Use of LUMEX and the conventional method to measure mercury concentration in ambient air at Caraguatatuba, São Paulo, Brazil. Paper 538 Luciana N Loureiro, Maria Cecilia G P Massa Maria Luiza B Tristão Ricardo Gutierres Petróleo Brasileiro S.A –Petrobras Centro de Pesquisas Leopoldo de Américo Miguez de Melo –CENPES Gerência de Avaliação e Monitoramento Ambiental –AMA Av. Horácio de Macedo, 950 –Ilha do Fundão –Rio de Janeiro –RJ - Brasil - CEP: 21941-915 [email protected] Luiz Drude de Lacerda Marcelo Dominguez de Almeida Laboratório de Estudos da Baixa Atmosfera Laboratório de Biogeoquímica Costeira Insituto de Ciências do Mar, Universidade Federal do Ceará Av. Abolição 3207, Fortaleza –CE - Brasil - CEP: 60.165-081

ABSTRACT The mercury (Hg) is a relatively abundant element in nature. Among the different chemical forms of mercury, the specie of major distribution is the elemental form, Hg0 (in gas phase), predominant in the atmosphere. The present paper summarizes the development of a methodology to effectively measure elemental mercury concentration in ambient air, applied in Caraguatatuba, São Paulo, Brazil. Within the development of this methodology were included many key aspects such as: preparation of a monitoring plan, selection of sampling sites, establishment of frequency campaigns, definition of sampling time and the location of automatic analyser. The developed methodology use an automatic analyser LUMEX RA915 +, operating at atomic absorption spectrometry that measures the concentrations of Hg0 in a certain period of time. In parallel, was used the conventional techniques of monitoring Hg 0 concentration, the manual gold trap method, for intercomparing of the methods. The main advantage of using an automatic analyser is avoiding subsequent analysis in the laboratory. The average concentration obtained during the campaigns was similar to the background air throughout uncontaminated areas. For all the selected sites in Caraguatatuba, the seasonal pattern of gaseous mercury was indicative of southern hemispherical background. The environmental behavior of Hg0, while exhibiting clear distinctions over diurnal and/or seasonal scales, was found to experience sharp changes for a very short time period, usually less than a few tens of seconds. The aplication of this methodology using an automatic analyser for monitoring Hg0concentration has presenting a potential for using in evaluations to understand the atmospheric behavior of Hg0 in several areas with reduction of time and costs.

1

INTRODUCTION Mercury (Hg) is a natural element of ubiquitous distribution in the planet (geoesphere, hydrosphere and biosphere) occurring in varying concentrations. The atmospheric mercury exists in the following forms: gaseous elemental mercury (Hg0, 90 to 99%), mercury bound to particulate matter (Hgp, <5%) and gaseous divalent mercury and reactive gaseous mercury (Hgr, <5%) 1,2,3,4. Elemental mercury gas (Hg0) is the most abundant form in the atmosphere, practically insoluble in water and low reactivity, which confers a long residence time in the atmosphere, allowing its transport over long distances. Due to this characteristic, mercury is considered a global pollutant 5 ,6, 7, 8, 9. The values of Hg0 in the northern hemisphere (~ 1.7 ng.m-3) tend to be higher than in the southern hemisphere (~ 1.2 ng.m-3) as a result of major natural and anthropogenic emissions of the northern hemisphere. Many authors consider 1.5 ng.m-3as an average base level of elemental mercury in the atmosphere 6,7, 8 ,9, 10. The environmental agencies sometimes request monitoring campaigns for determination of the background gaseous elemental mercury concentrations (Hg0) on the influence region of natural gas process units to issue permits for implementation and upgrades or processes modifications of industrial plants11. This study was carried through to evaluate the natural variability of elemental mercury concentration (Hg0) in the atmosphere on the influence region of Caraguatatuba Natural Gas Treatment Unit (UTCGA), in Caragruatatuba, a coastal municipality in the State of São Paulo, southeastern Brazil, before the beginning of the operation of the unit The present study aimed to develop a monitoring strategy included aspects such as: preparation of a monitoring plan, selection of sampling sites, establishment of frequency campaigns, definition of sampling time and the location of automatic analyser. Simultaneously, in two of the four monitoring points, to test the accuracy and reability of the real time in situ Hg0 measurements, the manual gold trap method, conventional technique of monitoring Hg 0 concentration, was used for comparison 10,11. The compilation and critical analysis of the results of four monitoring campaigns of Hg0 in the period of February to May 2008 was compareted with the results reported in the literature (for natural areas under similar anthropogenic influence in Brazil and abroad). The final results will subsidize the definition of the future monitoring campaigns in terms of scope and frequency sampling for monitoring the concentration of Hg0, during the operation period of UTGCA. The evaluation of the results obtained by monitoring the concentrations of elemental mercury (Hg°) is necessary to understanding the dynamic nature of this element in atmosfere. The main advantage of using an automatic analyser is not need sampling and subsequent analysis in the laboratory, allowing a high sampling frequency, what reduces the time and cost of this kind of monitoring campaigns.

STUDY SCOPE The definition of monitoring campaigns of gaseous elemental mercury concentrations (Hg0) was based on internationally recognized studies, which have procedures described in technical articles published in sources of high reliability and acceptability by the scientific community 12, 13, 14. Based on technical articles on the subject and the experience of the team involved in similar studies, each day of monitoring campaign should be occur in a continuous period of 8 to 12 hours, including periods of different sunlight intensity, because the concentration of Hg0 may be influenced by the degree of insolation and the variation in temperature associated 5, 15, 16, 17, 18, 19, 20. Moreover, shortterm changes in local weather conditions can also cause significant variations in the concentrations of Hg0 in the atmosphere, which must be monitored and considered as part of the natural of the local background 16. The predicting time for starting and ending of each day of monitoring 2

campaigns was 08: 00 a.m to 08:00 p.m, for four consecutive days in a single point of monitoring 16. The four campaigns were distributed over four months, covering the end of the rainy season (summer) and the beginning of dry period (autumn) in the region, in order to verify the behavior of elemental mercury in the atmosphere under the weather changes. The four monitoring points (1, 2, 3 and 4) was located in the city of Caraguatatuba. Table 1 shows the location of these monitoring points and the dates of the campaigns of sampling21. Table 1. Monitoring points and Campaigns dates

Monitoring Points

Localization (UTM)

Campaigns dates

1

451249

7385156

Day 1 Feb 14/ 2008

Day 2 Feb 15/2008

Day 3 Feb 16/2008

Day 4 Feb 17/2008

2

447943

7385234

Day 1 Mar 04/2008

Day 2 Mar 05/2008

Day 3 Mar 06/2008

Day 4 Mar 07/2008

3

455912

7384721

Day 1 Apr 23/ 2008

Day 2 Apr 24/ 2008

Day 3 Apr 25/ 2008

Day 4 Apr 26/ 2008

4

457404

7386710

Day 1 May 27/ 2008

Day 2 May 28/ 2008

Day 3 May 29/ 2008

Day 4 May 30/ 2008

STUDY AREA The monitoring campaigns were performed in the municipality of Caraguatuba, Brazil, situated at approximately 200 km far away from São Paulo City. The study area is situated close a hilly area and coastline. The study area was defined based on a previous atmospheric dispersion modelling study and involved rural and urban areas of Caraguatatuba 21. The location of four monitoring points (1, 2, 3 and 4) was the same of the quality air and meteorologic station as showed in figure 1. Figure 1 –The four monitoring points in Caraguatatuba, SP, Brazil.

3

METHODOLOGY The evaluation of the results obtained by monitoring the concentrations of elemental mercury (Hg0) is necessary to understanding the dynamic nature of this element in atmosphere 10. The traditional techniques for monitoring Hg0 (ASTM D-5954), in open air areas, uses sample, preconcentration and subsequent analysis in laboratory. The stage of pre-concentration can vary from minutes to several hours, depending on the combination of the following factors: Hg0 concentration, the number of cartridges available for monitoring and the equipament detection limit of the analyser used 10, 11. For these reasons, the monitoring of natural environments, which generally has low concentrations of Hg °, demands a high pre-concentration time, usually from 6 to 12 hours 16. Thus, if occur an increase in the concentration of Hg0 in a small part of the day, this increase would be diluted over time of sampling. Normally, it is not possible to determine the natural variations using this technique, in the subsequent interpretation of the results of monitoring. In regard to reduce the inherent difficulties of to use the traditional techniques, many types of equipment (analysers / detectors) to measure in real time and easy handling in the field have been developed10,16. The continuous analysers of Hg0 perform the automatic pre-concentration and detection in the same instrument, with appropriate limits of detection and frequency of sampling. These analysers usually use the technique of atomic fluorescence spectrometry (AFS) to achieve the sensitivity required. However, the performance of these analysers is strongly influenced by polyatomic molecules presents in air, making necessary the use of carrier gas with high degree of purity, typically argon, which dificult its use in environmental monitoring. In this study was used the analyser Lumex RA - 915 + that performs continuous measures of Hg, using atomic absorption spectrometry. In this analyser, the problems of interference with absorption in region of ultraviolet spectrum (such as some organic compounds and SO2) were solved by coupling a Zeeman Effect corrector. The analyser is portable, low power consumption, doesn’t require special carrier gases and has low limits of detection and high frequency of data acquisition (minimum time between sampling of 1 s) 10,11,16. The Lumex RA - 915 + analyser was interconnected to a portable microcomputer and programmed for monitoring, with records of concentration of Hg0 and environmental temperature. All procedures recommended by the manufacturer of equipment were carefully followed in daily routine work such as heating of the lamp of the equipment before each monitoring for a time of 30 minutes 10,11,16. Besides the storage of the measurements made by the analyser on the hard disk drive of 4

microcomputer, a spreadsheet was used for the field notes of any event that could contribute to any changes in the behavior profile of the concentrations of Hg0 observed during the monitoring. In the monitoring campaign the equipments of traditional technique and the Lumex analyser RA 915 + were set to about 1.5 meters of soil. The traditional technique for monitoring Hg0 was only used in points 3 and 4, due to difficulty logistic. The samples were colleted on traps (quartz tube with a mixture of gold). The air was kept at 1l/min using a small pump, for a period of 12 hours and the traps were set to same position of Lumex (1.5 meters of soil).

RESULTS AND DISCUSSION The main results of using the equipment LUMEX RA915 + and traditional technique were summarized in table 2 and 3. 0

-3

Table 2. Concentration of Hg (ng m ) obtained with LUMEX RA915 +

LUMEX

Monitoring campaigns

Day 1o. 2o. 3o. 4o. 1o. 2o. 3o. 4o. 1o. 2o. 3o. 4o. 1o. 2o. 3o. 4o.

1a.

2a.

3a.

4a.

Day Average Hg (ng m-3) 1.0 0.4 0.6 0.6 1.0 0.8 0.6 0.6 0.6 0.6 0.6 0.7 0.7 0.6 0.4 0.6

Std.Dev. Hg (ng m-3) 0.7 0.4 0.5 0.5 0.5 0.5 0.4 0.4 0.5 0.5 0.5 0.5 0.5 0.5 0.3 0.6

Table 3. Concentration of Hg 0 (ng m-3) obtained with the traditional technique

Traditional technique

Monitoring campaigns 3

Day 1o.

Day Average Hg (ng m-3) 0.60

Std.Dev. Hg (ng m-3) 0.01 5

4

2o.

1.30

0.04

3o.

0.80

0.08

4o.

0.70

0.01

1o.

0.90

0.07

2o.

1.00

0.08

3o.

0.50

0.12

4o.

1.00

0.24

The Hg0 concentrations varied between 0.4 to 1.0 ng.m-3 and from 0.5 to 1.3 ng.m-3, using the LUMEX and the gold cartridge method respectively. The environmental behavior of Hg 0 did not exhibite clear distinctions over seasonal scales (summer in the 1 st and 2 nd campaigns and autumn in the 3 rd and 4 rd campaigns) in the region. After observation Hg 0 concentration results from Lumex througt a day (approximatively 20000 results), it is possible to observe biggest results in the period of morning (probably because degree of insolation and the variation in temperature associated) compared with evening.

CONCLUSION The measurement of elemental mercury concentration on the atmosphere in influence region of UTCGA gives a preliminary identification of the behavior this element in the Caraguatatuba’s atmosphere. The application of exposed methodology for determination of Hg0 concentration in parallel with the conventional was an opportunity to compare results obtained from these different methods. And the results suggest that it is possible to replace the conventional method, resulting in optimizing the response time and logistics of sampling. The development of this methodology for continuous monitoring of the concentration of Hg0 in ambient air (using equipment LUMEX) is necessary and included issues with great relevance such: Ø Setting the period of campaign, Ø Determination of the appropriate period and local for measurement; Ø Ideal conditions for assembly and installation of equipment in the monitoring point; The comparison between the results obtained in these 4 monitoring points with other similar studies, reported in the scientific community, shows that the results found are similar with the values expected for uncontaminated areas (the southern hemisphere ~ 1.2 ng.m-3), showing a daily average ranging from 0.4 to 1.0 ng m-3 and maximum values ranging from 1.11 to 5.22 ng m-3, using the continuous analyzer. The using of gold trap method results the daily average concentration ranging from 0.5 to 1.3 ng m-3. The results demonstrated a good intercomparision between using continuous analyser and gold trap method. The development of this strategy of using an automatic analyser for monitoring Hg0 concentration showed consistent results under field conditions, confirming its potential for using to understand the atmospheric behavior of Hg0 with significant reduction of time and costs. These campaigns were the first initiative to show the background of the area as the concentration of elemental mercury. In general, the relations of the measured Hg0 concentrations with meteorological and air quality parameters not allowed the identification of direct relations with the concentration of elemental mercury in air. What suggest the using of longer time series that to allow an assessment of interannual variations of these parameters.

6

For all the selected sites in Caraguatatuba, the seasonal pattern of gaseous mercury was indicative of southern hemispherical background and the environmental behavior of Hg0 doesn´t exhibit good distinctions over seasonal scales during the monitoring period of time.

REFERENCES 1. Abbot, M. L., Lin, C. J., Martian, P., and Einerson, J. J., 2008. Atmospheric mercury near Salmon Falls Creek Reservoir in southern Idaho. Applied Geochemistry 23, 438-453. 2. Lohman, K., Seigneur, C., Gustin, M., and Lindberg, S., 2008. Sensitivy of the global atmospheric cycle of mercury to emissions. Applied Geochemistry 23, 454-466. 3. Kim, K. H., Ebinghaus, R., Schroeder, W. H., Blanchard, P., Kock, H. H., Steffen, A., Froude, F. A., Kim, M. Y., Hong, S. M., and Kim, J. H., 2005. Atmospheric mercury concentrations from several observatory sites in the northern hemisphere. Journal of Atmospheric Chemistry 50, 1-24. 4. Lindberg, S., Bullock, R., Ebinghaus, R., Engstrom, D., Feng, X. B., Fitzgerald, W., Pirrone, N., Prestbo, E., and Seigneur, C., 2007. A synthesis of progress and uncertainties in attributing the sources of mercury in deposition. Ambio 36, 19-32. 5. Almeida, M. D., Marins, R. V., Paraquetti, H. H. M., and Lacerda, L. D., 2008. Methodology optimization and application for measure of gaseous elemental mercury in the Amazon Atmosphere. Journal of the Brazilian Chemical Society in press 6. Slemr, F. and Langer, E., 1992. Increase in Global Atmospheric Concentrations of Mercury Inferred from Measurements over the Atlantic-Ocean. Nature 355, 434-437. 7. Temme, C., Blanchard, P., Steffen, A., Banic, C., Beauchamp, S., Poissant, L., Tordon, R., and Wiens, B., 2007. Trend, seasonal and multivariate analysis study of total gaseous mercury data from the Canadian atmospheric mercury measurement network (CAMNet). Atmospheric Environment 41, 5423-5441. 8. Temme, C., Slemr, F., Ebinghaus, R., and Einax, J. W., 2003. Distribution of mercury over the Atlantic Ocean in 1996 and 1999-2001. Atmospheric Environment 37, 1889-1897. 9. Wangberg, I., Munthe, J., Pirrone, N., Iverfeldt, A., Bahlman, E., Costa, P., Ebinghaus, R., Feng, X., Ferrara, R., Gardfeldt, K., Kock, H., Lanzillotta, E., Mamane, Y., Mas, F., Melamed, E., Osnat, Y., Prestbo, E., Sommar, J., Schmolke, S., Spain, G., Sprovieri, F., and Tuncel, G., 2001. Atmospheric mercury distribution in Northern Europe and in the Mediterranean region. Atmospheric Environment 35, 3019-3025. 10. Munthe, J., Wängberg, I., Pirrone, N., Iverfeldt, R., Ferrara, R., Ebinghaus, R., Feng,X., Ferrara, R., Gardfeldt, K., Keeler, G., Lanzillotta, E., Lindberg, S.E., Lu, J., Mamane, Y., Prestbo, E., Schmolke, S., Schroeder, W.H., Sommar, J., Sprovieri, F., Stevens, R.K., Stratton, W., Tuncel, G. and Urba, A., 2001. Intercomparison of methods for sampling and analysis of atmospheric mercury species. Atmospheric Environment 35, 3007-3017. 11. Spiric, Z. and Mashyanov, N. R., 2000. Mercury measurements in ambient air near natural gas processing facilities. Fresenius Journal of Analytical Chemistry 366, 429-432. 12. Engle, M. A., Tate, M. T., Krabbenhoft, D. P., Kolker, A., Olson, M. L., Edgerton, E. S., Dewild, J. F., and McPherson, A. K., 2008. Characterization and cycling of atmospheric mercury along the central US Gulf Coast. Applied Geochemistry 23, 419-437. 13. EPA, 1997. Mercury Report to Congress, Vol VI: Characterization of Human Health and Wildlife Risks from anthropogenic Mercury Emissions in the United States. U.S. Environmental Protection Agency, Washington 14. Fadini, P. S. and Jardim, W. F., 2001. Is the Negro River Basin (Amazon) impacted by naturally occurring mercury? Science of the Total Environment 275, 71-82. 15. Fostier, A. H. and Michelazzo, P. A. M., 2006. Gaseous and particulate atmospheric mercury concentrations in the Campinas Metropolitan Region (Sao Paulo State, Brazil). Journal of the Brazilian Chemical Society 17, 886-894. 16. Gosar, M., Pirc, S., Sajn, R., Bidovec, M., Mashyanov, N. R., and Sholupov, S. E., 1997. 7

Distribution of mercury in the atmosphere over Idrija, Slovenia. Environmental Geochemistry and Health 19, 101-110. 17. Lacerda, L.D.; Guerra, T.; Castilhos, Z.C.; Hatje, V.; Canesin, F.P. & Cunha, L.C. 1995. Estimated heavy metals emissions to the atmosphere from energy generation in Brazil. Proc. 10th International Conf. on Heavy Metals in the Environment, Hamburgo, 2:81-83. 18. Magarelli, G. and Fostier, A. H., 2005. Influence of deforestation on the mercury air/soil exchange in the Negro River Basin, Amazon. Atmospheric Environment 39, 7518-7528. 19. Amouroux, D., Wasserman, J. C., Tessier, E., and Donard, O. F. X., 1999. Elemental mercury in the atmosphere of a tropical Amazonian forest (French Guiana). Environmental Science & Technology 33, 3044-3048. 20. Artaxo, P., de Campos, R. C., Fernandes, E. T., Martins, J. V., Xiao, Z. F., Lindqvist, O., Fernandez-Jimenez, M. T., and Maenhaut, W., 2000. Large scale mercury and trace element measurements in the Amazon basin. Atmospheric Environment 34, 4085-4096. 21. Environmental Basic Plan - Review 1st , BIODINÂMICA , Rio de Janeiro, 2008.

ACKNOWLEDGMENTS The authors wish to express their gratitude to the Petrobras technicians Bruno Rodrigues de Moura e Mauricio Guerrante for close collaboration and incentive throughout this study.

KEYWORDS Mercury, atmosphere, monitoring campaign, Trap, automated analyzer and Lumex.

8