J. Chil. Chem. Soc., 48, N 2 (2003)

José Celis¹, Roberto Morales², Claudio Zaror³, Juan Inzunza⁴, Roberto Flocchini⁵, Omar Carvacho⁵

¹Facultad de Ingeniería Agrícola, Universidad de Concepción, Casilla 537, Chillán, Chile.
²Facultad de Ciencias, Universidad de Chile, Casilla 653, Santiago, Chile.
³Facultad de Ingeniería, Universidad de Concepción, Casilla 160-C, Concepción, Chile.
⁴Facultad de Ciencias Físicas y Matemáticas, Universidad de Concepción, Casilla 160-C, Concepción, Chile.
⁵Crocker Nuclear Laboratory, Universidad de California, Davis, CA 95616, USA.
(Received: December 4, 2002 - Accepted: March 5, 2003)

Inhalable particulate matter (PM₁₀) measurements were performed in six different sites in the city of Chillán, Chile, during September 2001 to September 2002. Chemical composition of PM₁₀ was performed to samples of 47 mm diameter Teflon membranes within the city of Chillán. The spatial and temporal variability of the chemical composition of PM₁₀ was evaluated taking into account additional data from meteorology and further air pollutants. The chemical analyses of PM₁₀ showed that carbonaceous substances and crustal material were the most abundant components of PM₁₀ during the winter and summer, respectively. The concentrations of PM₁₀ were higher during the cold season than during the warm season. This was explained mainly due to the massive use of wood as fuel for residential heating within the city of Chillán, producing a dense smoke cloud in those days of atmospheric stability. The PM₁₀ concentrations were higher in the downtown area of the city of Chillán, where also the chemical composition was more variable due to urban traffic and other anthropogenic sources.

Key Words: PM₁₀, aerosols, urban atmosphere, particulate matter, air pollution.

* Corresponding author: Tel.: 56-42-208808; fax: 56-042-275303; e-mail: jcelis@udec.cl

INTRODUCTION

Several epidemiological studies have demonstrated a direct association between atmospheric inhalable particulate matter (PM₁₀) and people's health. Exposure to increased levels of PM₁₀ shows a high correlation with the increase of the respiratory diseases, pulmonary damage, and mortality among population ^1,2,3). In addition the studies have indicated more consistent effects for high concentrations of the fine and inhalable particles with heath effects than for others atmospheric pollutants ^{4, 5)}.

From the point of view of the origin, the PM₁₀ can be classified in primary and secondary ⁶⁾. The primary particles are released into the atmosphere directly, whereas secondary particles are formed within the atmosphere by gas to particle conversion from gaseous precursor substances such as the volatile SO₂, NO₂, NH₃, and organic compounds ⁷⁾. Aerosol particles are produced by natural and human activities. Typical natural sources are those that come from the sea and give origin to saline particles or wind-blown dust. Man generates particulate matter as a result of industrial activities, traffic and combustion processes ⁸⁾.

The chemical composition of particulate matter may vary within a broad range according to the sources of the particles and the conditions of their dispersion ^{9,10, 11)}. In this sense, a similarity in studies made in cities of industrialized countries has been observed ^1,2,3,4,5), whereas in cities of the developing countries the variability is enormous due to large polluting loads and dust caused by local winds. In Chile, studies concerning particulate matter have been focused in the city of Santiago ^{12, 13, 14)}, and few investigations have been performed in cities along the territory with a population ranging between 50,000 and 250,000. Nevertheless, these cities are considered to be growing more rapidly in Latin America, and those that present major expectations having a sustainable development (sustainable economic growth, social fairness and environmental protection). It is necessary to avoid that these cities reproduce the environmental problems of the metropolitan areas, so is important to evaluate quickly and strategically its environmental potentialities and limitations. The objective of this research was to analyse the chemical composition of the particulate matter PM₁₀ in the city of Chillán, considering its temporary and space variability.

EXPERIMENTAL

This study was carried out in the city of Chillán, Chile, which has a population of 162,930 people (according to the Chilean Censo 2002) living in an urbanized surface of 2,010 hectares. It is located in the northern central part of the Biobío region, between 3634´S latitude and 72 W longitude, and 144 m altitude above sea level. The surrounding lands are the result of fluvial and volcanic deposits, characterized by a flat topography with smooth slopes. These deposits were transported from the Andes Mountain by the rivers, as a result of huge volcanic and torrential events. Its climate is mediterranean with prolonged dry station (6 months) followed by a humid period, and 1100 mm annual average precipitation.

The study was performed between 1 September 2001 to 30 September 2002. The concentrations of particulate matter PM₁₀ were measured in six stationary points within the city of Chillán, by using IMPROVE monitors equipped with a Sierra Andersen 246b cyclone on 47 mm Teflon filters (Gelman Scientific, Ann Arbor, MI). The samplers were positioned taking into account traffic and stationary sources with significant contribution to the emissions (downtown, near to hospital, industries, and other sites), considering prevailing N-S wind direction (Fig 1). The capture of particles was made with a flow rate of 16.7 L/min. Monitoring daily programming time was between the 0:00 to 24:00 h for each equipment.

Fig. 1 Location of the monitors in the city of Chillán, Chile (where M1, ...., M5, corresponds to monitoring stations in different sites within the city, with Mc located in the downtown area, of heaviest vehicular traffic in the city).

In each sampling site, the monitoring program allowed to collect particulate matter PM₁₀. The concentration (mg/m³) of PM₁₀ was determined by using a Cahn 31 microbalance, so filters were weighed before and after each 24-h sampling of continuous monitoring. For the determination of the chemical compounds in the filters, 25 elements (S, Cl, K, Ca, Ti, V, Cr, Mn, Fe, Ni, Cu, Zn, As, Pb, Se, Br, Rb, Sr, Zr, Mo, Na, Mg, Al, Si, P) were analysed by means of PIXE (Particle Induced X-ray Emission), by using a 18 MeV proton beam of energy generated by the CNL cyclotron at the University of California, Davis. Also, energy-dispersive X-ray fluorescence spectrometry was used for analyzing elemental concentration. A description of these techniques is in the literature available ^{15, 16)}. Also, energy-dispersive X-ray fluorescence spectrometry (ED-XFR) was used for analyzing elemental concentrations in a laboratory at the Universidad de Concepción, Concepción, Chile.

Filters were used to determine organic carbon (OC) and elemental carbon (EC) by means of an optical analyser thermal DRI model 4000X. The concentration of organic matter (OM) was obtained multiplying the content of OC by a factor 1.4 as indicated by some authors ^{9, 17)}. The analysis to detect ammonium (NH4⁺) was made by means of a spectrometer Lambda 2 Perkin Elmer. The ions nitrate (NO₃^-), sulphate (SO₄^{2 -}) and chloride (Cl ^-) were determined by ion chromatography (Dionex DX 100).

In order to calculate the proportion of soil (wind-blown dust) in PM₁₀, the equation proposed by Aldape et al.¹⁸⁾ was used here: Soil = 2.20Al + 2.49Si + 1.63Ca + 1.94Ti + 2.38Fe, where Si, Al, Fe, Ca, and Ti are the concentrations of those elements in mg/m³.

Potassium is in natural form in the ground, but also it is present in the smoke of the combustion from vegetative biomass, so producing soluble K. In Chillán, wood is a common fuel for housing heating, so producing visually dense smoke clouds in cold months. In order to calculate the amount of non-geologic potassium, the equation suggested by Eldred et al. ¹⁵⁾ was used here: K_nos = K ­ 0.6Fe, where K and Fe are the concentrations of those elements (mg/m³).

Each monitor was equipped to handle 4 cassettes plus an internal programmable clock, which allowed changing filters and equipment maintenance every five days. Monthly, in each site of monitoring it was prepared a mixed sample for the chemical analyses: 10 filters for analysis of ions, 3 filters to determine carbon, and 10 filters for determination of elemental analysis. For the mixed samples the selected filters were shredded to small pieces of about 1 cm², macerated with nitric acid (65%) and centrifuged. The weather data (temperature, wind speed and relative humidity) were obtained from the meteorological station at the Universidad de Conception, located in the limit of the city of Chillán. Simultaneously, these variables were measured at downtown during the present research.

RESULTS AND DISCUSSION

Table 1 summarizes the seasonal and average chemical composition of PM₁₀ in Chillán during 2001 and 2002: Spring (21 September - 21 December 2001), Summer (22 December 2001 - 21 March 2002), Autumn (22 March - 21 June 2002), and Winter (22 June - 21 September 2002). Each seasonal value is the result of the average of six stations with 90 days continuous monitoring. The total mass concentration (mg/m³) annual average of 84.42 ± 11.68 was higher than the limit of 50 mg/m³ established by EPA, and the levels found in the Chilean cities of Santiago and Temuco ¹⁴⁾.

Table 1. Chemical composition of PM₁₀ (mg/m³) in the atmospheric air of the city of Chillán, Chile, between September of 2001 to September of 2002. The maximum value indicates the highest concentration registered in the period.

As shown in Fig. 2, the main constituent of PM₁₀ was found to be organic matter (31%), followed by elemental carbon (16%), soil (22%), nitrate (10%) and ammonium (7%), whereas most prominent elements were Si, K, Fe, Ca, Al, Cl, Zn, Ti, y Cu. The components OM, EC, and NO₃ result from combustion gases of vehicles and firewood burning ^19,20), whereas the ammonium is originated from agricultural activities ³⁾. The nitrate concentration is directly associated with the traffic of the city because it is formed from the NO₂. The Cl comes from the ocean and is an element that normally is present in soluble form in nature, so is generally part of the atmosphere of most Chilean cities ²¹⁾.

Fig. 2 Relative percentage of chemical composition of PM10 within the city of Chillán, Chile, during September of 2001 to September of 2002.

PM₁₀ was clearly higher during the cold season (March-August) than during the warm months (September-February). The same seasonal pattern was observed for NO₃, OM, EC, V, Zn, K, Mo, Mn, Cl, and Ni. On the other hand, higher concentrations of Si, Al, Ti, Ca, Sb, and Fe were registered during summertime (December-March) than during cold months. This finding are similar to results registered in recent studies abroad ^10,19,7,21), suggesting that air pollution during the cold season originates from anthropogenic sources, whereas in summertime comes from natural sources. The high concentration of Si, Al, Fe, and Ca during summertime can be explained by re-suspension of dust from roads and surrounding agricultural lands. Originally, soils of the Chilean Central Valley were mostly formed from volcanic ashes, therefore easily release many particles by action of the wind and farming activities, especially when they are dried. In addition with that, 60% of the urban streets in the city of Chillán still remain unpaved ²²⁾.

A seasonal behaviour was observed for lead. There was an increase in May, June and July as compared to December, January and February. It is necessary to emphasize that the Chilean Petroleum Company eliminated the use of lead in gasoline by March of 2001. For that reason, present lead levels were lower than the levels found in a preliminary study performed in 1998 in the city of Chillán ²³⁾, and are similar to the city of Santiago, where lead is also decreasing following the same tendency ²⁴⁾. Copper is an abundantly natural element in Chilean soils, therefore its behaviour can strongly be related to the fraction of PM₁₀ originated by wind-blown dust. With regard to manganese, human population may be exposed to manganese released into the air by the combustion of the unleaded gas that contains the organic manganese compound, methylcyclopentadienyl-tricarbonyl-manganese as anti-detonator.

The SO₄ concentration in the atmosphere of Chillán does not appear to be a crucial factor, in coincidence with data reported by Paoletti et al. ²⁵⁾. The highest sulphur concentration corresponded to the monitoring station located near the local hospital (M₃), which has an energy plant with coal-fired boiler. High concentrations of V and Ni (products of oil combustion emissions), Br, Pb, EC, and OM (all components from the vehicle combustion) were measured at monitoring station M_c, located at downtown area. In the city of Chillán usually circulate public cars, buses and trucks that have been retired from circulation in the metropolitan region (Santiago) due to strict regulations from local environmental authorities¹⁴⁾, so for that reason they are the main responsible for these compounds.

In Table 2 monthly averages PM₁₀ concentrations are related to temperature, wind speed and relative humidity. In general, a higher concentration of PM₁₀ along with lower temperature and wind speed, and increasing relative humidity was noted. This finding agrees with other previous studies ^7,18). Temperatures were lower during autumn and wintertime, so probably most of the particles become condensed instead of volatilising. Temperatures measured at downtown area were in average 22.5% higher than sites at urban limit, whereas wind speed was approximately 60% lower, suggesting a strong influence of human activity in the inhalable particulate matter.

Table 2. Monthly average concentration of PM₁₀ and meteorological parameters (temperature, wind speed, and relative humidity) between September 2001 to September 2002 of Chillán, Chile. The average is the mean value of the parameter in the entire period of study.

M2 is the monitoring station at the Universidad de Concepción, Campus Chillán. Mc is the monitoring station at downtown

As previously noted by Aldape et al. ¹⁸⁾, high correlations were found for the elements Pb, S, V, Mn and for Fe, Si, Al, Ca, Ti (Table 3). In the first case, a clear anthropogenic origin was observed, whereas the second group is a result of geologic sources. It is important to notice that there was a low correlation for K, as compared to Si and Fe, which suggests the presence of smoke in the atmosphere of Chillán, as a result of wood combustion from residential houses and municipal waste incinerators.

Table 3. Pearson correlation for the elements measured in the city of Chillán (September 2001 to September 2002).

Fig. 3 indicates that inhalable particulate matter is dominated with mineral particles to which occasionally smaller particles are attached; also aggregates of soot are present in this sample, which is in coincidence with other studies ^{25, 26)}.

Fig. 3 Electronic microscopy of a PM10 sample collected in the city of Chillán, Chile (June 5, 2002).

A clear temporary variability for OM, NO₃, NH₄, soil and K_nos was observed (Fig. 4). A seasonal pattern for EC was not observed. These results were similar to those of Röösli et al. ⁷⁾. The values of OM, NO₃, and K_nos were higher during the cold months, whereas the concentrations of the NH₄ were higher during springtime. On the other hand, the highest soil concentrations (Si, Al, Ca, K, Fe and Ti,) were measured in warm months.

Fig. 4 Monthly average variation of the chemical components of PM10 for the period between September of 2001 to September of 2002

The fraction of potassium contained in products of the combustion from housing heating (K_nos) was prominent in cold months (between April to July), when wood is used massively in the city of Chillán, producing a dense smoke cloud that remains to very low height and that does not dissipate in those days with atmospheric stability (low temperature and wind speed, with high relative humidity).

In Fig. 5, a space and temporary variability of the chemical compounds at each monitoring site in the city of Chillán is showed. The EC, OM, and NO₃ concentrations were higher at downtown area (at monitoring site Mc). No seasonal pattern was clearly observed for EC. Ammonium concentrations were slightly superior at the limits of the city (M₂) and during springtime, which is time coinciding with applications of pesticides and fertilizers on agricultural lands that surround Chillán.

Fig. 5 Temporal and spatial variation of the chemical composition of PM10 in the city of Chillán, Chile, between September of 2001 to September of 2002.

It is possible to note (Fig. 6) that higher soil concentration was clearly manifested between December to March (maximum in January), at those sites far apart of downtown (monitoring stations M₁ and M₂). The reason for that is because these points were at the periphery of the city of Chillán, so received direct emissions from lands that surround the urban area, and also for re-suspension of dust from roads produced by urban and rural traffic and that is transported by prevailing wind direction (during spring and summertime was from south).

Fig. 6 Inhalable particulate matter concentration in the soil fraction and wood combustion of Chillán (September of 2001 to September of 2002).

It is also necessary to note that Chillán has a 60 percent of streets without paving. Non-geologic potassium levels (K_nos) from the combustion of wood were present during the cold season (April to July), being the highest at downtown.

CONCLUSIONS

The results obtained in this research showed that there was a distribution pattern of the chemical components in the inhalable particular matter as a function of the anthropogenic activity within the city of Chillán. The chemical analyses of PM₁₀ revealed that carbonaceous substances (EC, OM), crustal material and inorganic substances of secondary origin (NO₃ and NH₄) are the predominant components of PM₁₀ in the city of Chillán during September of 2001 to September of 2002 (approximately 85%). A clear temporal variability was observed, because there was a higher concentration of PM₁₀ during cold season than during the warm months. Also, spatial variability was noted as downtown of Chillán resulted more contaminated by chemical compounds (OM, NO₃, EC, K_nos, Ti, K, Pb, Mn, Ni, V, and Mo) compared to surrounding areas of the city. Nevertheless, because Chillán is surrounded by agricultural lands, these perimetrical areas presented higher ammonium concentrations during springtime, whereas received large loads of geologic material in suspension in summertime. In general, there were no high SO₄ concentrations in the city of Chillán, which can be explained for there are few industries nearly. However, it is a matter of a worry the presence of this compound in the hospital location, with large combustion installations burning fuel coal the year throughout.

The results, even though from limited coverage, indicate that the inhalable atmosphere of the city of Chillán is to be considered as a problem of anthropogenic origin during autumn and winter, in coincidence with significant amounts of wood used as fuel for housing heating and with the periods of atmospheric stability.

It is recommended to accomplish a cadastre of emissions for the city of Chillán, thus to identify the contributions from stationary point sources and traffic pollutants.

ACKNOWLEDMENTS

This research was part of the corresponding author's Ph. D. thesis in Environmental Sciences at the Centro EULA of the Universidad de Concepción, Chile. It was partly supported by projects P.I. 200.132.004-1.1 and Instrumentación Científica "Sistema de Medición de Variables Ambientales", both from the Dirección de Investigación of the Universidad de Concepción, and by Fondecyt project 1000828. Also, many thanks to the Crocker Nuclear Laboratory of the University of California (Davis), for providing the IMPROVE PM₁₀ monitors and allowing some of the chemical analyses.

REFERENCES

1. D. Dockery, C. Pope, X. Xu, J. Spengler, J. Ware, M. Fay, B. Ferris, F. Speizer. N. Engl. J. Med., 329, 1753-1759 (1993).         [ Links ]

2. G. Hoek, J. Schwartz, B. Groot, P. Eilers. Arch. Environ. Health, 52, 455-463 (1997).F.         [ Links ]

3. R. Harrison, J. Yin, J. Sci. Total Environ., 249, 85-101 (2000).         [ Links ]

4. J. Schwartz, D. Dockery, L. Neas. J. Air Waste Manage. Assoc., 46, 2-14 (1996).         [ Links ]

5. J. Samet, F. Dominici, F. Curriero, I. Coursac, S. Zeger. N. Engl. J. Med., 343, 1742-1749 (2000).         [ Links ]

6. M. Préndez, P. Ulriksen. La contaminación del aire y sus efectos. En: Contaminación Atmosférica de Santiago: Estado Actual y Soluciones, Ed. Sandoval, Préndez y Ulriksen, Universidad de Chile, Santiago, Chile. pp. 23-36 (1993).         [ Links ]

7. M. Röösli, G. Theis, N. Künzli, J. Staehelin, P. Mathys, L. Oglesby, M. Camenzind, Ch. Braun-Fahrländer. Atmos. Environ., 35, 3701- 3713 (2001).         [ Links ]

8. X. Querol, A. Alastuey, S. Rodríguez, F. Plana, E. Mantilla, C. Ruiz. Atmos. Environ., 35, 845-858 (2001).         [ Links ]

9. R. Eldred, T. Cahill, R. Flocchini. J. Air Waste Manage. Assoc., 47, 194-203 (1997).         [ Links ]

10. K. Müller, K. Atmos. Environ., 33, 1679-1685 (1999).         [ Links ]

11. J. Chow, J. Watson, S. Edgerton, E. Vega. Sci. Total Environ., 287, 177-201 (2002).         [ Links ]

12. M. Préndez, J. Ortíz. Bol. Soc. Chil. Quím., 27, 283-285 (1982).         [ Links ]

13. M. Préndez, M. Carrasco. Bol. Soc. Chil. Quím., 27, 310-312 (1982).         [ Links ]

14. UChile (Universidad de Chile). Informe País: Estado del Medio Ambiente en Chile. Centro de Análisis de Políticas Públicas, LOM Ediciones, Santiago, Chile. pp.33-72 (1999).         [ Links ]

15. R. Eldred, T. Cahill, P. Feeney. Nucl. Instr. and Meth., B22, 289- 295 (1987).         [ Links ]

16. J.R. Brook, T.F. Dann, R.T. Burnett. J. Air Waste Manage. Assoc., 47, 2-19 (1997).         [ Links ]

17. B. Kim, S. Teffera, M. Zeldin, M. J. Air Waste Manage. Assoc., 50, 2034-2044 (2000).         [ Links ]

18. Aldape, J. Flores, R. Díaz, R., J. Miranda, T. Cahill, J. Morales. I. J. PIXE 1, 373-388 (1991).         [ Links ]

19. A. Turnbull, R. Harrison. Atmos. Environ., 34, 3129-3137 (2000).         [ Links ]

20. J. Gillies, A. Gertler, J. Sagebiel, W. Dippel. Environ. Sci. Technol., 35, 1054-1063 (2001).         [ Links ]

21. N. Hrepic, P. Mladinic, C. Díaz, M. Prendes. Bol. Soc. Chil. Quím., 28, 477-479 (1983).         [ Links ]

22. PLADECO (Plan de Desarrollo Comunal), 1998. Municipalidad de Chillán, Dirección de Planificación. 105 p.         [ Links ]

23. J. Celis, O. Carvacho, R. Flocchini, J. Cañumir, A. Flores. Agro- Ciencia, 18, 133-142 (2002).         [ Links ]

24. SESMA (Servicio de Salud Metropolitano del Ambiente). Caracterización de elementos inorgánicos presentes en el aire de la Región Metropolitana 1997-2000. Ministerio de Salud, Laboratorio de Salud Ambiental, Santiago de Chile. 42 p (2002).         [ Links ]

25. L. Paoletti, B. De Berardis, M. Diociaiuti. Sci. Total Environ., 292, 265-275 (2002).         [ Links ]

26. L. Greenwell, T. Moreno, T. Jones, R. Richards. Free Radical Res., 32, 898-905 (2002).         [ Links ]