toyota, 76 kW) in neutral at 1500 rpm and in later experiements
by a non-road engine (Perkins, 12 kW) at idle or with 9 kW load. When burner
and engine operation stabilized, the combustion aerosol was injected into
a 143 m3 Teflon environmental chamber situated on the top of the laboratory
building. The aerosol was monitored for at least 24 hours. Between experiments
the chamber was flushed with particle free air (volumetric flow 1000 m3/h).
NO and NOx concentrations in the chamber were monitored by a chemiluminescent analyzer and O3 by an UV-absorption analyzer. Air temperature, relative humidity, total solar radiation (TSR) and total ultraviolet radiation (TUVR) in the chamber were registered continuously.
Particle size was measured every 5 minutes by scanning mobility particle sizers (SMPS) consisting of a differential mobility analyzer (DMA) and a condensation particle counter (CPC): the size range 15-750 nm with a TSI 3071A+3022A and 3-90 nm with a TSI 3085+3025 system. The amount of volatile compounds on particles was determined with a thermodenuder system. In this method a sample flow is heated in an oven up to 360 °C and cooled in an absorbing environment. The change in particle size produces information on the volatile fraction of particles. Submicron particles (PM1) were collected on Teflon and quartz fibre filters using parallel sampling, which technique enables correction for gas adsorption on quartz filters. Filter samples were analyzed by a thermal-optical method (Ålander and Raunemaa, 1997), which produces the OC/EC ratio in the sample particles.
The amount of photo-active compounds on particles was measured using a photoelectric aerosol sensor (PAS) system. In PAS particles are illuminated with energetic UV light to ionize PAH molecules on the particle surface. The suitability of the method was reviewed in the project (Scotto di Marco et al., 2001), but only a few experiments were carried out. The sensor was deduced to monitor only relative changes of PAHs, for absolute concentrations the system should be calibrated.
Smog chamber experiments with pure organic compounds
To gain information over large reaction range on the ozone and particle
formation potential of hydrocarbon compounds, a mixture of VOC (acetone,
methyl ethyl ketone, toluene, xylene or alpha-pinene) and NO2 was exposed
to UV light in a collapsible 6 m3 Teflon smog chamber. The UVA lamps (Blacklight
350) can be switched on to produce UV irradiance from zero to 48 W/m2,
which covers the UV irradiance variations on a sunny day outdoors.
Particle formation properties of the chamber material and laboratory air had to be investigate first, and the experiments were done in a 0.3 m3 Teflon chamber using the same UV light system. Only VOCs present in the laboratory air or in the compressed air supply were found to be significant. To prevent their interference, a chemical filter system was installed in the chamber flush air.
In the experiments, hydrocarbons from liquid sources (purity 99.5-99.8 %, for alpha-pinene 97 %) were introduced into the chamber through an evaporation vessel. Hydrocarbon concentrations of 0.5-1.0 ppm and NOx concentrations of 0.1-0.3 ppm were mixed in the chamber air. Hydrocarbons and their photochemical products were analyzed by gas chromatography every 20 minutes.
Lung exposure estimates
The lung exposure was estimated by calculating particle deposition
into Extrathorasic 1 (ET1), Extrathorasic 2 (ET2), Bronchial (BB), Bronchiolar
(bb) and Alveolar-Interstitional (AI) regions by the LUDEP model (ICRP,
1994). The deposition calculations were carried out for an adult male in
light work applying the experimental results for 3-750 nm particles and
assuming particle density of 1.7 g/cm3. For simplicity spherical particle
shape was assumed, even if diesel particles are known to be chain aggregates.
The exposure to organic compounds in combustion derived fine particles was based primarily on particle size and number concentration. The ratio (OC/TC) was proposed to indicate the organic species content in the particles. As particle size determines the deposition in different parts of respiratory tract, number concentration gives the total particulate burden and OC the amount of harmful compounds. When the aim of the work was to study bulk changes in particles, no species specific organics analysis was performed.
3 Results and discussion
The results in 2000-2001 are discussed. Earlier work (1998-1999) has been presented by Raunemaa et al. (1999), Leskinen et al. (2000a) and Leskinen et al. (2000b).
The mean particle size of fresh wood combustion aerosol and diesel exhaust
aerosol was around 100 nm. Approximately half of the particles at this
size deposit in the respiratory tract (Figure 1), mostly in the alveolar
region (30-35 %). As the initial undiluted concentration of particles was
around 2.5x105 #/cm3, this would amount an alveolar deposition
of around 8x1010 particles or 100 µg per inhaled m3 of
air.
a)
b)
c)
Figure 1: The organic (OC) and elemental (EC) carbon fraction of the total carbon (TC) and particle deposition fractions to different parts of the respiratory tract (abbreviations in the text) in fresh and 24 h aged (a) wood combustion, (b) diesel exhaust (no load), (c) diesel exhaust (9 kW load) aerosol.
The volatility of wood combustion aerosol is different from that of diesel exhaust aerosol. In fresh wood combustion aerosol 18 % of particle volume was found volatile at 360 °C, whereas for diesel exhaust aerosol the volatility was 36 % for idle operation and 17 % with 9 kW load. The OC/TC ratio was 67 % in wood combustion particles, 72 % in particles from diesel engine without load and 58 % with 9 kW load. The OC/PM1 ratio was 5-14 %, and people inhaling fresh flue gas from wood combustion or diesel exhaust are exposed to around 10 µg/m3 of organic particulate matter in the alveolar region, assuming the concentration values in the chamber experiments.
When combustion particles age for 24 hours, their size increases to around 200 nm, i.e. 2.5-fold, and their number and surface area decreases due to coagulation and removal processes. This leads to lower respiratory deposition; accordingly only 20 % of the particles were estimated to deposit in the alveolar region.
When total surface area of particles falls below the concentration 109 nm2/cm3, formation of new ultrafine particles is observed, with prerequisite that enough solar radiation is available. Ultrafine particles could be observed in the experiments only after growth to 15 nm size. Their average number concentration was in the order of 103 #/cm3 which corresponds a typical unpolluted air. The formation of ultrafine particles increased the alveolar deposition from 20% to 30-40 %, while the aerosol mass deposition in alveolar region remained at 15-20 %.
The new particles appeared to be more volatile than primary particles, which decreased in volume by more than 90 %. When heating to 140 °C their volume decreased by 55 % in wood and by 92 % in diesel aerosol, and at 250 °C particles evaporated completely. The OC/TC ratio increased simultaneously to 100 % from initial 67 % in wood and from 72 % in diesel exhaust (no load) aerosol. Despite no new particle formation was seen in diesel exhaust with 9 kW load, the OC/TC ratio increased to 84 % from 58 % indicating condensation of organic compounds on existing particles. Alveolar mass deposition was estimated to be 1-2 µg per inhaled m3 of air and OC/PM1 ratio 18-24 %. This amounts an alveolar exposure to organics of around 0.3 µg/m3.
When solar irradiance was low, e.g. due to cloud coverage, no new particle formation was observed even if the surface area was below the proposed limit. This is assumed to be connected with deceleration in particle formation process at low UV exposure. The effect was examined in systematic way in the ozone and particle formation experiments in the smog chamber. Using the terpene compound, alpha-pinene, an increase in UV irradiance from low (8 W/m2) to high (48 W/m2) resulted in threefold ozone concentration, known to accelerate new particle formation. Ozone forming potential of organics increases also with the molecular weight of the organic compound.
4 Conclusions
The physical properties (e.g. size and concentration) and organic content of combustion derived fine particles were shown to vary during aerosol ageing, which can be interpreted as different exposure to organic compounds in fresh and aged aerosol. The changes are different for aerosols from wood combustion or diesel engine. Even if the emissions were equal, their particles age differently depending on environmental conditions. New particles were shown to enhance alveolar deposition and their organic content is high. This suggests an increased exposure to organics when new particle formation occurs. Particle formation potential was seen to be linked also to molecular weight of hydrocarbons, which means that more persistent organic compounds might be present in new particles.
The results indicate that knowledge of the combustion aerosol composition is essential when estimating organic species burden in the lungs. Especially particle surface properties are important, as the surface area was shown to be related to new particle formation potential. Particle size distribution instead of bulk mass concentration was seen important when estimating the lung exposure. Mass deposition could not discern the changes in alveolar deposition.
5 References
Dockery, D.W., Pope, C.A., Xu, X., Spengler, J.D., Ware, J.H., Fay, M.E., Ferris, B.G. Jr. and Speizer, F.E. (1993) An Association between Air Pollution and Mortality in Six U.S. Cities. N. Engl. J. Med. 329(24), 1753-1759.
International Commission on Radiological Protection (ICRP) (1994) Publication 66: Human Respiratory Tract Model for Radiological Protection. Edited by ICRP. Annals of the ICRP, 24/1-3, 482 pages.
Pope, C.A., Thun, M.J., Namboodiri, M.M., Dockery, D.W., Evans, J.S., Speizer, F.E. and Heath, C.W. (1995) Particulate Air Pollution as a Predictor of Mortality in a Prospective Study of U.S. Adults. Am. J. Respir. Crit. Care Med. 151, 669-674.
Leskinen, A., Lehtonen, M., Ålander, T. and Raunemaa, T. (2000a) Transformation and Assessment of Exposure to Organic Compounds in Combustion Derived Fine Particles. In Proceedings of the Mid-Term Symposium of the Finnish Research Programme on Environmental Health (Juuti, S. and Leinonen, H., Eds.), pp. 172-176.
Leskinen, A., Yli-Tuomi, T., Ålander, T. and Raunemaa, T. (2000b) Lung Exposure Estimates in Transformation of Diesel Engine Exhaust. In Proceedings of the 2000 European Aerosol Conference (Dublin, 4.-8.9.2000), J. Aerosol Sci. 31, S506-S507.
Raunemaa, T., Yli-Tuomi, T., Leskinen, A., Tissari, J. and Ålander, T. (1999) Wood Combustion Aerosol and Lung Exposure Estimates. In Proceedings of the 1999 European Aerosol Conference (Prague, 6.-10.9.1999), J. Aerosol Sci. 30, S735-S736.
Scotto di Marco, G., Leskinen, A. and Raunemaa, T. (2001) Aerosol Photoemission (APE): Literature Review. The Publication Series of the Departments of Environmental Sciences, University of Kuopio, 2/2000, 22 pages.
Strommen, M. R. and Kamens, R. M. (1997) Development and Application of a Dual-Impedance Radial Diffusion Model to Simulate the Partitioning of Semivolatile Organic Compounds in Combustion Aerosols. Environ. Sci. Technol. 31, 2983-2990.
Ålander, T. and Raunemaa, T. (1997) Thermal-Optical Method in
Automotive Particle Emission Carbon Analysis. J. Aerosol Sci. 28, S543-S544.