- Email: [email protected]
Solubility of deposited airborne heavy metals
Atmospheric Research 89 (2008) 396–404
Contents lists available at ScienceDirect
Atmospheric Research j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / a t m o s
Solubility of deposited airborne heavy metals Sibel C. Cizmecioglu a,⁎, Aysen Muezzinoglu b a b
Dokuz Eylul University, Graduate School of Natural and Applied Sciences, 35100 Bornova, Izmir, Turkey Dokuz Eylul University, Department of Environmental Engineering, Kaynaklar Campus, 35160 Buca, Izmir, Turkey
a r t i c l e
i n f o
Keywords: Heavy metals Dry deposition flux Wet deposition flux Air pollution Izmir Solubility of heavy metals
a b s t r a c t Toxic effects of heavy metals in water and soil environments are important. Quantifying the heavy metal concentrations and their solubilities in dry and wet deposition samples is part of atmospheric research. Soluble fractions of the deposited air pollutants are important in food chain mechanisms as heavy metals may cause ecotoxic impacts. In this study, the solubilities of Cr, Cd, Pb, Cu, Zn, and Ni were investigated in deposition samples for total, dissolved, and suspended fractions after collection in a surrogate, water-surface sampler in Izmir, Turkey, during October 2003 to June 2004. To find overall solubility of each metal in dry and wet deposition samples, concentrations in soluble and suspended phases of aqueous solutions were analyzed separately. Ratios between total and dissolved forms and the metals in the same forms were analyzed and evaluated statistically. It was found that the deposited metal fluxes were significantly correlated in wet deposition with the highest correlation between Cd and Pb in the soluble and total forms. Comparatively smaller correlations were found between these metal fluxes in dry deposition samples. Results of this study showed the importance of metal pollution, especially ecotoxic properties of heavy metals in wet deposition far more than dry deposition. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Transfer of metals into water and soil from the atmosphere is a significant part of their biogeochemical cycles. Heavy metals in atmospheric precipitation may create ecotoxic effects in the receiving water and soil environments; however,heavy metals are bioavailable only in their soluble forms. Heavy metals persist for a long time, leading to an increase in the environmental concentrations of these contaminants in polluted soils and waters (Mowat, 2000). Metallic pollutants may be transported in the atmosphere over long distances with very small particles. These particles when aggregated or washed out by rain are called atmospheric deposition, respectively, dry and wet. Dry deposition of particles occurs by direct impact and gravitational settling on land or water surfaces. In wet deposition, aerosols and gases are dissolved or suspended in water droplets or ice crystals (Azimi et al., 2003). Besides the long range transport processes, significant dry and wet deposition ⁎ Corresponding author. Tel./fax: +90 232 453 0922. E-mail address: [email protected] (S.C. Cizmecioglu). 0169-8095/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.atmosres.2008.03.012
also occur locally due to different sources of varying mineralogical composition and relative contributions. In addition, the importance of leaching effects for heavy metals may vary among different sites (Voutsa and Samara, 2002). The percentage of dissolved metals in deposited matter depends largely on the metal ionic form. Hydrolysis rates for heavy metal salts in water bodies or soil waters also depend on anion and cation balance, redox potential, and pH of these waters, as well as the size distribution and chemical nature of the depositing particles (Morselli et al., 2004). The abundance of crustal material and the pH of precipitation are two major parameters that determine the solubilities of elements. Ratios of soluble metals to total amounts coming from atmospheric to environmental waters may depend on the particle size and shape, molecular form, chemical speciation, and environmental conditions. For example, smaller-sized particles with longer atmospheric residence times provide greater opportunities for contribution from several different emission sources and may enhance the atmospheric reactions that change metal speciation (Voutsa and Samara, 2002). However, Heal et al. (2005) have reported
S.C. Cizmecioglu, A. Muezzinoglu / Atmospheric Research 89 (2008) 396–404
that a greater proportion of water-soluble metal is indicative of anthropogenic rather than crustal sources. Water pH is another important factor determining the ratio of dissolved and suspended heavy metal fractions. It is therefore important to know about the probability of acid rain to estimate the possibility of high-hydrolysis rates for metals. The effect of pH is the highest on moderately soluble elements with most having anthropogenic sources. Crustal-originated components of an element in the atmospheric particles are less soluble than the anthropogenic components. Because of these solubility differences, elements are found to be more soluble in urban aerosols than in rural aerosols, which are under the influence of crustal material. Since atmospheric concentrations of anthropogenic elements, such as Cr, Pb, Cd, Zn, etc., are dominated by anthropogenic components in urban areas, the difference in solubility of a given element in anthropogenic and crustal components does not play a significant role in the overall solubility
397
of that element in precipitation (Kaya and Tuncel, 1997). Several studies have reported solubilities of trace elements in rainwater by analyzing the filtrate (soluble fraction) and filter (insoluble fraction) (Kaya and Tuncel, 1997; Golomb et al., 1997). In the present study, Cr, Cd, Pb, Cu, Zn, and Ni were investigated in the deposition samples for total, dissolved, and suspended fractions after collection in a surrogate water-surface sampler (WSS) in Izmir, Turkey, during October 2003 to June 2004. To find the overall solubility of each metal in dry and wet deposition samples, concentrations in soluble and suspended phases in aqueous solutions were analyzed separately, summed up, and crosschecked with analytical results conducted in unfiltered samples. The differences between calculated and measured concentrations on the same samples were statistically insignificant. Data sets indicating the total and dissolved forms of each metal and the metals in the same form were analyzed and evaluated statistically.
Fig. 1. Dissolved and suspended fractions of heavy metals in dry deposition samples.
398
S.C. Cizmecioglu, A. Muezzinoglu / Atmospheric Research 89 (2008) 396–404
2. Materials and methods Deposition sampling was conducted on a platform 3 m above the ground at the Dokuz Eylul University Kaynaklar Campus, so that the re-entrainment from the soil surface was minimized. The site was located in a growing forest about 10 km south of the city center. A WSS was used for deposition sampling. Technical drawings and operating principles of the WSS are designated in the literature (Shahin et al., 1999; Odabasi et al., 1999; Tasdemir, 1997; Yi et al., 1997). Twelve dry deposition samples and 13 rainwater samples were collected daily between October 2003 and June 2004. Dry deposition samples were collected when there was no rain. Sampling time was 24 h for dry deposition samples. Duration of rainwater sampling varied with the period of rain. At the end of the sampling period, all of the water in the WSS system was transferred to a clean plastic bottle with a plastic stopper and carried to the laboratory where the volume was measured. After thorough shaking, a 250-mL aliquot was taken from each sample and immediately filtered through a 0.45-μm Sartorius membrane filter. A 100-mL portion of the filtrate was acidified to pH 2 by nitric acid (Merck, 65%) and refrigerated in tightly covered plastic bottles until the heavy metals were analyzed for dissolved components. In addition, filters were digested and analyzed for heavy metals to find suspended and undissolved fractions of metals. Another 100mL aliquot of the fresh unfiltered sample was acidified with nitric acid to pH 2 and preserved for total heavy metal analyses. Filtrates were digested with HNO3 for analyses of suspended fraction. To note effects of pH on solubility of the heavy metals in rainwater and receiving water of the dry deposition sampler, pH values were measured and recorded. Heavy metal concentrations were measured using a PerkinElmer Model 700 atomic absorption spectrophotometer (AAS) equipped with a graphite furnace. For Zn analysis, an AAS with flame technique was used in the same instrument to avoid erratic results in the graphite furnace due to high concentrations in the samples. Background correction was applied using a deuterium lamp with the two-line method. A quality assurance test also was applied during the analysis. In order to verify AAS readings, calibration curves were tested by using two different methods, and efficiencies were calculated as deviations from the known concentrations. At the end of each analysis run, 3 mixed test solutions containing known amounts of the stock solution and added by acidified distilled water, similarly prepared as in the acidification step of the samples to pH 2, were analyzed. Results were corrected for the metal concentrations in added acid water mixture blanks. In this test, the heavy metal concentration was chosen, so that the test solution at the end was near the third calibration point. For the second test, solutions were prepared by transferring known volumes of the heavy metal stock solution into 500-mL volumetric flasks and adding portions of randomly selected samples. This procedure was repeated using three different samples chosen at random. The difference between the measured heavy metal quantity in the sample and the results from analyzing the prepared solutions should correspond to the known heavy metal quantity from the stock solution. Any deviation from the expected mass gave information about the reliability of test results. Measurement
results in both analytical performance tests gave average agreements of 93–97% for Cr, 100% for Cd, 100% for Pb, 93– 101.3% for Cu, 92–96% for Zn, and 97–114% for Ni. The STATLETS™ statistical program was used to correlate analyses of data obtained in the present study. 3. Results 3.1. Dissolved and suspended fractions in dry and wet deposition samples Dissolved and suspended fractions of dry and wet deposition samples are shown in Figs. 1 and 2, respectively. The ratio of insoluble to soluble fractions was quite high in the first two rain samples for all heavy metals except Cd. It must be noted that the first two samples were taken during rain events following elongated dry periods. The soluble average percentages of the Cr, Cd, Pb, Cu, Zn, and Ni in dry deposition samples obtained in this study have been compared with literature data in Table 1. The ratios of soluble fractions of heavy metals in the dry deposition were about 44–100% of total metals in deposition except for one or two samples having somewhat less soluble Cr and Cd forms in deposition. This result is in contradiction with the results of Morselli et al. (2003), who reports lower soluble fractions in dry deposition than in wet deposition at Bologna, Italy. By contrast, Voutsa and Samara (2002) have reported higher soluble fractions for Cd, Pb, Cu, Zn, and Ni in the Greater Thessaloniki area of Greece. Significantly higher labile fractions of these metals were found suggesting the dominance of readily available metal forms in atmospheric particles. The average percentages of soluble Cr, Cd, Pb, Cu, Zn, and Ni in rainwater samples obtained in this study and in literature data are given in Table 2. The ratios of soluble fractions of heavy metals in rainwater were within 37–100% of the total metals in deposition except for one or two samples having somewhat less soluble Cr and Cd forms in deposition. Another exception is Ni, which has a larger range of soluble fractions in wet deposition than in dry deposition. Kaya and Tuncel (1997) reported lower solubilities for heavy metals in Ankara rainwater as compared to our results except for Cd. However, the results reported by Morselli et al. (2003) were similar to our solubility values in wet deposition samples. The solubility of Cu in rainwater samples that were sampled at Büyükçekmece Lake, Istanbul, was found to be 8.4% by Başak and Alagha (2004). Solubilities for heavy metals in deposition samples were appreciably higher at Izmir than at other places in the world. This difference may originate from regional climatic characteristics, proximity of pollution sources, differences in sampling methods and analytical procedures, amounts of precipitation, and period between precipitation events. Composition of metals emitted into the atmosphere in the form of eolian dust or aerosols is due mainly to anthropogenic activities. Eolian dust or aerosols are removed by dry or wet deposition and cause damage to surface water and organisms. Two main sources of eolian dust strongly affect the composition of atmospheric aerosols and precipitation in the Mediterranean area. One source is eolian dust transported from North Africa, and the other source is the pollution aerosol transported from Europe (Al-Khashman, 2005).
S.C. Cizmecioglu, A. Muezzinoglu / Atmospheric Research 89 (2008) 396–404
399
Fig. 2. Dissolved and suspended fractions of heavy metals in wet deposition samples.
Some heavy metals may originate essentially from the scavenging of aerosol particles by precipitation between the clouds and the ground. Medium range transport may have an
Table 1 Soluble fractions of the dry deposited heavy metals Soluble fraction in dry deposition, % Cr
Cd
Pb
Cu
Zn
Ni
Heal et al. (2005)
–
50.0
≤10
–
33.0
35.0– 45.0 –
50.0
Morselli et al. (2003) Voutsa and Samara (2002) This study
35.0– 45.0 –
25.0
–
–
80.0
66.0
90.0
93.0
82.0
90.4 ± 12.8
75.4 ± 21.6
91.8 ± 10.3
87.5 ± 12.0
89.7 ± 11.5
98.8 ± 4.0
important impact on this scavenging process (Deboudt et al., 2004). Particulate scavenging is the predominant mode by which trace metals are incorporated into rainwater. It must be stressed, however, that only the soluble fraction of heavy metals in rainwater reach the collector (Hu and Balasubramanian, 2003). It is also believed that elements from anthropogenic sources exist mainly as water-soluble forms (Voutsa and Samara, 2002). 3.2. Correlations between metal concentrations in dry and wet deposition samples Different correlation coefficients have been reported in the literature in relation to several factors. A statistically significant relationship between two heavy metals indicates that the source and characteristics for both heavy metals may be the same, although the metal concentrations themselves
400
S.C. Cizmecioglu, A. Muezzinoglu / Atmospheric Research 89 (2008) 396–404
Table 2 Soluble fractions of the wet deposited heavy metals Soluble fraction in wet deposition, %
Özsoy and Örnektekin (2005) Morselli et al. (2003) Kaya and Tuncel (1997) This study
Cr
Cd
Pb
Cu
Zn
Ni
20.1
52.3
46.1
43.4
77.7
43.9
88.0
68.0
61.0
–
74.0
–
35.0
88.0
40.0
49.0
43.0
72.0
80.1 ± 16.2
87.2 ± 15.5
92.6 ± 15.0
92.7 ± 11.3
88.5 ± 16.2
95.2 ± 17.4
might be showing considerable variability. Any variability may be due to changing meteorological conditions affecting the transport time of pollutants, chemical speciation, and strength of emissions sources contributing pollution to the air. Multiple regression analysis was applied to determine the relation among dissolved and total concentrations of dry deposition samples. Significant relationships were found between dissolved concentrations of Pb and Ni and total concentrations of Pb and Ni at the 95% confidence level, respectively (Fig. 3). Wu et al. (2006) reported a correlation coefficient of 0.51 between Cr and Cu in total suspended particulates, a correlation coefficient of 0.67 between Pb and Cu for coarse (PM2.5–10) particulates, and a correlation coefficient of 0.66 between Zn and Cu for fine (PM2.5) particulates. Tasdemir et al. (2006) have reported that correlation coefficients were 0.677, 0.673, 0.711, and 0.797 between Ni–
Fig. 3. Correlations for dry deposition samples: (a) dissolved concentrations and (b) total concentrations.
Fig. 4. Correlations between dissolved concentrations of metals in wet deposition samples.
Cu, Ni–Zn, Ni–Cd, and Cd–Pb, respectively. Halstead et al. (2000) found correlations between Cd–Pb, Cd–Zn, and Pb– Zn as 0.790, 0.804, and 0.602, respectively. Correlation calculations show a significant correlation between Cd–Cu and Zn–Ni concentrations, in the dissolved form, in rainwater at the 95% confidence level (Fig. 4). Fig. 5 shows statistically significant relationships between Pb–Cu, Pb–Zn, Pb–Ni, and Zn–Ni total rainwater concentrations. Significant relationships between these heavy metals indicate that they show similar behavior under certain physical conditions such as wind speed, air temperature, rainfall amount, etc. (Tanner and Wong, 2000). Elements that are emitted from different sources may show correlations due to similar physical properties of the particles that carry them (Kaya and Tuncel, 1997). Statistical relationships were investigated between meteorological parameters and metal concentrations in rainwater samples. There were statistically significant relationships among the total rainwater concentrations of Pb, Zn, and Ni with air temperature as 0.707, 0.863, and 0.834 (p b 0.01), respectively. Statistically significant relationships were found among the total rainwater concentrations of Pb, Zn, and Ni and wind speeds as 0.778, 0.811, and 0.620 (p b 0.01), respectively. In addition, the dissolved and total heavy metal concentrations in wet deposition samples were found uncorrelated with the volume of rain on sampling days. This outcome was in contrast to the results of Hou et al. (2005) who reported that metal concentrations in rainwater negatively correlated with the volume of rainwater, indicating that washout was the main mechanism whereby metals were incorporated into rainwater. Statistically significant relationships were not found between
S.C. Cizmecioglu, A. Muezzinoglu / Atmospheric Research 89 (2008) 396–404
401
Fig. 5. Correlations between total concentrations of metals in wet deposition samples.
rainy period on the sampling day with dissolved and total heavy metal concentrations in rainwater samples. 3.3. Correlations between fluxes of heavy metals in dry and wet deposition samples Generally, the correlation coefficients among the metal fluxes were much better in wet deposition than in dry deposition. This observation parallels the results of Maneux et al. (1999) who reported that the dissolved fraction of heavy metals in rain was greater than the fraction of suspended particulate. Significant relationships were not found between heavy metal fluxes in dry deposition samples when total and soluble fractions were separately tested; highest correlations were found between the Pb and Ni soluble flux and total flux fractions, respectively (Fig. 6). This indicates that these metals had a common source, possibly of traffic origin at the study site. There was no significant difference between the deposition fluxes of metal among seasons in the present study. Likewise, no seasonal trends in the total particle mass of heavy metals were reported by Sakata and Marumoto (2004) and Sweet et al. (1998) at their respective study sites. Azimi et al. (2005) reported that these elements were mostly (N95%) of non-crustal origin. Bozlaker (2002) gives the composition of Cd, Zn, Pb, and Ni in dry deposition indicating their dominant anthropogenic origin exhibited by their high crustal enrichment factor (EFcrust) values (N10) at the same site as the present study. Particles originating from resuspension of dust, which is an important source for many trace metals (Sweet et al., 1998), have an important influence on atmospheric deposition (Azimi et al., 2005). The wet deposited fluxes of different metals are significantly interrelated except Zn (Tables 3 and 4). In soluble form, the most
Fig. 6. Correlations for dry deposition samples: (a) dissolved fluxes and (b) total fluxes.
402
S.C. Cizmecioglu, A. Muezzinoglu / Atmospheric Research 89 (2008) 396–404
Table 3 Correlation coefficients of soluble heavy metal fluxes in wet deposition samples Cr Cr Cd Pb Cu Zn Ni
2
r p r2 p r2 p r2 p r2 p r2 p
1.00 0.00 0.318 <0.05 0.060 N 0.05 0.105 N 0.05 0.026 N 0.05 0.256 N 0.05
Cd
1.00 0.00 0.575 <0.01 0.745 <0.01 0.010 N0.05 0.461 <0.05
Pb
Cu
Zn
Ni
1.00 0.00 0.436 < 0.05 0.066 N 0.05 0.551 < 0.01
Total
1.00 0.00 0.078 N 0.05 0.192 N 0.05
1.00 0.00 0.083 N0.05
1.00 0.00
Table 4 Correlation coefficients of total heavy metal fluxes in wet deposition samples
Cd Pb Cu Zn Ni
Cr
Cd
1.00 0.00 0.246 N 0.05 0.023 N 0.05 0.064 N 0.05 0.007 N 0.05 0.420 < 0.05
1.00 0.00 0.491 < 0.01 0.518 < 0.01 0.000 N 0.05 0.466 < 0.01
Pb
1.00 0.00 0.424 <0.05 0.288 N0.05 0.513 <0.01
Cu
Metal
r2 p r2 p
Cr
Cd
Pb
Cu
Zn
Ni
0.549 b0.01 0.588 b0.01
0.335 b0.05 0.344 b0.05
0.146 N0.05 0.174 N0.05
0.591 b0.01 0.468 b0.01
0.243 N0.05 0.117 N0.05
0.286 N0.05 0.181 N0.05
Figures in italics and bold indicated statistically significant values at the 95% confidence level.
important relationship was found between Cd and Cu (r2 = 0.745; p b 0.01). In total form, however, a significant relation was found between Cd and Pb (r2 = 0.518; p b 0.01). There were statistically significant relationships between wind speed values and dissolved and total fluxes of Zn (r2 = 0.779; p b 0.01 and r2 = 0.876; p b 0.01, respectively). The statistical relationships between air temperatures on sampling days and dissolved and total fluxes of Cr have been found to be r2 = 0.379 (p b 0.05) and r2 = 0.358 (p b 0.05), respectively. There was not a statistically significant relationship between dissolved and total fluxes of heavy metals in wet deposition samples and relative humidity values at the 95% confidence level. In this study, statistical relationships between daily rain volumes and daily soluble metal fluxes have been found to be r2 = 0.446 (p b 0.05) for Cr; r2 = 0.477 (p b 0.01) for Cd; r2 = 0.476 (p b 0.01) for Pb, and r2 = 0.875 (p b 0.01) for Ni. Daily rain volumes and daily total fluxes of Cr (r2 = 0.504; p b 0.01), Cd (r2 = 0.490; p b 0.01), Pb (r2 = 0.306; p b 0.05), and Ni (r2 = 0.745; p b 0.01) were correlated, too. These are in contradiction with Takeda et al. (2000) who reported that Cr, Zn, Cu, and Ni fluxes do not correlate with the amount of precipitation except for Cr and Zn. However these results are in parallel with the Pb results determined by Kim (1998). Sakata et al. (2006) noted that the annual wet deposition fluxes of As, Hg, Cr, Cd, Pb, Mn, V, Sb, and Se were correlated with the annual precipitation amount.
r2 p r2 p r2 p r2 p r2 p r2 p
Sample fraction Soluble
Figures in italics and bold indicated statistically significant values at the 95% confidence level.
Cr
Table 5 Correlations between soluble and total concentration and flux values in dry deposition samples
Zn
Ni
Sweet et al. (1998) stated that wet deposition of metals was more closely related to the amount of precipitation than the dependence of the concentration of metals in precipitation to rain volume. The results of Luo (2001) showed that all wet deposition fluxes of soluble chemical species and elements in insoluble materials have strong correlations with rain intensity. Kaya and Tuncel (1997) reported that the low, wet deposition fluxes of measured parameters found in Ankara, Turkey, were due to low precipitation amounts in the sampling period. Tanner and Wong (2000) indicated several occasions where low rainfall volume, small rainfall intensity, and low wind speed produce high concentrations of trace metals in bulk deposition in Hong Kong with pH values down to 3.6. Sakata et al. (2006) noted that the annual wet deposition fluxes of As, Hg, Cr, Cd, Pb, Mn, V, Sb, and Se correlated with the annual precipitation amount. In Izmir, the relationships between the daily rain period and daily dissolved and total fluxes of heavy metals have not been found. Maneux et al. (1999) reported that although elements display generally parallel variations with time, they sometimes follow independent behaviors (e.g., Pb and Cd), suggesting that they may derive from different geographical and/or pollution sources. 3.4. Correlations between concentrations and fluxes of dry and wet deposition samples Table 5 gives the correlation coefficients of dry deposition fluxes with the concentrations for soluble and total parts of metals in the samples. Significant relationships have been observed between soluble and total concentrations and fluxes of Cr, Cd, and Cu. Wu et al. (2006) have reported the relationships between fluxes and total suspended, coarse and fine particulate concentrations of metallic elements and found significant relations for Cr, Pb, Cu and Zn. Their results indicated that the flux was highly correlated with coarse particle concentrations for most of the elements because particles larger Table 6 Correlations between soluble and total concentration and flux values in wet deposition samples
1.00 0.00 0.076 N 0.05 0.203 N 0.05
Sample fraction 1.00 0.00 0.066 N 0.05
Soluble 1.00 0.00
Figures in italics and bold indicated statistically significant values at the 95% confidence level.
Total
Metal
r2 p r2 p
Cr
Cd
Pb
Cu
Zn
Ni
0.123 N0.05 0.084 N0.05
0.502 < 0.01 0.471 < 0.01
0.431 <0.05 0.436 <0.05
0.792 <0.01 0.729 <0.01
0.663 < 0.01 0.713 < 0.01
0.004 N0.05 0.000 N0.05
Figures in italics and bold indicated statistically significant values at the 95% confidence level.
S.C. Cizmecioglu, A. Muezzinoglu / Atmospheric Research 89 (2008) 396–404
than 2.5 µm had high deposition velocities. Maneux et al. (1999) have reported that the dissolved fraction of heavy metals in rain was greater than the suspended fraction. Table 6 shows correlations between soluble and total concentration and flux values in wet deposition samples. Statistically significant correlations were found between soluble and total concentrations and fluxes of Cd and Cu. Luo (2001) has reported the strong correlation coefficient of wet deposition fluxes with the concentrations for Cr. This observation indicated that the mechanism for wet deposition was controlled by factors such as rain volume and duration along with concentration. 3.5. Impact of pH on the solubility of heavy metals in the deposition samples The pH of dry deposition samples obtained from the WSS ranged between 5.6 and 7.8 with an average of 6.6. The only notable correlations were found between pH values for dissolved and total dry deposition concentrations of Pb and Ni, respectively, at the 95% confidence level (Fig. 7). Correlation coefficient between total Pb concentrations and pH values was equal to 0.552 (p b 0.01). This low dependence may have been caused by the relatively narrow range of pH values in the rain at Izmir. The pH in collected rainwater samples ranged between 5.1 and 7.7 with an average of 6.5. Dissolved and total metal concentrations and pH values of rainwater samples at the 95% confidence level did not correlate well in the present study. Acid rain (i.e., rain with pH b 5.0) did not occur during the sampling period. Yet, no significantly low pH was notable in the samples.
403
This indicates that the strongly acidic rain phenomena did not exist in Izmir during the study period. This is in contrast to the high quantity of SO2 emissions in the area (Dincer et al., 2003) but in parallel with the findings of Al-Momani et al. (1995) who have concluded that the neutralizing capacity of aerosols of crustal origin in precipitation are highly due to the excessive limestone coverage in the Izmir area. Al-Momani et al. (1998) also have indicated that the excessive alkaline material in crustal aerosols in the Eastern Mediterranean region has a strong neutralizing impact on rainwater. Similarly, Gülsoy et al. (1999) indicated the frequency of acidic pH to be about one-fifth of the total rain incidences in 1996 for Istanbul, Turkey. In addition, Akkoyunlu and Tayanc (2003) showed that although the SO2− 4 concentration in precipitation was high, so was the Ca2+ as a neutralizing factor of acidity in the rain. Tuncer et al. (2001) determined that approximately 95% of acidity in precipitation samples was neutralized, particularly in the summer season, and the neutralizing agents were primarily CaCO3 and NH3 from Central Anatolia. Kaya and Tuncel (1997) investigated the effect of pH on solubilities of elements in rainwater samples. They found that solubilities of all elements were higher in samples with pH values lower than 5.0, although pH did not have the same effect on solubility for all elements. Results from the present study have shown the importance of metal pollution in rainwater especially concerning the ecotoxic properties of heavy metals rather than dry deposition, which occurred over the long-term. Results also showed that pH was a relatively unimportant property in creating ecotoxic effects due to enhanced solubility.
Fig. 7. Relation between concentrations and pH values of dry deposition samples: (a) soluble concentrations and (b) total concentrations.
404
S.C. Cizmecioglu, A. Muezzinoglu / Atmospheric Research 89 (2008) 396–404
4. Conclusion In the present study, the soluble fractions of metals in aqueous environmental samples due to wet and dry deposition from the air have been evaluated. Statistical correlations between concentrations in water and in the deposition fluxes of metals have been studied. These correlations are discussed in view of the environmental conditions such as meteorology and source effects. This discussion in relation to the soluble fraction ratios for deposition in Izmir indicated that the studied metals were ready to impose ecotoxic impacts in the water–soil environments and on biota. The impact was enhanced by highly soluble fractions in the dry deposition, as well. Sudden increases in heavy metal deposition in the dissolved form can be an important input into biochemical cycles and may create impacts that are more significant. However, statistical evidence shows that the relationships between concentrations of metals in wet and dry deposition are not high. This was true also for dry deposition flux samples. However, there are significant correlations between metals in wet deposition fluxes when soluble and total fractions accompanying these fluxes are considered. This may be due to common sources for these metals available only under rainy conditions in contrast to the significance of nearby sources and soil resuspension under the dry weather conditions. Acknowledgments Dokuz Eylül University Research Funds and TÜBİTAK 100Y104 project are acknowledged for partial financial support. Dokuz Eylül University Air Pollution Laboratory staff is gratefully acknowledged for sampling and analytical help. References Akkoyunlu, B.O., Tayanç, M., 2003. Analyses of wet and bulk deposition in four different regions of Istanbul, Turkey. Atmospheric Environment 37, 3571–3579. Al-Khashman, O.A., 2005. Ionic composition of wet precipitation in the Petra Region, Jordan. Atmospheric Research 78, 1–12. Al-Momani, I.F., Ataman, O.Y., Anwari, M.A., Tuncel, S., Köse, C., Tuncel, G., 1995. Chemical composition of precipitation near an industrial area at Izmir, Turkey. Atmospheric Environment 29, 1131–1143. Al-Momani, I.F., Aygun, S., Tuncel, G., 1998. Wet deposition of major ions and trace elements in the eastern Mediterranean basin. Journal of Geophysical Research-Atmospheres 103, 8287–8300. Azimi, S., Ludwig, A., Thevenot, D.R., Colin, J.L., 2003. Trace metal determination in total atmospheric deposition in rural and urban areas. Science of the Total Environment 308, 247–256. Azimi, S., Rocher, V., Muller, M., Moilleron, R., Thevenot, D.R., 2005. Sources, distribution and variability of hydrocarbons and metals in atmospheric deposition in an urban area (Paris, France). Science of the Total Environment 337, 223–239. Başak, B., Alagha, O., 2004. The chemical composition of rainwater over Büyükçekmece Lake, Istanbul. Atmospheric Research 71, 275–288. Bozlaker, A., 2002. Trace metals in airborne particles and their dry deposition in Izmir. MS Thesis, Graduate School of Natural and Applied Sciences, Dokuz Eylül University, Izmir, Turkey. Deboudt, K., Flament, P., Bertho, M.L., 2004. Cd, Cu, Pb and Zn concentrations in atmospheric wet deposition at a coastal station in Western Europe. Water, Air and Soil Pollution 151, 335–359. Dincer, F., Muezzinoglu, A., Elbir, T., 2003. SO2 levels at forested mountains around Izmir, Turkey and their possible sources. Water, Air and Soil Pollution 147, 331–341. Golomb, D., Ryan, D., Eby, N., et al., 1997. Atmospheric deposition of toxics onto Massachusetts Bay-I. Metals. Atmospheric Environment 31, 1349–1359. Gülsoy, G., Tayanç, M., Ertürk, F., 1999. Chemical analyses of the major ions in the precipitation of Istanbul, Turkey. Environmental Pollution 105, 273–280.
Halstead, M.J.R., Cunninghame, R.G., Hunter, K.A., 2000. Wet deposition of trace metals to a remote site in Fiordland, New Zealand. Atmospheric Environment 34, 665–676. Heal, M.R., Hibbs, L.R., Agius, R.M., Beverland, I.J., 2005. Total and watersoluble trace metal content of urban background PM10, PM2.5 and black smoke in Edinburgh, UK. Atmospheric Environment 39, 1417–1430. Hou, H., Takamatsu, T., Koshikawa, M.K., Hosomi, M., 2005. Trace metals in bulk precipitation and throughfall in a suburban area of Japan. Atmospheric Environment 39, 3583–3595. Hu, G.P., Balasubramanian, R., 2003. Wet deposition of trace metals in Singapore. Water, Air, and Soil Pollution 144, 285–300. Kaya, G., Tuncel, G., 1997. Trace element and major ion composition of wet and dry deposition in Ankara, Turkey. Atmospheric Environment 31, 3985–3998. Kim, G., 1998. Atmospheric inputs and upper ocean biogeochemistry of trace elements and radionuclide tracers in the Atlantic. Ph.D. Thesis, University of Delaware. Luo, W., 2001. Wet-deposition fluxes of soluble chemical species and the elements in insoluble materials. Atmospheric Environment 35, 2963–2967. Maneux, E., Grousset, F.E., Buat-Ménard, P., Lavaux, G., Rimmelin, P., Lapaquellerie, Y., 1999. Temporal patterns of the wet deposition of Zn, Cu, Ni, Cd and Pb: the Arcachon Lagoon (France). Water, Air, and Soil Pollution 114, 95–120. Morselli, L., Olivieri, P., Brusori, B., Passarini, F., 2003. Soluble and insoluble fractions of heavy metals in wet and dry atmospheric depositions in Bologna, Italy. Environmental Pollution 124, 457–469. Morselli, L., Brusori, B., Passarini, F., et al., 2004. Heavy metals monitoring at a Mediterranean natural ecosystem of Central Italy. Trends in different environmental matrixes. Environmental International 30, 173–181. Mowat, F.S., 2000. Toxicity Assessment of Complex Contaminant Mixtures in Louisiana Sediments. Ph.D. dissertation. Tulane University, New Orleans, LA, USA. Odabasi, M., Sofuoglu, A., Vardar, N., Tasdemir, Y., Holsen, T.M., 1999. Measurement of dry deposition and air–water exchange of polycyclic hydrocarbons with the water surface sampler. Environmental Sciences 33, 426–434. Özsoy, T., Örnektekin, S., 2005. Elemental composition of rainwater in Mersin, an urban site in the North Eastern Mediterranean. Proceedings of the Third International Symposium on Air Quality Management at Urban, Regional and Global Scales & 14th IUAPPA Regional Conference 18–27. 26–30 September 2005. Istanbul, Turkey. Sakata, M., Marumoto, K., 2004. Dry deposition fluxes and deposition velocities of trace metals in the Tokyo Metropolitan Area measured with a water surface sampler. Environmental Science and Technology 38, 2190–2197. Sakata, M., Marumoto, K., Narukawa, M., Asakura, K., 2006. Regional variations in wet and dry deposition fluxes of trace elements in Japan. Atmospheric Environment 40, 521–531. Shahin, U.M., Zhu, X., Holsen, T.M., 1999. Dry deposition of reduced and reactive nitrogen: a surrogate surface approach. Environmental Science and Technology 33, 2113–2117. Sweet, C.W., Weiss, A., Vermette, S.J., 1998. Atmospheric deposition of trace metals at three sites near the Great Lakes. Water, Air, and Soil Pollution 103, 423–439. Takeda, K., Marumoto, K., Minamikawa, T., Sakugawa, H., Fujiwara, K., 2000. Three-year determination of trace metals and the lead isotope ratio in rain and snow depositions collected in Higashi-Hiroshima, Japan. Atmospheric Environment 34, 4525–4535. Tanner, P.A., Wong, A.Y.S., 2000. Soluble trace metals and major ionic species in the bulk deposition and atmosphere of Hong Kong. Water, Air and Soil Pollution 122, 261–279. Tasdemir, Y., 1997. Modification and evaluation of a water surface sampler to investigate the dry deposition and air–water exchange of polychlorinated biphenyls (PCBs). Ph.D. Thesis, Illinois Institute of Technology, Chicago, IL. Tasdemir, Y., Kural, C., Cindoruk, S.S., Vardar, N., 2006. Assessment of trace element concentrations and their estimated dry deposition fluxes in an urban atmosphere. Atmospheric Research 81 (1), 17–35. Tuncer, B., Bayar, B., Yesilyurt, C., Tuncel, G., 2001. Ionic composition of precipitation at the Central Anatolia (Turkey). Atmospheric Environment 35, 5958–6002. Voutsa, D., Samara, C., 2002. Labile and bioaccessible fractions of heavy metals in the airborne particulate matter from urban and industrial areas. Atmospheric Environment 36, 3583–3590. Wu, Y.S., Fang, G.C., Chen, J.C., Lin, C.P., Huang, S.H., Rau, J.Y., Lin, J.G., 2006. Ambient air particulate dry deposition, concentrations and metallic elements at Taichung Harbor near Taiwan Strait. Atmospheric Research 79, 52–66. Yi, S.M., Holsen, T.M., Noll, K.E., 1997. Comparison of dry deposition predicted from models and measured with a water surface sampler. Environmental Sciences and Technology 31, 272–278.
Contents lists available at ScienceDirect
Atmospheric Research j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / a t m o s
Solubility of deposited airborne heavy metals Sibel C. Cizmecioglu a,⁎, Aysen Muezzinoglu b a b
Dokuz Eylul University, Graduate School of Natural and Applied Sciences, 35100 Bornova, Izmir, Turkey Dokuz Eylul University, Department of Environmental Engineering, Kaynaklar Campus, 35160 Buca, Izmir, Turkey
a r t i c l e
i n f o
Keywords: Heavy metals Dry deposition flux Wet deposition flux Air pollution Izmir Solubility of heavy metals
a b s t r a c t Toxic effects of heavy metals in water and soil environments are important. Quantifying the heavy metal concentrations and their solubilities in dry and wet deposition samples is part of atmospheric research. Soluble fractions of the deposited air pollutants are important in food chain mechanisms as heavy metals may cause ecotoxic impacts. In this study, the solubilities of Cr, Cd, Pb, Cu, Zn, and Ni were investigated in deposition samples for total, dissolved, and suspended fractions after collection in a surrogate, water-surface sampler in Izmir, Turkey, during October 2003 to June 2004. To find overall solubility of each metal in dry and wet deposition samples, concentrations in soluble and suspended phases of aqueous solutions were analyzed separately. Ratios between total and dissolved forms and the metals in the same forms were analyzed and evaluated statistically. It was found that the deposited metal fluxes were significantly correlated in wet deposition with the highest correlation between Cd and Pb in the soluble and total forms. Comparatively smaller correlations were found between these metal fluxes in dry deposition samples. Results of this study showed the importance of metal pollution, especially ecotoxic properties of heavy metals in wet deposition far more than dry deposition. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Transfer of metals into water and soil from the atmosphere is a significant part of their biogeochemical cycles. Heavy metals in atmospheric precipitation may create ecotoxic effects in the receiving water and soil environments; however,heavy metals are bioavailable only in their soluble forms. Heavy metals persist for a long time, leading to an increase in the environmental concentrations of these contaminants in polluted soils and waters (Mowat, 2000). Metallic pollutants may be transported in the atmosphere over long distances with very small particles. These particles when aggregated or washed out by rain are called atmospheric deposition, respectively, dry and wet. Dry deposition of particles occurs by direct impact and gravitational settling on land or water surfaces. In wet deposition, aerosols and gases are dissolved or suspended in water droplets or ice crystals (Azimi et al., 2003). Besides the long range transport processes, significant dry and wet deposition ⁎ Corresponding author. Tel./fax: +90 232 453 0922. E-mail address: [email protected] (S.C. Cizmecioglu). 0169-8095/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.atmosres.2008.03.012
also occur locally due to different sources of varying mineralogical composition and relative contributions. In addition, the importance of leaching effects for heavy metals may vary among different sites (Voutsa and Samara, 2002). The percentage of dissolved metals in deposited matter depends largely on the metal ionic form. Hydrolysis rates for heavy metal salts in water bodies or soil waters also depend on anion and cation balance, redox potential, and pH of these waters, as well as the size distribution and chemical nature of the depositing particles (Morselli et al., 2004). The abundance of crustal material and the pH of precipitation are two major parameters that determine the solubilities of elements. Ratios of soluble metals to total amounts coming from atmospheric to environmental waters may depend on the particle size and shape, molecular form, chemical speciation, and environmental conditions. For example, smaller-sized particles with longer atmospheric residence times provide greater opportunities for contribution from several different emission sources and may enhance the atmospheric reactions that change metal speciation (Voutsa and Samara, 2002). However, Heal et al. (2005) have reported
S.C. Cizmecioglu, A. Muezzinoglu / Atmospheric Research 89 (2008) 396–404
that a greater proportion of water-soluble metal is indicative of anthropogenic rather than crustal sources. Water pH is another important factor determining the ratio of dissolved and suspended heavy metal fractions. It is therefore important to know about the probability of acid rain to estimate the possibility of high-hydrolysis rates for metals. The effect of pH is the highest on moderately soluble elements with most having anthropogenic sources. Crustal-originated components of an element in the atmospheric particles are less soluble than the anthropogenic components. Because of these solubility differences, elements are found to be more soluble in urban aerosols than in rural aerosols, which are under the influence of crustal material. Since atmospheric concentrations of anthropogenic elements, such as Cr, Pb, Cd, Zn, etc., are dominated by anthropogenic components in urban areas, the difference in solubility of a given element in anthropogenic and crustal components does not play a significant role in the overall solubility
397
of that element in precipitation (Kaya and Tuncel, 1997). Several studies have reported solubilities of trace elements in rainwater by analyzing the filtrate (soluble fraction) and filter (insoluble fraction) (Kaya and Tuncel, 1997; Golomb et al., 1997). In the present study, Cr, Cd, Pb, Cu, Zn, and Ni were investigated in the deposition samples for total, dissolved, and suspended fractions after collection in a surrogate water-surface sampler (WSS) in Izmir, Turkey, during October 2003 to June 2004. To find the overall solubility of each metal in dry and wet deposition samples, concentrations in soluble and suspended phases in aqueous solutions were analyzed separately, summed up, and crosschecked with analytical results conducted in unfiltered samples. The differences between calculated and measured concentrations on the same samples were statistically insignificant. Data sets indicating the total and dissolved forms of each metal and the metals in the same form were analyzed and evaluated statistically.
Fig. 1. Dissolved and suspended fractions of heavy metals in dry deposition samples.
398
S.C. Cizmecioglu, A. Muezzinoglu / Atmospheric Research 89 (2008) 396–404
2. Materials and methods Deposition sampling was conducted on a platform 3 m above the ground at the Dokuz Eylul University Kaynaklar Campus, so that the re-entrainment from the soil surface was minimized. The site was located in a growing forest about 10 km south of the city center. A WSS was used for deposition sampling. Technical drawings and operating principles of the WSS are designated in the literature (Shahin et al., 1999; Odabasi et al., 1999; Tasdemir, 1997; Yi et al., 1997). Twelve dry deposition samples and 13 rainwater samples were collected daily between October 2003 and June 2004. Dry deposition samples were collected when there was no rain. Sampling time was 24 h for dry deposition samples. Duration of rainwater sampling varied with the period of rain. At the end of the sampling period, all of the water in the WSS system was transferred to a clean plastic bottle with a plastic stopper and carried to the laboratory where the volume was measured. After thorough shaking, a 250-mL aliquot was taken from each sample and immediately filtered through a 0.45-μm Sartorius membrane filter. A 100-mL portion of the filtrate was acidified to pH 2 by nitric acid (Merck, 65%) and refrigerated in tightly covered plastic bottles until the heavy metals were analyzed for dissolved components. In addition, filters were digested and analyzed for heavy metals to find suspended and undissolved fractions of metals. Another 100mL aliquot of the fresh unfiltered sample was acidified with nitric acid to pH 2 and preserved for total heavy metal analyses. Filtrates were digested with HNO3 for analyses of suspended fraction. To note effects of pH on solubility of the heavy metals in rainwater and receiving water of the dry deposition sampler, pH values were measured and recorded. Heavy metal concentrations were measured using a PerkinElmer Model 700 atomic absorption spectrophotometer (AAS) equipped with a graphite furnace. For Zn analysis, an AAS with flame technique was used in the same instrument to avoid erratic results in the graphite furnace due to high concentrations in the samples. Background correction was applied using a deuterium lamp with the two-line method. A quality assurance test also was applied during the analysis. In order to verify AAS readings, calibration curves were tested by using two different methods, and efficiencies were calculated as deviations from the known concentrations. At the end of each analysis run, 3 mixed test solutions containing known amounts of the stock solution and added by acidified distilled water, similarly prepared as in the acidification step of the samples to pH 2, were analyzed. Results were corrected for the metal concentrations in added acid water mixture blanks. In this test, the heavy metal concentration was chosen, so that the test solution at the end was near the third calibration point. For the second test, solutions were prepared by transferring known volumes of the heavy metal stock solution into 500-mL volumetric flasks and adding portions of randomly selected samples. This procedure was repeated using three different samples chosen at random. The difference between the measured heavy metal quantity in the sample and the results from analyzing the prepared solutions should correspond to the known heavy metal quantity from the stock solution. Any deviation from the expected mass gave information about the reliability of test results. Measurement
results in both analytical performance tests gave average agreements of 93–97% for Cr, 100% for Cd, 100% for Pb, 93– 101.3% for Cu, 92–96% for Zn, and 97–114% for Ni. The STATLETS™ statistical program was used to correlate analyses of data obtained in the present study. 3. Results 3.1. Dissolved and suspended fractions in dry and wet deposition samples Dissolved and suspended fractions of dry and wet deposition samples are shown in Figs. 1 and 2, respectively. The ratio of insoluble to soluble fractions was quite high in the first two rain samples for all heavy metals except Cd. It must be noted that the first two samples were taken during rain events following elongated dry periods. The soluble average percentages of the Cr, Cd, Pb, Cu, Zn, and Ni in dry deposition samples obtained in this study have been compared with literature data in Table 1. The ratios of soluble fractions of heavy metals in the dry deposition were about 44–100% of total metals in deposition except for one or two samples having somewhat less soluble Cr and Cd forms in deposition. This result is in contradiction with the results of Morselli et al. (2003), who reports lower soluble fractions in dry deposition than in wet deposition at Bologna, Italy. By contrast, Voutsa and Samara (2002) have reported higher soluble fractions for Cd, Pb, Cu, Zn, and Ni in the Greater Thessaloniki area of Greece. Significantly higher labile fractions of these metals were found suggesting the dominance of readily available metal forms in atmospheric particles. The average percentages of soluble Cr, Cd, Pb, Cu, Zn, and Ni in rainwater samples obtained in this study and in literature data are given in Table 2. The ratios of soluble fractions of heavy metals in rainwater were within 37–100% of the total metals in deposition except for one or two samples having somewhat less soluble Cr and Cd forms in deposition. Another exception is Ni, which has a larger range of soluble fractions in wet deposition than in dry deposition. Kaya and Tuncel (1997) reported lower solubilities for heavy metals in Ankara rainwater as compared to our results except for Cd. However, the results reported by Morselli et al. (2003) were similar to our solubility values in wet deposition samples. The solubility of Cu in rainwater samples that were sampled at Büyükçekmece Lake, Istanbul, was found to be 8.4% by Başak and Alagha (2004). Solubilities for heavy metals in deposition samples were appreciably higher at Izmir than at other places in the world. This difference may originate from regional climatic characteristics, proximity of pollution sources, differences in sampling methods and analytical procedures, amounts of precipitation, and period between precipitation events. Composition of metals emitted into the atmosphere in the form of eolian dust or aerosols is due mainly to anthropogenic activities. Eolian dust or aerosols are removed by dry or wet deposition and cause damage to surface water and organisms. Two main sources of eolian dust strongly affect the composition of atmospheric aerosols and precipitation in the Mediterranean area. One source is eolian dust transported from North Africa, and the other source is the pollution aerosol transported from Europe (Al-Khashman, 2005).
S.C. Cizmecioglu, A. Muezzinoglu / Atmospheric Research 89 (2008) 396–404
399
Fig. 2. Dissolved and suspended fractions of heavy metals in wet deposition samples.
Some heavy metals may originate essentially from the scavenging of aerosol particles by precipitation between the clouds and the ground. Medium range transport may have an
Table 1 Soluble fractions of the dry deposited heavy metals Soluble fraction in dry deposition, % Cr
Cd
Pb
Cu
Zn
Ni
Heal et al. (2005)
–
50.0
≤10
–
33.0
35.0– 45.0 –
50.0
Morselli et al. (2003) Voutsa and Samara (2002) This study
35.0– 45.0 –
25.0
–
–
80.0
66.0
90.0
93.0
82.0
90.4 ± 12.8
75.4 ± 21.6
91.8 ± 10.3
87.5 ± 12.0
89.7 ± 11.5
98.8 ± 4.0
important impact on this scavenging process (Deboudt et al., 2004). Particulate scavenging is the predominant mode by which trace metals are incorporated into rainwater. It must be stressed, however, that only the soluble fraction of heavy metals in rainwater reach the collector (Hu and Balasubramanian, 2003). It is also believed that elements from anthropogenic sources exist mainly as water-soluble forms (Voutsa and Samara, 2002). 3.2. Correlations between metal concentrations in dry and wet deposition samples Different correlation coefficients have been reported in the literature in relation to several factors. A statistically significant relationship between two heavy metals indicates that the source and characteristics for both heavy metals may be the same, although the metal concentrations themselves
400
S.C. Cizmecioglu, A. Muezzinoglu / Atmospheric Research 89 (2008) 396–404
Table 2 Soluble fractions of the wet deposited heavy metals Soluble fraction in wet deposition, %
Özsoy and Örnektekin (2005) Morselli et al. (2003) Kaya and Tuncel (1997) This study
Cr
Cd
Pb
Cu
Zn
Ni
20.1
52.3
46.1
43.4
77.7
43.9
88.0
68.0
61.0
–
74.0
–
35.0
88.0
40.0
49.0
43.0
72.0
80.1 ± 16.2
87.2 ± 15.5
92.6 ± 15.0
92.7 ± 11.3
88.5 ± 16.2
95.2 ± 17.4
might be showing considerable variability. Any variability may be due to changing meteorological conditions affecting the transport time of pollutants, chemical speciation, and strength of emissions sources contributing pollution to the air. Multiple regression analysis was applied to determine the relation among dissolved and total concentrations of dry deposition samples. Significant relationships were found between dissolved concentrations of Pb and Ni and total concentrations of Pb and Ni at the 95% confidence level, respectively (Fig. 3). Wu et al. (2006) reported a correlation coefficient of 0.51 between Cr and Cu in total suspended particulates, a correlation coefficient of 0.67 between Pb and Cu for coarse (PM2.5–10) particulates, and a correlation coefficient of 0.66 between Zn and Cu for fine (PM2.5) particulates. Tasdemir et al. (2006) have reported that correlation coefficients were 0.677, 0.673, 0.711, and 0.797 between Ni–
Fig. 3. Correlations for dry deposition samples: (a) dissolved concentrations and (b) total concentrations.
Fig. 4. Correlations between dissolved concentrations of metals in wet deposition samples.
Cu, Ni–Zn, Ni–Cd, and Cd–Pb, respectively. Halstead et al. (2000) found correlations between Cd–Pb, Cd–Zn, and Pb– Zn as 0.790, 0.804, and 0.602, respectively. Correlation calculations show a significant correlation between Cd–Cu and Zn–Ni concentrations, in the dissolved form, in rainwater at the 95% confidence level (Fig. 4). Fig. 5 shows statistically significant relationships between Pb–Cu, Pb–Zn, Pb–Ni, and Zn–Ni total rainwater concentrations. Significant relationships between these heavy metals indicate that they show similar behavior under certain physical conditions such as wind speed, air temperature, rainfall amount, etc. (Tanner and Wong, 2000). Elements that are emitted from different sources may show correlations due to similar physical properties of the particles that carry them (Kaya and Tuncel, 1997). Statistical relationships were investigated between meteorological parameters and metal concentrations in rainwater samples. There were statistically significant relationships among the total rainwater concentrations of Pb, Zn, and Ni with air temperature as 0.707, 0.863, and 0.834 (p b 0.01), respectively. Statistically significant relationships were found among the total rainwater concentrations of Pb, Zn, and Ni and wind speeds as 0.778, 0.811, and 0.620 (p b 0.01), respectively. In addition, the dissolved and total heavy metal concentrations in wet deposition samples were found uncorrelated with the volume of rain on sampling days. This outcome was in contrast to the results of Hou et al. (2005) who reported that metal concentrations in rainwater negatively correlated with the volume of rainwater, indicating that washout was the main mechanism whereby metals were incorporated into rainwater. Statistically significant relationships were not found between
S.C. Cizmecioglu, A. Muezzinoglu / Atmospheric Research 89 (2008) 396–404
401
Fig. 5. Correlations between total concentrations of metals in wet deposition samples.
rainy period on the sampling day with dissolved and total heavy metal concentrations in rainwater samples. 3.3. Correlations between fluxes of heavy metals in dry and wet deposition samples Generally, the correlation coefficients among the metal fluxes were much better in wet deposition than in dry deposition. This observation parallels the results of Maneux et al. (1999) who reported that the dissolved fraction of heavy metals in rain was greater than the fraction of suspended particulate. Significant relationships were not found between heavy metal fluxes in dry deposition samples when total and soluble fractions were separately tested; highest correlations were found between the Pb and Ni soluble flux and total flux fractions, respectively (Fig. 6). This indicates that these metals had a common source, possibly of traffic origin at the study site. There was no significant difference between the deposition fluxes of metal among seasons in the present study. Likewise, no seasonal trends in the total particle mass of heavy metals were reported by Sakata and Marumoto (2004) and Sweet et al. (1998) at their respective study sites. Azimi et al. (2005) reported that these elements were mostly (N95%) of non-crustal origin. Bozlaker (2002) gives the composition of Cd, Zn, Pb, and Ni in dry deposition indicating their dominant anthropogenic origin exhibited by their high crustal enrichment factor (EFcrust) values (N10) at the same site as the present study. Particles originating from resuspension of dust, which is an important source for many trace metals (Sweet et al., 1998), have an important influence on atmospheric deposition (Azimi et al., 2005). The wet deposited fluxes of different metals are significantly interrelated except Zn (Tables 3 and 4). In soluble form, the most
Fig. 6. Correlations for dry deposition samples: (a) dissolved fluxes and (b) total fluxes.
402
S.C. Cizmecioglu, A. Muezzinoglu / Atmospheric Research 89 (2008) 396–404
Table 3 Correlation coefficients of soluble heavy metal fluxes in wet deposition samples Cr Cr Cd Pb Cu Zn Ni
2
r p r2 p r2 p r2 p r2 p r2 p
1.00 0.00 0.318 <0.05 0.060 N 0.05 0.105 N 0.05 0.026 N 0.05 0.256 N 0.05
Cd
1.00 0.00 0.575 <0.01 0.745 <0.01 0.010 N0.05 0.461 <0.05
Pb
Cu
Zn
Ni
1.00 0.00 0.436 < 0.05 0.066 N 0.05 0.551 < 0.01
Total
1.00 0.00 0.078 N 0.05 0.192 N 0.05
1.00 0.00 0.083 N0.05
1.00 0.00
Table 4 Correlation coefficients of total heavy metal fluxes in wet deposition samples
Cd Pb Cu Zn Ni
Cr
Cd
1.00 0.00 0.246 N 0.05 0.023 N 0.05 0.064 N 0.05 0.007 N 0.05 0.420 < 0.05
1.00 0.00 0.491 < 0.01 0.518 < 0.01 0.000 N 0.05 0.466 < 0.01
Pb
1.00 0.00 0.424 <0.05 0.288 N0.05 0.513 <0.01
Cu
Metal
r2 p r2 p
Cr
Cd
Pb
Cu
Zn
Ni
0.549 b0.01 0.588 b0.01
0.335 b0.05 0.344 b0.05
0.146 N0.05 0.174 N0.05
0.591 b0.01 0.468 b0.01
0.243 N0.05 0.117 N0.05
0.286 N0.05 0.181 N0.05
Figures in italics and bold indicated statistically significant values at the 95% confidence level.
important relationship was found between Cd and Cu (r2 = 0.745; p b 0.01). In total form, however, a significant relation was found between Cd and Pb (r2 = 0.518; p b 0.01). There were statistically significant relationships between wind speed values and dissolved and total fluxes of Zn (r2 = 0.779; p b 0.01 and r2 = 0.876; p b 0.01, respectively). The statistical relationships between air temperatures on sampling days and dissolved and total fluxes of Cr have been found to be r2 = 0.379 (p b 0.05) and r2 = 0.358 (p b 0.05), respectively. There was not a statistically significant relationship between dissolved and total fluxes of heavy metals in wet deposition samples and relative humidity values at the 95% confidence level. In this study, statistical relationships between daily rain volumes and daily soluble metal fluxes have been found to be r2 = 0.446 (p b 0.05) for Cr; r2 = 0.477 (p b 0.01) for Cd; r2 = 0.476 (p b 0.01) for Pb, and r2 = 0.875 (p b 0.01) for Ni. Daily rain volumes and daily total fluxes of Cr (r2 = 0.504; p b 0.01), Cd (r2 = 0.490; p b 0.01), Pb (r2 = 0.306; p b 0.05), and Ni (r2 = 0.745; p b 0.01) were correlated, too. These are in contradiction with Takeda et al. (2000) who reported that Cr, Zn, Cu, and Ni fluxes do not correlate with the amount of precipitation except for Cr and Zn. However these results are in parallel with the Pb results determined by Kim (1998). Sakata et al. (2006) noted that the annual wet deposition fluxes of As, Hg, Cr, Cd, Pb, Mn, V, Sb, and Se were correlated with the annual precipitation amount.
r2 p r2 p r2 p r2 p r2 p r2 p
Sample fraction Soluble
Figures in italics and bold indicated statistically significant values at the 95% confidence level.
Cr
Table 5 Correlations between soluble and total concentration and flux values in dry deposition samples
Zn
Ni
Sweet et al. (1998) stated that wet deposition of metals was more closely related to the amount of precipitation than the dependence of the concentration of metals in precipitation to rain volume. The results of Luo (2001) showed that all wet deposition fluxes of soluble chemical species and elements in insoluble materials have strong correlations with rain intensity. Kaya and Tuncel (1997) reported that the low, wet deposition fluxes of measured parameters found in Ankara, Turkey, were due to low precipitation amounts in the sampling period. Tanner and Wong (2000) indicated several occasions where low rainfall volume, small rainfall intensity, and low wind speed produce high concentrations of trace metals in bulk deposition in Hong Kong with pH values down to 3.6. Sakata et al. (2006) noted that the annual wet deposition fluxes of As, Hg, Cr, Cd, Pb, Mn, V, Sb, and Se correlated with the annual precipitation amount. In Izmir, the relationships between the daily rain period and daily dissolved and total fluxes of heavy metals have not been found. Maneux et al. (1999) reported that although elements display generally parallel variations with time, they sometimes follow independent behaviors (e.g., Pb and Cd), suggesting that they may derive from different geographical and/or pollution sources. 3.4. Correlations between concentrations and fluxes of dry and wet deposition samples Table 5 gives the correlation coefficients of dry deposition fluxes with the concentrations for soluble and total parts of metals in the samples. Significant relationships have been observed between soluble and total concentrations and fluxes of Cr, Cd, and Cu. Wu et al. (2006) have reported the relationships between fluxes and total suspended, coarse and fine particulate concentrations of metallic elements and found significant relations for Cr, Pb, Cu and Zn. Their results indicated that the flux was highly correlated with coarse particle concentrations for most of the elements because particles larger Table 6 Correlations between soluble and total concentration and flux values in wet deposition samples
1.00 0.00 0.076 N 0.05 0.203 N 0.05
Sample fraction 1.00 0.00 0.066 N 0.05
Soluble 1.00 0.00
Figures in italics and bold indicated statistically significant values at the 95% confidence level.
Total
Metal
r2 p r2 p
Cr
Cd
Pb
Cu
Zn
Ni
0.123 N0.05 0.084 N0.05
0.502 < 0.01 0.471 < 0.01
0.431 <0.05 0.436 <0.05
0.792 <0.01 0.729 <0.01
0.663 < 0.01 0.713 < 0.01
0.004 N0.05 0.000 N0.05
Figures in italics and bold indicated statistically significant values at the 95% confidence level.
S.C. Cizmecioglu, A. Muezzinoglu / Atmospheric Research 89 (2008) 396–404
than 2.5 µm had high deposition velocities. Maneux et al. (1999) have reported that the dissolved fraction of heavy metals in rain was greater than the suspended fraction. Table 6 shows correlations between soluble and total concentration and flux values in wet deposition samples. Statistically significant correlations were found between soluble and total concentrations and fluxes of Cd and Cu. Luo (2001) has reported the strong correlation coefficient of wet deposition fluxes with the concentrations for Cr. This observation indicated that the mechanism for wet deposition was controlled by factors such as rain volume and duration along with concentration. 3.5. Impact of pH on the solubility of heavy metals in the deposition samples The pH of dry deposition samples obtained from the WSS ranged between 5.6 and 7.8 with an average of 6.6. The only notable correlations were found between pH values for dissolved and total dry deposition concentrations of Pb and Ni, respectively, at the 95% confidence level (Fig. 7). Correlation coefficient between total Pb concentrations and pH values was equal to 0.552 (p b 0.01). This low dependence may have been caused by the relatively narrow range of pH values in the rain at Izmir. The pH in collected rainwater samples ranged between 5.1 and 7.7 with an average of 6.5. Dissolved and total metal concentrations and pH values of rainwater samples at the 95% confidence level did not correlate well in the present study. Acid rain (i.e., rain with pH b 5.0) did not occur during the sampling period. Yet, no significantly low pH was notable in the samples.
403
This indicates that the strongly acidic rain phenomena did not exist in Izmir during the study period. This is in contrast to the high quantity of SO2 emissions in the area (Dincer et al., 2003) but in parallel with the findings of Al-Momani et al. (1995) who have concluded that the neutralizing capacity of aerosols of crustal origin in precipitation are highly due to the excessive limestone coverage in the Izmir area. Al-Momani et al. (1998) also have indicated that the excessive alkaline material in crustal aerosols in the Eastern Mediterranean region has a strong neutralizing impact on rainwater. Similarly, Gülsoy et al. (1999) indicated the frequency of acidic pH to be about one-fifth of the total rain incidences in 1996 for Istanbul, Turkey. In addition, Akkoyunlu and Tayanc (2003) showed that although the SO2− 4 concentration in precipitation was high, so was the Ca2+ as a neutralizing factor of acidity in the rain. Tuncer et al. (2001) determined that approximately 95% of acidity in precipitation samples was neutralized, particularly in the summer season, and the neutralizing agents were primarily CaCO3 and NH3 from Central Anatolia. Kaya and Tuncel (1997) investigated the effect of pH on solubilities of elements in rainwater samples. They found that solubilities of all elements were higher in samples with pH values lower than 5.0, although pH did not have the same effect on solubility for all elements. Results from the present study have shown the importance of metal pollution in rainwater especially concerning the ecotoxic properties of heavy metals rather than dry deposition, which occurred over the long-term. Results also showed that pH was a relatively unimportant property in creating ecotoxic effects due to enhanced solubility.
Fig. 7. Relation between concentrations and pH values of dry deposition samples: (a) soluble concentrations and (b) total concentrations.
404
S.C. Cizmecioglu, A. Muezzinoglu / Atmospheric Research 89 (2008) 396–404
4. Conclusion In the present study, the soluble fractions of metals in aqueous environmental samples due to wet and dry deposition from the air have been evaluated. Statistical correlations between concentrations in water and in the deposition fluxes of metals have been studied. These correlations are discussed in view of the environmental conditions such as meteorology and source effects. This discussion in relation to the soluble fraction ratios for deposition in Izmir indicated that the studied metals were ready to impose ecotoxic impacts in the water–soil environments and on biota. The impact was enhanced by highly soluble fractions in the dry deposition, as well. Sudden increases in heavy metal deposition in the dissolved form can be an important input into biochemical cycles and may create impacts that are more significant. However, statistical evidence shows that the relationships between concentrations of metals in wet and dry deposition are not high. This was true also for dry deposition flux samples. However, there are significant correlations between metals in wet deposition fluxes when soluble and total fractions accompanying these fluxes are considered. This may be due to common sources for these metals available only under rainy conditions in contrast to the significance of nearby sources and soil resuspension under the dry weather conditions. Acknowledgments Dokuz Eylül University Research Funds and TÜBİTAK 100Y104 project are acknowledged for partial financial support. Dokuz Eylül University Air Pollution Laboratory staff is gratefully acknowledged for sampling and analytical help. References Akkoyunlu, B.O., Tayanç, M., 2003. Analyses of wet and bulk deposition in four different regions of Istanbul, Turkey. Atmospheric Environment 37, 3571–3579. Al-Khashman, O.A., 2005. Ionic composition of wet precipitation in the Petra Region, Jordan. Atmospheric Research 78, 1–12. Al-Momani, I.F., Ataman, O.Y., Anwari, M.A., Tuncel, S., Köse, C., Tuncel, G., 1995. Chemical composition of precipitation near an industrial area at Izmir, Turkey. Atmospheric Environment 29, 1131–1143. Al-Momani, I.F., Aygun, S., Tuncel, G., 1998. Wet deposition of major ions and trace elements in the eastern Mediterranean basin. Journal of Geophysical Research-Atmospheres 103, 8287–8300. Azimi, S., Ludwig, A., Thevenot, D.R., Colin, J.L., 2003. Trace metal determination in total atmospheric deposition in rural and urban areas. Science of the Total Environment 308, 247–256. Azimi, S., Rocher, V., Muller, M., Moilleron, R., Thevenot, D.R., 2005. Sources, distribution and variability of hydrocarbons and metals in atmospheric deposition in an urban area (Paris, France). Science of the Total Environment 337, 223–239. Başak, B., Alagha, O., 2004. The chemical composition of rainwater over Büyükçekmece Lake, Istanbul. Atmospheric Research 71, 275–288. Bozlaker, A., 2002. Trace metals in airborne particles and their dry deposition in Izmir. MS Thesis, Graduate School of Natural and Applied Sciences, Dokuz Eylül University, Izmir, Turkey. Deboudt, K., Flament, P., Bertho, M.L., 2004. Cd, Cu, Pb and Zn concentrations in atmospheric wet deposition at a coastal station in Western Europe. Water, Air and Soil Pollution 151, 335–359. Dincer, F., Muezzinoglu, A., Elbir, T., 2003. SO2 levels at forested mountains around Izmir, Turkey and their possible sources. Water, Air and Soil Pollution 147, 331–341. Golomb, D., Ryan, D., Eby, N., et al., 1997. Atmospheric deposition of toxics onto Massachusetts Bay-I. Metals. Atmospheric Environment 31, 1349–1359. Gülsoy, G., Tayanç, M., Ertürk, F., 1999. Chemical analyses of the major ions in the precipitation of Istanbul, Turkey. Environmental Pollution 105, 273–280.
Halstead, M.J.R., Cunninghame, R.G., Hunter, K.A., 2000. Wet deposition of trace metals to a remote site in Fiordland, New Zealand. Atmospheric Environment 34, 665–676. Heal, M.R., Hibbs, L.R., Agius, R.M., Beverland, I.J., 2005. Total and watersoluble trace metal content of urban background PM10, PM2.5 and black smoke in Edinburgh, UK. Atmospheric Environment 39, 1417–1430. Hou, H., Takamatsu, T., Koshikawa, M.K., Hosomi, M., 2005. Trace metals in bulk precipitation and throughfall in a suburban area of Japan. Atmospheric Environment 39, 3583–3595. Hu, G.P., Balasubramanian, R., 2003. Wet deposition of trace metals in Singapore. Water, Air, and Soil Pollution 144, 285–300. Kaya, G., Tuncel, G., 1997. Trace element and major ion composition of wet and dry deposition in Ankara, Turkey. Atmospheric Environment 31, 3985–3998. Kim, G., 1998. Atmospheric inputs and upper ocean biogeochemistry of trace elements and radionuclide tracers in the Atlantic. Ph.D. Thesis, University of Delaware. Luo, W., 2001. Wet-deposition fluxes of soluble chemical species and the elements in insoluble materials. Atmospheric Environment 35, 2963–2967. Maneux, E., Grousset, F.E., Buat-Ménard, P., Lavaux, G., Rimmelin, P., Lapaquellerie, Y., 1999. Temporal patterns of the wet deposition of Zn, Cu, Ni, Cd and Pb: the Arcachon Lagoon (France). Water, Air, and Soil Pollution 114, 95–120. Morselli, L., Olivieri, P., Brusori, B., Passarini, F., 2003. Soluble and insoluble fractions of heavy metals in wet and dry atmospheric depositions in Bologna, Italy. Environmental Pollution 124, 457–469. Morselli, L., Brusori, B., Passarini, F., et al., 2004. Heavy metals monitoring at a Mediterranean natural ecosystem of Central Italy. Trends in different environmental matrixes. Environmental International 30, 173–181. Mowat, F.S., 2000. Toxicity Assessment of Complex Contaminant Mixtures in Louisiana Sediments. Ph.D. dissertation. Tulane University, New Orleans, LA, USA. Odabasi, M., Sofuoglu, A., Vardar, N., Tasdemir, Y., Holsen, T.M., 1999. Measurement of dry deposition and air–water exchange of polycyclic hydrocarbons with the water surface sampler. Environmental Sciences 33, 426–434. Özsoy, T., Örnektekin, S., 2005. Elemental composition of rainwater in Mersin, an urban site in the North Eastern Mediterranean. Proceedings of the Third International Symposium on Air Quality Management at Urban, Regional and Global Scales & 14th IUAPPA Regional Conference 18–27. 26–30 September 2005. Istanbul, Turkey. Sakata, M., Marumoto, K., 2004. Dry deposition fluxes and deposition velocities of trace metals in the Tokyo Metropolitan Area measured with a water surface sampler. Environmental Science and Technology 38, 2190–2197. Sakata, M., Marumoto, K., Narukawa, M., Asakura, K., 2006. Regional variations in wet and dry deposition fluxes of trace elements in Japan. Atmospheric Environment 40, 521–531. Shahin, U.M., Zhu, X., Holsen, T.M., 1999. Dry deposition of reduced and reactive nitrogen: a surrogate surface approach. Environmental Science and Technology 33, 2113–2117. Sweet, C.W., Weiss, A., Vermette, S.J., 1998. Atmospheric deposition of trace metals at three sites near the Great Lakes. Water, Air, and Soil Pollution 103, 423–439. Takeda, K., Marumoto, K., Minamikawa, T., Sakugawa, H., Fujiwara, K., 2000. Three-year determination of trace metals and the lead isotope ratio in rain and snow depositions collected in Higashi-Hiroshima, Japan. Atmospheric Environment 34, 4525–4535. Tanner, P.A., Wong, A.Y.S., 2000. Soluble trace metals and major ionic species in the bulk deposition and atmosphere of Hong Kong. Water, Air and Soil Pollution 122, 261–279. Tasdemir, Y., 1997. Modification and evaluation of a water surface sampler to investigate the dry deposition and air–water exchange of polychlorinated biphenyls (PCBs). Ph.D. Thesis, Illinois Institute of Technology, Chicago, IL. Tasdemir, Y., Kural, C., Cindoruk, S.S., Vardar, N., 2006. Assessment of trace element concentrations and their estimated dry deposition fluxes in an urban atmosphere. Atmospheric Research 81 (1), 17–35. Tuncer, B., Bayar, B., Yesilyurt, C., Tuncel, G., 2001. Ionic composition of precipitation at the Central Anatolia (Turkey). Atmospheric Environment 35, 5958–6002. Voutsa, D., Samara, C., 2002. Labile and bioaccessible fractions of heavy metals in the airborne particulate matter from urban and industrial areas. Atmospheric Environment 36, 3583–3590. Wu, Y.S., Fang, G.C., Chen, J.C., Lin, C.P., Huang, S.H., Rau, J.Y., Lin, J.G., 2006. Ambient air particulate dry deposition, concentrations and metallic elements at Taichung Harbor near Taiwan Strait. Atmospheric Research 79, 52–66. Yi, S.M., Holsen, T.M., Noll, K.E., 1997. Comparison of dry deposition predicted from models and measured with a water surface sampler. Environmental Sciences and Technology 31, 272–278.