Evaluation of Air Quality in Volta Redonda , the Main Metallurgical Industrial City in Brazil

A maior indústria de aço do Brasil está situada na cidade de Volta Redonda, RJ. A produção de coque neste tipo de indústria é a maior fonte de emissão de compostos aromáticos. Com objetivo de implementar medidas de controle da poluição nas indústrias a FEEMA, agência ambiental do Estado do Rio de Janeiro, avaliou a qualidade do ar naquela cidade em Dec/95-Mai/96 e Abr-Mai/99. Na primeira campanha, material particulado total (MPT), benzeno, tolueno, xilenos (BTX) e dióxido de enxofre (SO 2 ) foram analisados e concentrações médias altas foram encontradas: 102 μg m, 70 μg m, 23 μg m, 2,5 μg m e 186 μg m, respectivamente. Após medidas drásticas de controle, a segunda avaliação (Abr-Mai/99), apresentou valores menores: 102 μg m (TSP); 62 μg m (PM 10 ), 23 μg m (B), 4 μg m (T), 0,6 μg m (X), 73 μg m (SO 2 ) além de metais. Esses resultados, principalmente para benzeno, indicam que o programa de monitoramento deve continuar até atingir a solução final para o problema.


Introduction
Volta Redonda, a city in Rio de Janeiro State, grew around the largest steel producing plant in Brazil.It has a current population of 250,000 inhabitants, and is the source of one of the state's main revenues. 1Its environmental problems are basically related to air pollution resulting from the type and location of the region's industrial activities.The most important source of organic pollutants in the monitored area is coke production at National Steel Company (CSN) (4.6 x 10 6 metric tons/year).Volta Redonda is located in a strategic area -halfway between São Paulo and Rio de Janeiro (BR 393 road).Thus, there is high air pollution potential due to vehicles motor exhaust from heavy traffic.Furthermore, many other metallurgical and cement factories were established there to supply CSN demand, contributing to the complex array of possible air pollution sources.
Depending on the type of industry, particles, sulfur dioxide and organic compounds constitute the main atmospheric pollutants. 1 The concentration, composition and particle size of suspended particulate matter are determined by such factors as meteorological properties of the atmosphere, topographical influences, emission sources, and particle parameters such as density, shape, and hygroscopicity. 2articulate matter less or equal to 10 µm in diameter is associated with adverse health effects including increased respiratory problems, mortality and increase of cancer in some urban areas. 3By the same token, the chemical composition of particles (Polycyclic Aromatic Hydrocarbons (PAHs), metals, acidity) can induce health-related effects. 2 Benzene, toluene and xylenes (BTX) are present in the environment because of emissions from automobiles, petrochemical industry and combustion processes.Due to the toxicity and mutagenic or carcinogenic properties of BTX, 4 aromatic hydrocarbon air pollution is a very important topic for environmental health.Although significant, few are the studies related to atmospheric pollution in Brazil.These researches, most related to the effects of traffic, have been basically carried out in large metropolis such as São Paulo, 5,6 Rio de Janeiro, [7][8][9][10][11] Porto Alegre, 12,13 Salvador 14 and in the Amazonian region. 15here are few studies concerning industrial areas.
Aiming to follow the implementation of pollution control measures in industries, FEEMA, the Rio de Janeiro State environmental agency, in partnership with Laboratório de Apoio ao Desenvolvimento Tecnológico (LADETEC/ UFRJ) and Department of Meteorology of Universidade Federal do Rio de Janeiro (IGEO/UFRJ) evaluated air quality in several sampling sites in Volta Redonda.Two evaluations were performed in Dec/95-May/96 and Apr-May/99.The parameters were chosen according to national and international recommendations.The objective of this study was to obtain information regarding air quality in the Municipal District of Volta Redonda, so as to permit the proposal of new pollution control policies, and to increase the population awareness about eventual high pollution levels and environmental degradation.The results can be used in the evaluation of the existing monitoring program and may contribute to the generation of new pollution control strategies to evaluate and restructure the existent network of antipollution devices/procedures.

Monitored area
The Municipal District of Volta Redonda is located on the margins of the Paraíba do Sul river, between the Mantiqueira and Mar mountain ranges, parallels 22 22' 11"   and 22 38' of South latitude, and the meridians 44 o 09'25" and 44 o 20' of West longitude.Its urban perimeter occupies an area of 54 km 2 corresponding to 0.4% of the total area of the state of Rio de Janeiro.Its altitude varies widely, from 363 m at Paraibas's river margins to 707 m (in the northeast tip) above the sea level, while the altitude at the central area is 380.3 m.According to a study published by FEEMA, 16 the climate of Volta Redonda can be classified as mesotermic with a compensated medium temperature of 21 o C. The annual maximum average is 27.8 o C and the monthly minimum average is 16.5 o C. The rainy season runs from October to March, including summer, and it averages 130 days of rain/ year.January and February are the months of larger pluviometric index.The mean relative air humidity is 77%, with August and September as the least humid months.
The monitoring was performed in five stations located at strategic places within the Municipal District perimeter.The choice of the sampling points in the monitoring evaluation was based on: first, the representativeness of areas with high population density or intense traffic; second, areas surrounding main pollution sources with the correct predominant wind direction.Using these criteria, four sampling points were selected, leeward (downwind) of the emission sources, while a point to windward (upwind) was chosen as a reference station.They were FEEMA (FE), Belmonte (BE), Retiro (RE), and Centro de Pesquisas stations (CP) to leeward and Aeroclube (AE) to windward (background).These stations are located in Figure 1. 1 The urban area of Volta Redonda is dominated mainly by metallurgic industries such as CSN (National Steel Company), FEM (Plant of Metallic Structures), CESBRA (Tin Company of Brazil), Voldac Casting, and cement industries like Tupy Cement and Votorantin Cement.These may also be considered as considerable emitting sources. 1 In the first evaluation (Dec/95-May/96), Belmonte, Retiro, Centro de Pesquisass and Aeroclube were equipped to measure the concentrations of total suspended particles (TSP) and Benzene, Toluene and Xylenes (BTX).FEEMA was able to measure TSP, BTX, sulphur dioxide (SO 2 ), wind speed and direction.In the second evaluation (Apr-May/ 99), TSP and metals in TPS (FEEMA, Belmonte, Retiro, Aeroclube and Conforto), inhalable particles (PM 10 ) and metals in PM 10 (FEEMA, Belmonte and Conforto), BTX (FEEMA and Retiro) and SO 2 (FEEMA) were collected and analyzed.Conforto substituted Centro de Pesquisas station.

Total suspended particles (TSP) and inhalable particles (PM 10 )
TSP was collected using a high-volume air sampler, Hi-Vol (Andersen, Smyrna, GA, USA) fitted with fiberglass filter of 8 x 10 inches, having an air flow of 1.1 to 1.7 m 3 min -1 for a continuous 24 h period.Particulate matter (PM), with a diameter up to 100 µm was retained in the filter.To analyze PM 10 a Hi-Vol sampler impactor capable of retaining particles ≤ 10 µm was used.TSP and PM 10 concentrations were gravimetrically determined.In the particulate matter, metals concentrations were determined in the second sampling.In the first evaluation (Dec/95-May/96), TSP sampling was effected each 24 h, 45 samples being collected.In the second evaluation (Apr-May/99), it was carried out every two days, corresponding to 10 samples.

Metals
The particles retained in the fiber glass filter (Hi-Vol) were submitted (1/6 of the filter) to a nitric acid 3 mol L -1 (Merck, Darmstadt, Germany) hot digestion.Metals dissolved in the sample were determined by Flame Atomic Absorption Spectrophotometry CG 904 model using hollow cathode lamp to each metal to be analyzed: Cadmium (Cd), Chromium (Cr), Copper (Cu), Iron (Fe), Lead (Pb), Manganese (Mn), Nickel (Ni), Vanadium (V) and Zinc (Zn). 17Blank filters and reagents were analyzed for each sampling for the determinations of possible contamination.The average metal content of the blanks represent less than 5% of the samples average metal content.Detection limit was 0.04 µg m -3 .Metals were sampled in a total of 11 days.

Benzene, toluene and xylenes (BTX)
Air was pumped (1 to 2 L min -1 ) through adsorbent tubes (Supelco ORBO 100, batch 6719, Bellefonte, PA, EUA), which contained two sections of activated charcoal.Mean volume of pumped air was 1.5 m 3 24h -1 .Solvent desorption with carbon disulfide (Tedia, Fairifield, OH, USA) was used. 2 µL of samples were injected into a 25 meter capillary column (CARBOWAX 20 M -HP-20M; 25 m x 0.2 mm, film thickness 0.2 µm) attached to a Hewlett-Packard 5890 GC, Series II with flame ionization detector (FID).Split mode 1:100 was used.Temperature of the injector and detector were 250 o C. The temperature of the column oven was maintained initially at 30 o C for 12 min, then increased at 25 o C per minute to 180 o C. BTX standard solutions were prepared from high purity standards obtained commercially (Riedel, Phillips and Ecibra).Dilutions were made with CS 2 to build a calibration curve with 5 different standard concentrations (13, 26, 52, 104, 208 mg L -1 ).The samples were injected in duplicate and blanks and standard solutions in triplicate.Blank values were around 5 mg L -1 .Standard deviations between sample injections were less than 10%.Dectection limit of method was 1 µg m -3 .BTX sampling, in the first evaluation (Dec/95-May/96), was realized continuously, during seven days every month, corresponding to 42 sampling days.In the second one, it was collected every two days, in a total of 11 days.In both evaluations, two samples were collected during 12 hours.

Sulphur dioxide (SO 2 )
During the campaign an automatic analyzer (Thermo Electron Corporation, Houston, TX, USA) was used continuously to measure the concentrations of sulphur dioxide.The employed methodology was based on fluorescence UV.The measurement of sulphur dioxide was registered every 10, 15 or 20 min and averaged for each hour.SO 2 was sampled for 122 days in the first evaluation (Dec/95-May/96) and 16 days in the second one (Apr-May/99).Calibration is made each 24h and annually by the manufacturer.

Meteorological data
In this first diagnostic phase, the daily synoptic charts of the schedule South America of 12:00 UTC (Universal Cardinale Teams) were used.They corresponded to the 9:00 hours local (in regular civil schedule), were elaborated and divulged by the Brazilian Navy Directorate of Hydrography and Navigation in agreement with information obtained from 4 meteorological stations from the Brazilian National Meteorological Institute (INMET), and the data from the National Center of Environmental Prediction (NCEP) of the United States.In the second monitoring evaluation, the results generated by the meteorological station at CSN were used.The wind regime came from the East, and varied from Northeast to Southeast. 1

Results and Discussion
In Brazil, the threshold limit for the majority of compounds in outdoor air is set by Conselho Nacional do Meio Ambiente (CONAMA). 18However, some pollutants still do not have any established air quality standards.In these cases, international standards (European Environmental Agency-EEA, 19 U.S. Environmental Protection Agency-USEPA 20 and World Health Organization -WHO 21 ) were adopted as reference.Due to the fact that the sampling period in both campaigns was less than one year, the results will be compared with 24 h standards suggested by the agencies; if not available the annual standard will be used.

Total suspended particles (TSP)
Table 1 presents averages, as well as minimum and maximum values of total suspended particles obtained during both evaluations in Volta Redonda and guideline values.All sampling sites in both evaluations did not exceed the daily limit for TSP according to CONAMA 18 (240 µg m -3 ) or EEA 19 (250 µg m -3 ).But, FEEMA station presented maximum concentration higher than the primary 24 h standard established by these agencies in both cases.
FEEMA and Belmonte stations presented the highest concentration averages, while Aeroclube presented the lowest in both evaluations.TSP average spatial distribution shows a clear increase from Northeast to Southwest (Figure 1).The isolines in Figure 1 refer to distribution of TSP concentrations in Volta Redonda according to wind direction.It can be stated that the gradient of TSP concentration is lower windward to CSN than leeward.The handling of iron ores and the burning of coal at CSN plant can origin the plume of pollution that reach FEEMA and Belmonte stations (leeward) but not reach Aeroclube (windward to CSN).Votorantin cement, which emits particulate matter in its process, seems not to influence TSP levels in Aeroclube station.Perhaps, the distance between station and plant is big enough to disperse or settle the particles.Low levels of TSP in this station were observed in both campaigns.But, Tupy cement can have some influence in PM mainly in FEEMA station due to their proximity.The highest levels (average and maximum) of TSP detected in both campaigns suggested this influence.
The average concentrations found for TSP in Volta Redonda (54-163 µg m -3 ) are lower than the average concentration found in polluted areas in Rio de Janeiro (68-198 µg m -3 ) according to FEEMA. 1 In another research, Azevedo et al. 9 found 27 to 348 µg m -3 TSP in industrial and heavy traffic areas in Rio de Janeiro.Dos Santos et al. 11 measured concentrations within 110-399 µg m -3 of TSP near the crop burning region in Campos dos Goytacazes, RJ, Brazil.In other cities (Table 2) average concentrations were 56-113 µg m -3 , i. e., lower than Volta Redonda.

Inhalable particles (PM 10 )
Inhalable particle concentrations were measured only in the second evaluation (Apr-May/99).Table 1 shows average, minimum and maximum concentrations of inhalable particles obtained for each monitoring station during the second campaign and suggested limits.The average concentrations for all stations do not exceed the national and international daily limits (150 µg m -3 ).Conforto registered the lowest concentrations that may probably be explained by the wind regime and the location of sampling sites.This fact confirms the results found for TSP.The ratio of PM 10 /TSP was greater than 0.5 for all points, highlighting the relevance of the PM 10 contribution to the TSP, that is, more than 50% correspond to PM 10 .In downtown Rio de Janeiro, average values within 57-100 µg m -3 PM 10 were detected, i.e., similar to the average values found in Volta Redonda (44-78 µg m -3 ). 1 According to Table 2, the concentrations found in other cities (24-96 µg m -3 ) are also similar.

Metals
Table 2 shows the concentrations of airborne trace metals at each station and international standard limits besides concentrations obtained in other cities.Zinc and iron presented the highest concentrations.Zinc concentration ranged from 2.5 to 7.0 µg m -3 in TSP and 0.006 to 4.3 µg m -3 in PM 10 , corresponding to 3-5% and 5-10% of metal concentration, respectively.Iron concentration ranged from 2.1 to 5.2 µg m -3 in TSP and 1.4 to 3.7 µg m -3 in PM 10 , representing 2-5% and 3-6% of metal concentration, respectively.Chromium and vanadium concentrations were below the detection limit.Lead concentrations did not exceed the daily standard established by EEA 19 (3 µg m -3 ) and USEPA 20 (1.5 µg m -3 quarterly average).For  manganese, if it was considered the annual WHO 21 standard for Europe (0.15 µg m -3 ) it was exceeded in all stations, except Aeroclube.Cd and Ni did not exceed the limit suggested by EEA 19 in TSP and PM 10 , 0.05 µg m -3 and 1 µg m -3 24h-average, respectively.Zn also did not surpass daily standard established by WHO 21 (100 µg m -3 ) in both kinds of particles.Cu and Fe do not have standards established by these agencies.
Half of airborne trace metals in Volta Redonda area are higher than average values reported for urban particulates.In PM 10 Cu, Fe, Mn and Zn and in TSP Fe, Mn and Zn presented average concentrations higher than other cities (Table 2).When compared to Brazilian levels, Fe concentrations found in Volta Redonda (2-5 µg m -3 in TSP and 1.5-4.0µg m -3 in PM 10 ) were similar to Rio de Janeiro 1 (1.5-3.0 µg m -3 in TSP and 1.5-2.0µg m -3 in PM 10 ).In the same way, zinc concentrations were 4.5-6.0µg m -3 in TSP in Rio de Janeiro 1 and 2.0-7.0 µg m -3 in TSP in Volta Redonda.For most of toxic trace metals (Pb, Cd, Zn, Ni , Cu) anthropogenic inputs are more important than natural sources. 25In Volta Redonda, anthropogenic sources seem to be responsible for the high metal concentrations, mostly industry related sources such as CSN (Al, Mn, Cr, Sn, Zn and mainly Fe), FEM (mainly Fe), CESBRA (Sb, Bi, Ni, Ag, Zn, Pb and mainly Sn) and Voldac (Mn, Cu and mainly Fe).The wind direction from Northeast to Southwest proves these results since that the highest metal concentrations were registered in Belmonte and FEEMA stations.It is believed that CESBRA and Voldac have minor influence in metal concentration in the stations due to distance, but they are also emitting.
These results are of some concern as it is well established that inhalable particles that have higher concentrations of many trace elements such Pb, Cd, Zn, Cr, Ni, Mn and Cu are correlated with pulmonary toxicity. 23,25nzene, toluene and xylenes (BTX) Table 1 presents the results of BTX levels obtained in the Municipal District of Volta Redonda in both evaluations.In the first evaluation (Dec/95-May/96) average concentrations for benzene, toluene, xylenes were 70 µg m -3 , 23 µg m -3 and 2.5 µg m -3 , respectively.In the second one (Apr-May/99), the arithmetic means were for benzene 23 µg m -3 , toluene 4 µg m -3 and xylenes 0.6 µg m -3 .Benzene concentrations in the second evaluation (Apr-May/99) presented values approximately four times lower than in the first one (1995-96).Toluene levels presented a reduction from five to seven times in this last sampling (1999).Xylenes were found in low levels in both evaluations.BTX in Volta Redonda and different cities are presented in Table 3. Reported average concentrations of benzene has been for South America cities 5-24 µg m -3 , Europe, North America and Oceania, < 5 µg m -3 and Asia, 7-31 µg m -3 (except Calcuta 28,555 µg m -3 ). 10,29enzene levels in Volta Redonda, in the first evaluation (Dec/95-May/96), were among the highest in cities.In the

Table 1 .
Average (range in parenthesis) concentrations of TSP, PM 10 , BTX and SO 2 during sampling period (Dec/95-May/96 and Apr-May/99) in Volta Redonda city and guideline limits for these compounds suggested by national and international agencies

Table 2 .
Heavy metals found in particulate matter in Volta Redonda and other cities and guideline values

Table 3 .
BTX average concentrations in Volta Redonda and other regions of the world nd, not detected; -, not determined.