Environ. Sci. Technol. 2006, 40, 3163-3167
Observations of Mercury-Containing Aerosols D . M . M U R P H Y , * ,† P . K . H U D S O N , †,‡,| D . S . T H O M S O N , †,‡ P . J . S H E R I D A N , †,§ A N D J . C . W I L S O N § Earth System Research Laboratory, National Oceanic and Atmospheric Administration, Boulder, Colorado 80305-3337, Cooperative Institute for Research in Environmental Sciences, University of Colorado, Boulder, Colorado 80309-0216, and University of Denver, Denver, Colorado 80208
In situ analyses with a laser ionization mass spectrometer have shown that a large fraction of aerosols in the bottom few kilometers of the stratosphere contain small amounts of mercury (1). Electron microscopy of particles collected near the tropopause has also detected mercury. The distribution of mercury onto many particles, including those less than 20 nm in diameter, indicates that the mercury is from local condensation of mercury compounds onto particles rather than transport of mercury-rich aerosols from surface sources. Although the results are only semiquantitative, they suggest that most of the mercury in the lower stratosphere is converted into the particulate phase. Mercury-containing particles were present at both middle latitudes and the tropics in two seasons. There is therefore good reason to believe that particulate mercury above the tropopause is global and could affect the atmospheric lifetime of mercury. There are indications that bromine and/ or iodine may be involved in the conversion of mercury from the gas to particle phase. Measurements at altitudes below 5 km did not find mercury in any particles despite sampling some particles that clearly originated in the stratosphere. This indicates that the particulate mercury from the lower stratosphere may be volatile enough to evaporate or decompose once particles reach warmer temperatures.
1. Introduction Mercury is of interest as a pollutant on a global scale. It is present in the atmosphere at about 1-5 ng m-3, or a mixing ratio of a fraction of a part per trillion by volume (pptv) (25). In many places, over 90% of this mercury is present as gas-phase elemental mercury (2). The small fraction of mercury attached to particles is still important, however, because the atmospheric lifetime of particulate mercury is much shorter than that of elemental gas-phase Hg (2, 3). In a previous work we presented mass spectrometer data showing that a large number of particles in the lowermost stratosphere contain small amounts of Hg (1). We have now extended these observations to other seasons. We also have new observations at lower altitudes. These extended observations can be analyzed to place bounds on the amount of * Corresponding author phone: (303)497-5640; fax: (303)497-5373; e-mail:
[email protected]. † National Oceanic and Atmospheric Administration. ‡ University of Colorado. § University of Denver. || Present address: University of Iowa, Iowa City, IA 52242. 10.1021/es052385x CCC: $33.50 Published on Web 04/07/2006
2006 American Chemical Society
FIGURE 1. Sample mass spectrum containing mercury. This particular particle was analyzed about 300 m above the tropopause at 25.3° N, 95° W at an ambient temperature of 202 K. The inset shows a detail of the data for the Hg ions along with the isotopic pattern of Hg convolved with the resolution of the mass spectrometer (offset for clarity). Of the Hg-containing particles measured during this mission, about 40% were very similar to this example, about 35% were very similar except for larger sulfate and/or NO+ peaks, about 10% were similar except they also contained a large potassium peak, and about 15% had other patterns. Hg condensed onto particles in the lower stratosphere. Finally, we present an electron microscopy analysis of a stratospheric sample that shows that Hg was also present on particles smaller than those sampled by the mass spectrometer.
2. Observations The mass spectra of over 70 000 individual particles have been obtained with the Particle Analysis by Laser Mass Spectrometry (PALMS) instrument aboard the NASA WB57F aircraft (6). In addition, the mass spectra of over 900 000 individual particles have been obtained at lower altitudes from the NOAA P3 aircraft. The PALMS instrument obtains a positive or negative mass spectrum from individual particles larger than about 200 nm in diameter. Analysis is done on the aircraft with the particles spending 2-150 ms in the inlet and about 1 ms in a vacuum. Separate collections of particles for later electron microscopy analysis were carried out from an ER-2 aircraft (7). Positive ion mass spectra of particles near and just above the tropopause consistently showed small peaks due to Hg (Figure 1). Mercury was almost always observed in particles with sulfates, organics, and iodine, as in Figure 1. Because sulfate does not readily produce positive ions, its abundance in the particle is greater than indicated by relative the size of the SO+ peak. Although mercury was often found in the lower stratosphere, it was only rarely observed in the relatively pure sulfuric acid particles characteristic of the main stratospheric aerosol layer. Mercury was present as a small peak in many mass spectra. At the altitudes of peak abundance just above the tropopause it was present in the majority of the mass spectra. In contrast, Hg was never unambiguously observed in about 200 000 positive ion mass spectra acquired from the NOAA P3 primarily over the northeastern United States in summer 2004. Nor was any Hg clearly observed in a similar number VOL. 40, NO. 10, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3163
FIGURE 3. Vertical profiles of the average ion current for Hg measured using single particle analysis during the 1998 WB-57F aerosol mission. Each point is the average of 310-450 spectra, except the P3 data point that indicates that no unambiguous mercury was found in about 200 000 spectra. To calculate valid averages of such small signals, the mass spectra were required to have a total ion current at least 2000 times the detection limit. To avoid large amounts of water and nitric acid condensation, the spectra were required to be in unsaturated air above 193 K.
FIGURE 2. Electron microscope image and elemental analysis of sample 19930514_14 collected at 15.2° N, 188° W, 17.1 km. The local tropopause was about 17.3 km. There was a bimodal size distribution of particles. Larger particles were mostly sulfate with lesser amounts of C and trace elements. Smaller particles (about 10 nm in diameter, dots in this figure) contained a significant amount of Hg, as shown by part b, which shows a summation spectrum from 10 of the small particles. Discrete Hg peaks are visible at 10.00 and 11.85 keV (Lr1 and Lβ1+2, respectively). The Mr1 peak (2.19 keV) is visible as a shoulder of the S Kr peak at 2.31 keV, and a very small Lγ1 peak is visible at 13.83 keV. The Cu peaks are from the TEM grid, and the Si peak is probably from the Si EDS detector. S, Fe, C, and O appear to be in these small particles, although the C and O peaks probably have some X-ray contribution from the thin film. of spectra taken over California and the nearby Pacific in 2002, although we rely more on the 2004 spectra because they had better signal-to-noise ratios. An independent measurement of Hg-containing particles from a single lower-stratospheric sample was performed in 1993. An aerosol sample, characterized by relatively high concentrations of very small particles, was collected by impaction on a thin carbon film and was analyzed using two scanning transmission electron microscopes (STEMs). The first electron microscope used a tungsten filament as an electron source and was not capable of obtaining useful X-ray composition spectra of the smaller (sub-50-nm) particles. The second microscope utilized a field-emission electron source capable of roughly an order of magnitude increase in the electron beam power density. This instrument (Vacuum Generators HB-5 field emission gun STEM coupled to a Noran windowless Si EDS detector) was capable of providing images 3164
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 10, 2006
and X-ray spectra of the population of very small (∼10 nm in diameter) particles in this sample. Figure 2a shows a darkfield STEM image of the bimodal particle distribution. The larger particles are composed predominantly of sulfate, with C, Fe, and other trace elements often present. The smaller particles were largely sulfates but also contained significant amounts (e.g., a few weight percent) of Hg. The spectrum shown in Figure 2b is a summation spectrum formed from the brief analyses of 10 of the small particles. A summation spectrum is presented because the high-intensity electron beam heats and volatilizes these particles within a few seconds, so this spectrum should be more representative of the original particles. The detection of particulate Hg in the same atmospheric altitude range by two independent analysis techniques aboard different aircraft is strong evidence that the layer of particulate Hg is real and not due to contamination. The electron microscope and PALMS can only measure Hg on different size particles because PALMS cannot detect particles smaller than 120 nm and the electron microscope is not sensitive enough to detect the small amount of Hg in the larger particles measured by PALMS. Further evidence is that the Hg signal measured by PALMS is consistently found at the same altitude relative to the tropopause and not at the same absolute pressure, temperature, time into a flight, or any other variable that would have a closer relationship to plausible contamination. The layer has been measured with two different inlets into the vacuum system: a glass-lined capillary during 19982000 and an aerodynamic focusing inlet in 2004. Figure 3 displays vertical profiles of the mercury signal. The absolute altitudes of the tropopause were about 4 km higher for the tropical points (10-25° N) than for the midlatitude points (28-45° N). The vertical profiles are much better correlated with the tropopause than with absolute altitude, indicating that the processes forming the layers had
FIGURE 4. Vertical profiles of estimated bromine and iodine peaks for the 1998 WB-57F aerosol mission. Each point is an average of 500 spectra with restrictions similar to Figure 3. to do with the tropopause or lowermost stratosphere rather than some pressure or altitude. Figure 4 shows vertical profiles for bromine and iodine. Br- displays similar vertical profiles to Hg+ (8). Iodine produces both positive and negative ions and so helps connect the vertical profiles of bromine, best measured in negative ion mode, with the Hg-containing particles measured in positive ion mode. Figure 5 shows that iodine and mercury were often present in the same positive ion mass spectra. Iodine and bromine were often present in the same negative ion mass spectra. This, together with the similar vertical profiles of Hg+ and Br-, suggests that mercury and bromine were present in the same particles whether or not they were chemically bound. The abundance and vertical profile of Hg ions in the mass spectra were similar on flights conducted from Houston in April-May 1998, in September 1999, and in January 2004. Mercury was detected in the lowermost stratosphere over the entire latitude range covered by the flights (approximately 5-45° N). The existence over such a wide range of latitudes and seasons suggests that the layer of Hg-containing particles may be global in extent.
3. Mercury Budget We estimate that the mass spectra indicate the conversion of a substantial fraction of the gas-phase Hg to particulate form in the lower stratosphere. To make this estimate, we assume that the mass mixing ratio of total Hg in the lower stratosphere is similar to that in the remote lower troposphere, which contains about 2 ng m-3 in the Northern Hemisphere (2), or about 1.7 parts per trillion by mass (pptm). The lifetime of mercury is long enough that total Hg should be well mixed vertically in the troposphere (9). Measured aerosol size distributions while PALMS detected Hg (10) showed about 1200 pptm of aerosol, assuming a density of 1.4 g cm-3 for the mixture of organics and sulfate that is typical of that region (1). Thus, conversion of all Hg in the lower stratosphere to particulate form would yield about 0.1% by mass in the particles, or about 0.05% by mole (the atomic weight of Hg is about twice that of other species in the particles such as sulfuric acid).
FIGURE 5. Estimated iodine and mercury peaks in single particles. Each point is a 100 particle average after sorting by the iodine signal, which was taken as mass 127 above mass 125-129. Zeros are plotted at 0.01 to make them visible on the log scales. The vast majority of the P3 data had zero for both peaks, but some had iodine with no Hg. Inspection of the mass spectra shows that the one nonzero mercury point in the P3 data was actually from other ions at mass 202. Clearly, the approximately 0.2% Hg ion current in the lower stratosphere (Figure 2) exceeds 0.05%, so it is possible that essentially all of the Hg in the lower stratosphere is converted to particulate form. In addition, Hg must ionize more easily than other species in the particles. Just how much more easily is difficult to assess without laboratory experiments that we have not conducted because of the possible toxicity of aerosolized Hg compounds. Our mass spectrometer has been calibrated for Na, Ni, Fe, and other meteoric metals in sulfuric acid (11). Even if it ionized as well as Na, 5-30% of the Hg would need to be in particulate form to explain our data. However, Hg has a very high ionization potential (10.4 eV), so it is unlikely that it is ionized as well as Na. The ionization efficiency of Hg(NO3)2 in sucrose solutions was investigated by Kaufmann using a LAMMA laser ionization mass spectrometer (12). He found that Hg ionized roughly 50 times less efficiently than Fe. This result must be applied cautiously to our work because the LAMMA instrument used a different laser wavelength and energy density than our mass spectrometer. Still, the use of an ionization efficiency of a factor of 50 less than Fe would require condensation of all of the gas-phase Hg to account for the measured Hg+ peaks. A lower limit on the amount of Hg can be obtained from our failure to detect Hg in any particles at low altitudes. The amount of Hg on particles in the lower stratosphere must be much larger than the approximately 1% present on particles near the surface (2). Besides the mass spectrum, we can estimate the size of each particle from the amount of light it scatters and, more recently, from the velocity it acquires as it transits the air jet into the vacuum. Both measures show that size distributions of Hg-containing particles measured by PALMS were reasonably similar to other particles in the lower stratosphere. VOL. 40, NO. 10, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3165
4. Discussion
5. Conclusions and Testable Predictions
There are several lines of evidence suggesting that the particulate Hg in the lower stratosphere originated locally from the gas phase rather than being transported from sources of mercury-rich particles. No single distant source could explain the presence of Hg on a majority of the particles just above the tropopause. The lack of Hg-containing particles at P3 aircraft altitudes is inconsistent with large surface sources of particulate Hg. The relatively high Hg content observed by electron microscopy on small particles, which have large surface-to-volume ratios, also suggests that the Hg condensed from the gas phase. Such small particles also have fairly short lifetimes before coagulation. Although the difference is not large, the maximum Hg signal was clearly observed in the stratosphere and not at the altitude with the coldest temperature (Figure 3). This is consistent with a chemical reaction rather than condensation at the coldest point. Although an enhanced solubility at low temperatures cannot be ruled out, the vapor pressure of elemental Hg is too large for simple condensation at even the coldest tropopauses. A possible reason for the larger value of particulate Hg in the stratosphere compared to the troposphere is the relative residence time of particles there. Particles in the upper troposphere are subject to scavenging and precipitation by clouds whereas particles in the lower stratosphere have no such loss mechanism. This, together with the slow exchange of air across the tropopause, results in a change in estimated particle lifetime from an average of about 1 to 3 weeks in the lower to upper troposphere (13) to a year in the stratosphere above 20 km (14). To produce an atmospheric lifetime of 0.5-2 years (9, 15-17), any reactions that convert gas-phase Hg0 to particulate HgII must be either slow or very localized. A slow reaction would not have time to proceed very far in the weeks that an average particle is in the troposphere but could convert much of the mercury in the lower stratosphere to particulate form. Another reason for the presence of Hgcontaining particles in the stratosphere could be the increased abundance of O3 and reactive Br, both of which have been implicated in the conversion of gas to particulate Hg (1820). Calvert and Lindberg (21) have questioned the validity of the Hg + O3 reaction. The decrease of the mercury signal more than 5 km above the tropopause (Figure 3) has a number of possible explanations. The simplest is dilution of the Hg-containing aerosol by sulfuric acid from stratospheric oxidation of SO2 and carbonyl sulfide (22), followed by sedimentation of such particles. It is also possible that the ionization efficiency of Hg by our PALMS instrument is higher in mixed organic/ sulfate particles than in pure sulfuric particles. Mixed organic particles become less common higher in the stratosphere. A speculative implication of these measurements is that the anthropogenic increase in atmospheric mercury might have increased the rate of growth of new particles formed near the tropical tropopause (23). Although the overall amount of mercury is much less than 1% of the total mass of particles in the lower stratosphere, newly formed particles have very high surface-to-volume ratios, and therefore anything condensing from the gas phase can strongly influence their growth. Figure 2 provides some support for the idea that the growth of particles less than 20 nm in diameter was influenced by the availability of Hg. The importance of this would be that the number of newly formed particles that survive to large sizes is set by a competition between coagulation and growth to larger sizes that coagulate less readily. The long-term anthropogenic changes in both atmospheric Hg (15) and the amount of reactive Br (24) could therefore conceivably affect the number distribution of particles in the stratosphere.
An important outstanding question is if conversion of Hg into particulate form in the lower stratosphere affects the global Hg lifetime. For comparison, the gas-phase species CH3Br, with a sharp vertical gradient in the bottom 5 km of the stratosphere, has a global lifetime of less than 2 years (24). Two conditions must be satisfied if stratospheric conversion onto particles is to be an important global sink of gas-phase Hg. First, as seems likely, the conversion must be a substantial fraction of the available Hg in the lower stratosphere. Second and less likely, the conversion to particulate form must be long lasting. If the Hg is volatile enough so that it evaporates as soon as the particles warm to temperatures characteristic of the lower troposphere, then the overall sink will be minimal. Likewise, the sink would be unimportant if the particulate Hg is released upon contact with water in a cloud. The lack of clear signals of Hg in particles at low altitudes suggests that the particulate Hg formed in the lower stratosphere does not survive long in the lower troposphere. Of the over 200 000 suitable positive ion mass spectra acquired from the P3 in 2004, dozens showed elemental signatures clearly from meteoric material in the stratosphere (1, 11). There were hundreds that had organic, bromine, and iodine signals essentially identical to the spectrum shown in Figure 1, but without the Hg. These spectra show that the sample size was large enough to see the small contribution of stratospheric particles to the lower troposphere. The absence of Hg is most easily explained by evaporation at the warmer temperatures in the lower troposphere. From these semiquantitative measurements of particulate Hg in the lower stratosphere, we make two testable predictions. First, there should be a measurable depletion of gasphase elemental Hg just above the tropopause. Second, thunderstorms that penetrate the tropopause should occasionally scavenge Hg from this elevated layer and produce excess Hg in precipitation. A similar phenomenon was observed with occasional deposition of radioactive iodine from bomb tests (25). The analogy to iodine is particularly strong because it often occurs in the same particles as Hg (Figure 5). At ground level, a modest correlation (r2 ) 0.54) of particulate Hg was observed with 7Be, which is a tracer of the stratosphere and upper troposphere (9). Some evidence exists for high Hg concentrations in precipitation from intense Florida thunderstorms (26), but the data are neither extensive enough nor are there enough independent measures of thunderstorm height to prove that the Hg came from the lower stratosphere.
3166
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 10, 2006
Literature Cited (1) Murphy, D. M.; Thomson, D. S.; Mahoney, M. J. In situ measurements of organics, meteoritic material, mercury, and other elements in aerosols at 5 to 19 kilometers. Science 1998, 282, 1664-1669. (2) Schroeder, W. H.; Munthe, J. Atmospheric MercurysAn overview. Atmos. Environ. 1998, 5, 809-822. (3) Slemr, F.; Schuster, G.; Seiler, W. Distribution, speciation and budget of atmospheric mercury. J. Atmos. Chem. 1985, 3, 407434. (4) Kim, K.-H.; Ebinghaus, R.; Schroeder, W. H.; Blanchard, P.; Kock, H. H.; Steffen, A.; Froude, F. A.; Kim, M.-Y.; Hong, S.; Kim, J. H. Atmospheric mercury concentrations from several observatory sites in the Northern Hemisphere. J. Atmos. Chem. 2005, 50, 1-24. (5) Kock, H. H.; Bieber, E.; Ebinghaus, R.; Spain, T. G.; Thees, B. Comparison of long-term trends and seasonal variations of atmospheric mercury concentrations at the two European coastal monitoring stations Mace Head, Ireland, and Zingst, Germany. Atmos. Environ. 2005, 39, 7549-7556. (6) Thomson, D. S.; Schein, M. E.; Murphy, D. M. Particle analysis by laser mass spectrometry WB-57F instrument overview. Aerosol Sci. Technol. 2000, 33, 153-169.
(7) Sheridan, P. J.; Brock, C., A.; Wilson, J. C. Aerosol particles in the upper troposphere and lower stratosphere: elemental composition and morphology of individual particles in northern midlatitudes. Geophys. Res. Lett. 1994, 21, 2587-2590. (8) Murphy, D. M.; Thomson, D. S. Halogen ions and NO+ in the mass spectra of aerosols in the upper troposphere and lower stratosphere. Geophys. Res. Lett. 2000, 27, 3217-3220. (9) Lamborg, C. H.; Fitzgerald, W. F.; Graustein, W. C.; Turekian, K. K. An examination of the atmospheric chemistry of mercury using 210Pb and 7Be. J. Atmos. Chem. 2000, 36, 325-338. (10) Jonsson, H. H.; Wilson, J. C.; Brock, C. A.; Knollenberg, R. G.; Newton, R.; Dye, J. E.; Baumgardner, D.; Borrmans, S.; Ferry, G. V.; Pueschel, R.; Woods, D. C.; Pitts, M. C. Performance of a focused cavity aerosol spectrometer for measurements in the stratosphere of particle-size in the 0.06-2.0 µm diameter range. J. Atmos. Oceanic Technol. 1995, 12, 115-129. (11) Cziczo, D. J.; Thomson, D. S.; Murphy, D. M. Ablation, flux, and atmospheric implications of meteors inferred from stratospheric aerosol. Science 2001, 291, 1772-1775. (12) Kaufmann, R. L. Laser-microprobe mass spectroscopy (LAMMA) of particulate matter. In Physical and Chemical Characterization of Individual Aerosol Particles; Spurny, K. R., Ed.; Ellis Horwood Limited: Chichester, U. K., 1986. (13) Koch, D. M.; Jacob, D. J.; Graustein, W. C. Vertical transport of tropospheric aerosols as indicated by 7Be and 210Pb in a chemical tracer model. J. Geophys. Res. 1996, 101, 18651-18666. (14) Godin, S.; Poole, L. R.; Bekki, S.; Deshler, T.; Larsen, N.; Peter, T. Global distributions and changes in stratospheric particles. In Scientific Assessment of Ozone Depletion; Report #44; World Meteorological Organization: Geneva, Switzerland, 1999; Chapter 3, pp 3.1-3.40. (15) Bergan, T.; Gallardo, L.; Rodhe, H. Mercury in the global troposphere, a three-dimensional model study. Atmos. Environ. 1999, 31, 1575-1585. (16) Shia, R.-L.; Seigneur, C.; Pai, P.; Ko, M.; Sze, N. D. Global simulation of atmospheric mercury concentrations and deposition fluxes. J. Geophys. Res. 1999, 104, 23747-23760. (17) Tokos, J. J. S.; Hall, B.; Calhoun, J. A.; Prestbo, E. M. Homogeneous gas-phase reactions of Hg0 with H2O2, O3, CH3I, and (CH3)2S: Implications for atmospheric Hg cycling. Atmos. Environ. 1998, 32, 823-827.
(18) Mason, R. P.; Sheu, G.-R. Role of the ocean in the global mercury cycle, Global Biogeochem. Cycles 2002, 16 (4), 1093. (19) Schroeder, W. J.; Anlauf, K. G.; Barrie, L. A.; Lu, J. Y.; Steffen, A.; Schneeberger, D. R.; Berg, T. Arctic springtime depletion of mercury. Nature 1998, 394, 331-332. (20) Boudries, H.; Bottenheim, J. W. Cl and Br atom concentrations during a surface boundary layer ozone depletion event in the Canadian high Arctic. Geophys. Res. Lett. 2000, 27, 517-520. (21) Calvert, J. G.; Lindberg, S. E. Mechanisms of mercury removal by O3 and OH in the atmosphere. Atmos. Environ. 2005, 29, 3355-3367. (22) Weisenstein, D. K.; Yue, G. K.; Ko, M. K. W.; Sze, N. D.; Rodriguez, J. M.; Scott, C. J. A two-dimensional model of sulfur species and aerosols. J. Geophys. Res. 1997, 102, 13019-13035. (23) Brock, C. A.; Hamill, P.; Wilson, J. C.; Jonsson, H. H.; Chan, K. R. Particle formation in the upper tropical tropospheresA source of nuclei for the stratospheric aerosol. Science 1995, 270, 6501653. (24) Kurylo, M. J.; Rodriguez, J. M.; Andreae, M. O.; Atlas, E. L.; Blake, D. R.; Butler, J. H.; Lal, S.; Lary, D. J.; Midgley, P. M.; Montzka, S. A.; Novelli, P. C.; Reeves, C. E.; Simmonds, P. G.; Steele, L. P.; Sturges, W. T.; Weiss, R. F.; Yokouchi, Y. Short-lived ozonerelated compounds. In Scientific Assessment of Ozone Depletion; Report #44; World Meteorological Organization: Geneva, Switzerland, 1999; Chapter 2, pp 2.1-2.56. (25) Reiter, E. R. Atmospheric Transport Processes, Part 4: Radioactive Tracers; TIC-217114; Technical Information Center, U. S. Department of Energy: Springfield, VA, 1978. (26) Guentzel, J. L.; Landing, W. M.; Gill, G. A.; Pollman, C. D. Atmospheric deposition of mercury in Florida: The FAMS project (1992-1994). Water, Air, Soil Pollut. 1995, 80, 393-402.
Received for review November 28, 2005. Revised manuscript received March 2, 2006. Accepted March 13, 2006. ES052385X
VOL. 40, NO. 10, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3167
