Volatile aromatic compounds in Mexico City atmosphere: levels and source apportionment

 

V. Mugica

Universidad Autónoma Metropolitana-Azcapotzalco. Av. San Pablo No. 180, Col. Reynosa Tamaulipas Azcapotzalco, 2200 México, D. F. Email: uma@correo.azc.uam.mx.

 

M. E. Ruiz

Instituto Mexicano del Petróleo, México, D.F.

 

J. Watson and J. Chow

Desert Research Institute, Reno, Nevada. U.S.A.

 

Received March 1, 2002; accepted September 12, 2002

 

RESUMEN

Se colectaron simultáneamente muestras de aire ambiente en tres sitios de la Ciudad de México durante el mes de marzo de 1997 con el fin de conocer las concentraciones y el origen de compuestos aromáticos utilizando el modelo de balance de masa de especies químicas (CMB). Los compuestos aromáticos volátiles representaron alrededor del 20% del total de hidrocarburos no metánicos presentes en las muestras matutinas colectadas. Las especies aromáticas volátiles mas abundantes en el ambiente fueron el tolueno y los xilenos, seguidos por 1,2,4 trimetilbenceno, benceno, etilbenceno, melaetiltolueno, npropilbenceno, isopropilbenceno, 1,3,5 trimetilbenceno y estireno. Se llevaron a cabo compañas de muestreo en cruceros, en una estación de autobuses, un estacionamiento y algunos sitios en los que se utilizan disolventes y destilados de: petróleo, para determinar los niveles de compuestos aromáticos volátiles a los que se expone la población. Se aplicó el modelo CMB para determinar la contribución de distintas fuentes a la presencia en la atmósfera de cada uno de los compuestos aromáticos más abundantes. El escape de vehículos fue la fuente principal de todos los compuestos aromáticos, especialmente los de gasolina, aunque el escape de vehículos a diesel y el asfaltado contribuyeron también en forma importante a la presencia de tolueno, xilenos, etilbenceno, propilbencenos y estireno. Las artes gráficas y el uso de pinturas impactó principalmente en la presencia de tolueno.

 

ABSTRACT

Samples of ambient air were simultaneously collected at three different sites of Mexico City in March of 1997 in order to quantify the most abundant volatile aromatic compounds and estimate the source contributions by application of the chemical mass balance model (CMB). Volatile aromatic compounds were around 20% of the total of non-methane hydrocarbons present in morning air samples. The most abundant volatile aromatic species in urban air were toluene and xylenes followed by 1,2,4 trimethylbenzene, benzene, ethylbenzene, metaethyltoluene, 1,3,5 trimetlhylbenzene , styrene, npropylbenzene, and isopropylbenzane. Sampling campaigns were carried out at crossroads, a bus station, a parking place, and areas where solvents and petroleum distillates are used, with the objective of determining people's exposure to volatile aromatic compounds. The CMB was applied for estimating the contribution of different sources to the presence of each one of the most abundant aromatic compounds. Motor vehicle exhaust was the main source of all aromatic compounds, especially gasoline-exhaust, although diesel exhaust and asphalt operations also accounted for toluene, xylenes, ethylbenzene, propylbenzenes, and styrene. Graphic arts and paint application had an important impact on the presence of toluene.

Key words: Mexico, volatile organic compounds, benzene, CMB.

 

1. Introduction

The presence of volatile aromatic compounds in urban air is of environmental concern because they are precursors to photochemical smog including the production of secondary aerosols and the formation of ozone, as well as some of them have adverse effects; on human health (Odum et al., 1997). Several studies have been conducted in other countries to determine the atmospheric concentration of these compounds, mainly benzene, toluene and xylenes. Dann and Wang (1995) and Cheng et al. (1997) in Canada; Broceo et al., (1997) in Italy; Grosjean et al. (1998) in Brazil; Thijsse et al. (1999) in Berlin; Rommelt et al. (1999) in Munich: Derwent et al. (2000) in the United Kingdom; Kourtidis et al. (2000) in Greece; Vukovich (2000) in the United States; Na and Kim (2001) in Korea. In Mexico the information about non-methane hydrocarbons (NMHC) and specifically of aromatic compounds in urban atmospheres is very scarce because measurements are difficult and expensive to undertake (Arriaga et al., 1997; Sosa, 2001).

Volatile aromatic compounds are used as constituent in motor fuels, as chemicals intermediates, as solvents for fats, inks, oils, paints, plastics, etc. Their presence in the ambient air is associated to fuel incomplete combustion and to fuel and solvent evaporation. Increased concern for benzene as a significant environmental pollutant arises from public exposure to the content of benzene in gasoline, due to requirements for unleaded fuels for automobiles equipped with catalytic exhaust converters (Bravo, 2000). Benzene is a known human carcinogen and exposure has been linked to an increased risk of several forms of leukemia, Lynge et al. (1997) reported the association of leukemia and other forms of cancer to benzene exposure in gas stations. Long term exposure to benzene, toluene, ethylbenzene, xylene isomers (ortho, meta and para), trímethylbenzencs (TMB), and styrene, among other aromatics, may cause loss of appetite, nausea, weight loss, fatigue, headache, nervousness and irritability and has been associated to chronic respiratory diseases, dermatitis and other diseases (Sittig, 1991). The World Health Organization (WHO, 1987) considered no safe limit for benzene, but some countries have adopted an air quality guide values for it (i.e Germany with 2.5 μ g/m³/24 h). The WHO guideline recommended a safe limit for toluene of 7.5 μg/m³/24 h).

On the other hand, volatile aromatic compounds have been associated to the formation of secondary organic aerosol (Odum et al., 1997). Mugica et al., (2002b) used the maximum incremental reactivity coefficients determined by Carter (1994) to rank different NMHC in relation to ozone formation. Xylenes, toluene, 1,2,4 trimethylbenzene_r and 1,3-5 trimethylbenzenc ranked within the ten compounds, that play a major role in ozone production in Mexico City. With the aim to control air pollution in Mexico City, the installation of catalytic converters in vehicles has been mandatory since 1991. As a consequence, lead in gasoline was eliminated antl aromatic content increased. The content of aromatics in gasoline fuels on volume base has been controlled since 1994 to a maximum of 25%, and the content of benzene to a maximum of 1%. A gasoline vapors recovery program at gas stations has been implemented in Mexico City. Nevertheless, the atmospheric concentration of benzene in Mexico City has been reported as one of the highest in the world (Arriaga et al., 1997). In 1995, the Federal District Administration conducted some studies in a specific site of Mexico City using a differential absorption spectrometer (OPSIS), founding a mean level of benzene of 45 μg/m³/.

The main purpose of this research was to investigate the origin of the most abundant aromatic volatile compounds in Mexico City atmosphere, as well as the exposure levels in specific sites.

Traditionally, the CMB model has been, applied in many cities to determine how several sources contribute to the presence of NMHC (Watson et al., 2001). In this study, the CMB model was used to determine the source apportionment to the presence of individual volatile aromatic compounds go it would be possible to establish specific control measures for each specific aromatic compound.

 

2. Methodology

2.1 Sampling

The sampling of ambient air and sources was carried out according to the EPA TO-14 method using pre-evacuated SUMMA canisters (EPA, 1988). The stainless steel canisters employed were cleaned in the laboratory by repeated evacuation and pressurization with humidified helium chromatography grade. Three different places were selected for air ambient sampling in February and March, 1997: the first one, La Merced, is a commercial site near downtown; the second place, Xalostoc, is an industrial zone northeast Mexico City, and the third one, Pedregal, is a residential area located southwest Mexico City. Sampling time was from 6:00 to 9:00 A.M., and a total of 36 samples were collected in the stations located at a height of 2 to 3 meters above ground level. 56 additional samples were taken in La Merced with sampling periods of six hours, from 6:00 A.M. to 12:00 P.M. and from 12:00 P.M. to 18:00 P.M.

To characterize vehicular emissions, twelve samples were taken at ground level along three crossroads with a high density of traffic and seven samples were collected at a bus station. The sampling time was two hours (Sanchez et al., 2001; Mugica et al, 2001). Five samples were collected at a theater parking 20 minutes once the garage was full (Mugica et al., 1998). Other exposure levels of aromatic compounds were measured at different sites where solvents and petroleum distillates are used. The selected places were; four graphic art workshops (one serigraphy and three offsets), three automotive painting sites, two places where architectonic coatings were being applied, two furniture shops where lacquers and varnishes were used. Sampling time period was 15 minutes. (Vega et al, 2000b).

2.2 Analysis

Canister samples were analysed using a Hewlett Packard Gas Chromatograph (5890 Model) with a flame ionization detector (FID) after concentration by collection in a cryogenically-cooled trap. The cryogenic method removes impurities and pre-concentrates the hydrocarbons in ambient air. The chromatographic capillary column used for C2-C14 hydrocarbon analysis was a J&W DB-1, made of fused silica, with an internal diameter of 0.32mm, length of 60m, and 1 μm phase thickness. The oven temperature program was -50°C for 3min, to 200°C at 8μC/min. The chromatograph was calibrated by injecting known amounts of certified standard calibration gases onto the chromatography column.

 

3. The CMB model

The CMB model allows the correlation of measured species concentrations in a site known as receptor and the emission sources that originate them. The CMB model consists of a least-squares solution to a set of mass-balance equations that express each receptor chemical concentration as a linear sum of products of source profiles abundances and source contributions. The source profiles abundances and the receptor concentrations with appropriate uncertainty estimates, are the input data for the CMB model. The CMB model calculates values for the contributions from each source type and the uncertainties of these values (Watson cf at, 1991). In recent years different source profiles have been developed in Mexico City to apply the CMB model (Mugica et al, 1998; 2001; Vega et al, 20D0a,b). The base of the receptor model is the Following equation, where the relationship between the concentrations of the chemical species measured in the receptor with those emitted in the source are expressed:

where C_i is the ambient concentration of the species i measured at the receptor site; P is the number of sources that contribute; F_ij is the fraction of source contribution: S_j is the calculated contribution of source j. The number of chemical species (I) must be greater than or equal to the number of sources (J) for a unique solution to this equation. The CMS software in current use is the CMB8, which operates in a Windows-base environment and accepts inputs and creates outputs in a wider variety of formats than previous CMB versions (Watson and Chow, 1998). One of the CMES output files is a table that shows the contribution of each source to the calculated ambient concentration for each species.

 

4. Results

4.1 Ambient air concentrations

Table 1 shows the content of each type of organic groups at the three sites. Total volatile aromatic compounds accounted for 18 to 22% of the total NMHC, In average, La Merced was the most polluted area followed by Xalostoc. Pedregal had concentrations lower than one third of the other two sites, although it presented the highest percentage of aromatic compounds of the total NMHC.

Thirty-three aromatic compounds were detected, and the most abundant species in the three sites were in decreasing order: toluene, xylenes (ortho, meta and para isomers), followed by 1,2,4 trimethylbenzene, benzene, ethylbenzeuc, metaethyltoluene, 1,3,5 triinethylbenzene, npropylbenzene, isopropylbenzene, and styrene. These species account for 80% of the total aromatic compounds. Figure 1 and Figure 2 show the median, minimum and maximum aromatic compound concentrations in a box plot presentation of the three sites. In all cases the highest concentrations were found In La Merced and the lowest in Pedregal. Benzene was about 6% of the total of volatile aromatic compounds with concentrations up to 30 μg/m³.

Table 2 shows the average concentration of each compound in La Merced, both in the morning and in the afternoon. Rush hours are mainly in the early morning, so the average of total NMHC concentrations contained in the six hour morning samples (6 A.M.-12 P.M.) was 33% lower than the average concentrations of the three hour samples. (6-9 A.M.), although the volatile aromatic concentrations in the six hour samples were only 20% lower than those observed in the three-hour morning samples. On the other hand, both NMHC and aromatic compound concentrations decreased around 50% from the morning to the afternoon due to an improved dispersion and as a result of photochemical reactions in the atmosphere. Arriaga's (1997) study reported that the aromatic compound concentrations did not present a decreasing trend from 1993 to 1996. The present study found that these concentrations have remained high.

Table 2 also shows the comparison with other studies carried out in different countries. Most of them show that the highest concentrations are observed in the early morning. It is Important to stress that it is difficult to compare our results with those observed in other cities due to the differences in several factors, such as sampling sites and sampling periods; nevertheless, these data are presented to show how high the air pollution levels in Mexico City are.

4.2 Street, and bus station levels

Table 3 shows the volatile aromatic concentrations found in several crossroads, a bus station, and a parking lot. Values are more than twice as concentrated than in ambient air. This implies that policemen, street vendors and pedestrians along the street or adjoining owners of highly traveled roads are exposed to high dose of volatile aromatics. The results of this study agree with the benzene exposure study carried out by Meneses and Hernández-Avila (1999) who reported that street vendors are exposed to 83 μg/m³ of benzene every day. Although the increased use of catalytic converters has lowered the level of volatile organic compounds from vehicle exhaust ami the benzene content in fuels is controlled, several studies have confirmed that a substantial amount of benzene is produced by thermal decomposition of other aromatic compounds such as toluene, xylenes, and trimethylbenzenes in catalytic converters (Casinhns, 1999). Therewere, benzene and other aromatic compounds must be importantly decreased in fuel contents.

4.3 Solvents and petroleum distillate use

The use of solvents and distillates of petroleum in residential or commercial areas is of environmental concern because many of them are handled carelessly in the open air where not only workers, but also pedestrians, inhabitants and vendors are exposed to the emission of toxic vapors with a high content of aromatic compounds. Other closed sources located in residential areas such as dry cleaners and graphic art workshops (offset and serigraphy) lack emission control equipment because there are no national regulations; in this regard, adjoined owners or neighbors are exposed to NMHC emissions. Table 4 shows the mean concentrations and standard deviations found in these kinds of settings. Toluene's measured concentrations followed by xylenes' were very high in the graphic art workshops and where paint and varnishes are applied. Alkyl and trimethylbenzenes had the highest concentrations in dry cleaners.

4.4. Source apportionment

Vega et al (2000b) and Mugica ct at (2002a) applied the CMB model to conduct source apportionment studies in Mexico City. These studies concluded thai the major sources of NMHC were motor vehicle exhaust (50-64%), handling and distribution of LP gas (15-25%), and asphalting operations (9-12%). In this study the CMB model was applied to determine the source contribution of each of the most abundant aromatic species. Dry cleaners and decreasing; were not included in the CMB analysis because an initial test showed negligible contributions to the total NMHC in the ambient air. In general, the measured and calculated mass of aromatic species had a good concordance with deviations smaller than 20%, except 1,2,4 trimethylbenzene since only around 50% could be apportioned. This means that there are other sources not considered in the receptor model application.

Figures 3, 4 and 5 show the major sources for each species (calculated mass base) in the three sites. The contributions of vehicular emissions (gasoline, diesel and evaporative) for all the considered volatile aromatic species were quite similar for La Merced arid Pedregal. The contributions to aromatic compounds from diesel exhaust and asphalt operations were highest in Xalostoc. In general, the gasoline vehicular exhaust was the major source of the considered aromatic compounds, with contributions up to 78% for metaethyltoluene, 70% for benzene, 70% for npropylbenzene, 68% for 135 trimethylbenzene and isopropylbenzene, 65% for styrene and 60% for toluene. The contribution of this source was lower for ethylbenzene and xylenes, with an average of 42% and 24%. Other contributors to benzene were diesel-powered engines with around 10% in La Merced and Pedregal, and 19% in Xalostoc. The increased emissions of benzene in Xalostoc can be explained by the large number of industries located in the area that use diesel for the operation of boilers and engines. The contribution of evaporative sources to benzene was higher in Xalostoc (11%) than in La Merced (3%) and Pedregal (5%), maybe due to the presence of open tanks for industrial purposes.

In addition to gasoline exhaust emissions, about 20% of toluene was emitted by diesel exhaust sources, 13% in average by asphalt operations, and up to 11% by architectural coatings and graphic arts. As well as benzene, the major contribution of gasoline vapors to the concentration of toluene was found in Xalostoc (5.4% vs. 1.5%  in the other two sites). In general, the contribution of diesel engines exhaust was very important to the presence of alkylated aromatics, such as xylenes, ethylbenzene and styrene, particularly lit Xalostoc. Metaethyltoluene and 1,3,5 trimethylbenzene had the lowest contributions by this source [up to 8 and 15%), The contribution of evaporative emissions to other aromatic compounds than benzene was less to 5%, while the highest values were found in Xalostoc. Asphalt operations were Important contributors to ethylbenzene, xylenes, 1,3,5 trimethylbenzene, isopropylbenzene, npropylbenzene, and styrene, with average percentages of 25, 24, 21, 18, 18, and 13%, respectively.

Emissions from cooking contributed with percentages less than 2% of aromatics, except for benzene, which readied up to 4% In some samples. These contributions can be explained by the presence of many complex compounds in the charcoal used in cooking. The contribution of landfills to volatile aromatic compounds was only found in La Merced, mainly for ethylbenzene and xylenes (both with 6%).

 

5. Conclusions

Non-methane hydrocarbons were measured in Mexico City in March, 1997 by means of gas chromatography. Ambient data collected from three sites in Mexico City show that the aromatic hydrocarbon fraction ranges from 18 to 22% of the total NMHC concentration. The composition of top species was similar from site to site. Benzene accounts for an average of 6% of the aromatic fraction, while toluene contributes 29%, and xylenes 24%. Exhaust and evaporative gases are the main sources of aromatic compounds In Mexico City atmosphere. The volatile aromatic concentrations at several crossroads and the parking garage were several times higher than ambient air concentrations, and entail a health risk for pedestrians, street vendors and policemen.

Although the CMB model estimations were found to have small contributions from sources which use solvents (such as graphic arts workshops and paint and varnishes application), this study showed that neighbors and people near these places arc exposed to high concentrations of aromatic compounds, especially toluene. This suggests that regulatory initiatives to control exposure are also needed in these settings, as well as some regulations of toluene's content in solvents. Vehicles are the main source of all emissions of aromatic compounds. Thus, a strong reduction of aromatic compound emissions from vehicle traffic appears to be mandatory. Asphalting operations should he also considered to diminish the presence of alkylbenzenes.

The continuous monitoring of volatile aromatic compounds, especially benzene, should be seen as an important and significant improvement of the existing network in order to upgrade its response to an effective strategy to control pollution. Long-term measurements are needed to provide important information on hazardous and reactive pollutants, such as human exposure, exposure trends, and estimates of source contributions, as well as to evaluate the progress achieved through the application of control strategies.

 

Acknowledgements

UAM-A, PEMEX and IMP supported this research as part of the project FIES-93-93-VI.

 

References

Arriaga, J., S. Escalona, A.D., Cervantes, R. Orduñez and T. López 1997. Seguimiento de COV en aire urbano de la ZMCM, 1992-1996. García Colin Scherer, L. Várela Ham, (Eds). In Contaminación Atmosférica II. El Colegio Nacional/Universidad Autónoma Metropolitana, México.         [ Links ]

Bravo, H.,R. Sosa, P. Sánchez, and M. Jaimes, 2000, "Potential urban ozone pollution as a product of an inadequate implementation of a best available control technology in developing countries". Proceedings of 93rd Annual Meeting of the Air &. Waste Management Association. Salt Lake City, Utah, U.S.A.         [ Links ]

Brocco, D., R. Fratancangeli, L. Lepore, M. Petricca and I. Ventrone, 1997. Determination of aromatic hydrocarbons in urban air of Rome. Atmospheric Environment, 31-4- 557- 566.         [ Links ]

Carter, W., 1994. Development of ozone reactivity scales for volatile organic compounds. J. Air Waste Manage Assoc, 44, 881-899.         [ Links ]

Casinhas, L., M. Fraser, D. Liang, and U. Boesl, (1999). A pyrolisis cell as simulator for an automobile catalytic converter. Vaccum, 52, 89-97.         [ Links ]

Cheng, I., I. Fu, R. Angle, and H. Sandhu, 1997. Seasonal variations of volatile organic compounds in Edmonton, Alberta. Atmospheric Environment, 31, 239-246.         [ Links ]

Dann, T. and D. Wang, 1995. Ambient air concentrations in Canada (1989-1993): Seasonal and day week variations, trends, and source influences. J. Air Waste Manage. Assoc., 45, 695-702.         [ Links ]

Derwent, R., T. Davies T., Delaney, G. Dollard, R. Field, P. Dumitrean, P. Nason, B. Jones, and S. Pepler, 2000. Analysis and interpretation of the continuous hourly monitoring data for 26 C2-C8 hydrocarbons at 12 United Kingdom sites during 1996. Atmospheric Environment, 34, 297-312.         [ Links ]

EPA, 1988, Compendium method TO-14. The determination of volatile organic compounds in ambient air using canister sampling and gas chromatography analysis. U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 2771.         [ Links ]

Grosjean, E., D. Grosjean and R. Rasinussen, 1998. Ambient concentrations, sources, emission rates and photochemical reactivity of C2-C10 hydrocarbons in Porto Alegre, Brazil. Environ Sci. Technot., 32, 2061-2069.         [ Links ]

Kourtudis, K., I. Ziomas, C. Zerefos, A. Gousopoulos, D. Balis, and P. Tzoumaka, 2000. Benzene and Toluene levels measured with a commercial DOAS system in Thessaloniki Greece. Atmospheric Environment 34, 1471-1480.         [ Links ]

Lynge E., A. Andersen, and R. Nilsson, 1997, Risk of cancer and exposure to gasoline vapors. Am. J. Epidemiology, 145, 449-458.         [ Links ]

Meneses, F., and Hernández-Avila M., 1999. A survey of personal exposures to benzene in Mexico City, Archives of Environmental Health, 2-10.         [ Links ]

Mugica, V., E. Vega, J. L. Arriaga and M. Ruiz, 1998. Determination of motor vehicle profiles for non-methane organic compounds in Mexico City metropolitan area. J. Air Waste Manage. Assoc., 48, 1060-1068.         [ Links ]

Mugica, V., E, Vega, J. Chow, G. Sánchez, E. Reyes, J. Watson, R. Egami and J. L. Arriaga, 2001. Volatile organic compounds emissions from gasoline and diesel powered vehicle. Atmósfera, 14, 29-38.         [ Links ]

Mugica, V., J. Watson, E. Vega, E. Reyes, M. Ruiz and J. Chow, (2002a). Source apportionment of NMHC in Mexico City. The Scientific World, 2. 344-860.         [ Links ]

Mugica, V., E. Vega, H. Ruiz, G. Sánchez, E. Reyes and A. Cervantes, (2002b). Photochemical reactivity and sources of individual VOCs in Mexico City. In Air Pollution VIII. WIT Press, U.K.         [ Links ]

Na, K. and Y. Kim, 2001. Seasonal characteristics of ambient volatile organic compounds in Seoul, Korea, Atmospheric Environment, 35, 2603-2614.         [ Links ]

Odum, J., T. Jungkamp, R. Griffin, H. Forstner, R. Flagan, and J. Seinfeld, 1997. Aromatice reformulated gasoline, and atmospheric organic aerosol formation. Environmental Science and Technology, 31, 1890-1897.         [ Links ]

Römmelt, H., A. Pfaller, G. Fruhmann and D. Nowak, 1999. Benzene exposures caused by traffic in Munich public transportation system during 1993-1997. The Science of Total Environment, 214, 197-203.         [ Links ]

Sánchez, G., E. Vega, E. Reyes, V. Mugica, J. Chow, J. Watson and J. L. Arriaga, 2001. Importance of the determination of nonmethane hydrocarbons to diminish the ozone concentrations in Mexico City. Electronic Proceeedings of 94 Annual Meeting & Exhibition of Air and Waste Management Association, Orlando, Florida, U.S. 24-28 June.         [ Links ]

Sittig, M., 1991. Handbook of toxic and hazardous chemicals and carcinogens. 3rd Ed., Noyes Publications. U.S.A. pp 207-209, 240-241. 1569, 1611, 1659, 1479.         [ Links ]

Sosa, R., 2001. Evaluación y control de benceno en la atmósfera de la Zona Metropolitana de la Ciudad de Mexico. PhD thesis. Universidad Nacional Autónoma de México. México.         [ Links ]

Thijsse, T., F. Van Oss, and P. Lenschow, 1999. Determination of Source Contributions to Ambient Volatile Organic Compound Concentrations in Berlín. J. Air Waste Manage Assoc 49, 1394-1404.         [ Links ]

Vega E., V. Mugica, L. Diaz and F. Ramos, 2000a. Estudio comparativo de perfiles do emisiones vehiculares en túnel y dinamómetro. Rev Int. Contam. Ambient., 16, 55-69.         [ Links ]

Vega, E., V. Mugica, R. Carmona, and E. Valencia, 2000b. Hydrocarbon source apportionment in Mexico City using the chemical mass balance receptor model Atmospheric Environment, 34, 4121-4129.         [ Links ]

Vukovich, F., 2000. Weekday/weekend differences in OH reactivity with VOCs and CO in Baltimore, Maryland. J. Air Waste Manage. Assoc., 50, 1843-1851.         [ Links ]

Watson, J., J. Chow and T. Pace, 1991. Chemical mass balance. In Receptor Modeling for Air Quality Management, Hopke, P.K. Ed. Elsevier Press, New York, NY, 83-116.         [ Links ]

Watson, J, and J. Chow, 1998. CMB8: Chemical Mass Balance Receptor Model Version 8. Course Air-301. Air & Waste Management Association.         [ Links ]

Watson, J., J. Chow and E. Fujita, 2001. Review of volatile organic compound source apportionment by chemical mass balance. Atmospheric Environment, 35, 1007-1584.         [ Links ]

WHO, 1987, Air quality guidelines for Europe. European Series 23. World Health Authority Regional Publications, Copenhagen, Denmark.         [ Links ]