Methylmercury determination in marine sediment and organisms by Direct Mercury Analyser

Analytica Chimica Acta 641 (2009) 32–36

Contents lists available at ScienceDirect

Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca

Methylmercury determination in marine sediment and organisms by Direct Mercury Analyser Chiara Maggi a , Maria Teresa Berducci a , Jessica Bianchi a,∗ , Michele Giani b , Luigi Campanella c a b c

Advanced Institute for Environmental Protection and Research – ISPRA, Via di Casalotti 300, Rome, Italy National Institute of Oceanography and Experimental Geophysics, Biological Oceanography Department, Via Piccard 54, 34100 Trieste, Italy Chemical Department University “La Sapienza”, P.le Aldo Moro 5, 00185 Rome, Italy

a r t i c l e

i n f o

Article history: Received 10 November 2008 Received in revised form 6 February 2009 Accepted 16 March 2009 Available online 25 March 2009 Keywords: Methylmercury Direct Mercury Analyser Marine sediments Biological tissues Antarctica

a b s t r a c t An analytical method for simple and rapid determination of methylmercury in sediment and organism samples is described. The proposed method employs the oxygen combustion-gold amalgamation using Direct Mercury Analyser (DMA-80) after complete removal of MeHg by organic extraction and back extraction to an aqueous medium. DMA-80 instrument is equally suitable for the analysis of solid and liquid materials and has a good detection limit. The analytical performance of this method was evaluated by analysis of certified reference materials (CRM-580, IAEA-405, DORM-2, DOLT-3, SRM-2976 and SRM2977) assessing its quality in terms of accuracy, repeatability and quantification limit. Furthermore total mercury and methylmercury have been analysed in sediment and organism samples collected during the XXI Italian Antarctic Expedition in Terra Nova Bay (Ross Sea, Northern Victoria Land). The results obtained show the validity of the proposed method as ready-to-use analytical method to analyse real samples. © 2009 Elsevier B.V. All rights reserved.

1. Introduction The biogeochemistry of mercury (Hg) has received, in the last decades, considerable attention because of the extreme toxicity of methylmercury (MeHg) and its ability to bioaccumulate in biota and to biomagnificate in aquatic food web. In fact, Hg is the only metal which bioaccumulates through all levels of the aquatic food chain; accordingly, Hg contamination in fish has been a widespread health concern. [1,2]. Also MeHg becomes biomagnified in the food chain through passage from bacteria, plankton, macroinvertebrates, herbivorous fish, piscivorous fish and, finally to humans. Each step results in an increased concentration of MeHg, which may end being many times higher in the animal than the initial concentration in the water [3]. MeHg blocks the binding sites of enzymes, and interferes with the protein synthesis as well as thymidine incorporation into DNA. Fetus and neonates are known to be high-risk groups for MeHg: it can be transferred to the fetus through the placenta and to offspring through breast milk and can be caused severe effects on neurological development [4–6]. One of the most catastrophic example in modern history is mercury pollution in Minamata, Japan, where waste water containing inorganic mercury was released into the Shiranui Sea between 1932

∗ Corresponding author. Tel.: +39 06 61570504; fax: +39 06 61561906. E-mail address: [email protected] (J. Bianchi). 0003-2670/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.aca.2009.03.033

and 1966 by the Chisso Factory. This inorganic mercury was transformed into MeHg through the aquatic ecosystem and ingested by a large number of people because of their traditional seafood-based diet. [3,7]. In several countries MeHg is frequently determinated in fish samples to control the contamination level prior market sales. Recently, monitoring of MeHg content in sediment has been started in research laboratories for the purpose of pollution monitoring and geochemical studies [8–11] A large assortment of microorganisms is capable of converting inorganic Hg2+ into CH3 Hg+ . Rates of MeHg formation depend on the amount of Hg that is available for methylation reactions, rather than on T-Hg concentrations, and on various physical, chemical and biological factors [12,13]: the microbial activity, composition of sediment and the organic content [14], the sulphidic character of sediment [15], the oxygen concentration, pH and the presence of inorganic and organic complexing agents [16]. In general the methylation rates are favoured in anaerobic sediment. Methylation rates are usually highest in the upper 2 cm of sediments where the microbiological activity is highest [17]. The determination of MeHg in sediments is not easy: the very low concentrations, and the presence of interfering substances (such as sulphides, humic acids, aminoacids and protein able to strongly bind methylmercury) are the main analytical complications [18]. In fact the most critical compartments for speciation are still linked to solid phase. Extraction is a very subtle step because the whole species content may not be liberated, and artefacts can occur

C. Maggi et al. / Analytica Chimica Acta 641 (2009) 32–36

so that some organomercury species can be destroyed or formed (interspecies exchange) [19]. It is important to note that because of low MeHg concentrations in marine matrices the measurement protocol must be carefully designed; besides analytical procedures for each of mercurial species are quite different. Numerous analytical methods based on the home-made hyphenations between an appropriate separation technique and a sensitive and selective detector (gas or liquid chromatography and mass spectrometry, fluorescence spectrometry, electron capture) have proposed to measure MeHg in sediment and marine organisms. Beside separation and detection, these methods involve several crucial points, as extraction, derivatization, preconcentration and clean-up, that are potential source of errors and may lead to low precision and lack of repeatability [20–23]. Since Governmental Bodies and Environmental Protection Agencies starts to be sensitive towards Hg pollution problem, it is necessary to have ready-to-use analytical methods which are simple, low cost and fast sample treatment. In this paper the analysis of MeHg in marine sediment and biota samples by separation of the MeHg from the sample matrix in order to avoid interference during the extraction reaction, then MeHg extraction with organic solvent has been carried out; after complete removal of MeHg by organic extraction, it has been transferred to an aqueous medium, than its analysis has been possible by AAS. The T-Hg concentration in all environmental samples has been also determined. The singularity of proposed method is the possibility to estimate MeHg content in solid matrices (sediment and biota) using the same instrument for T-Hg determination [24]. This method allows to perform routine analysis of methylmercury in sediment and marine organisms by Direct Mercury Analyser (DMA-80) [25]. DMA-80 instrument does not require any pre-treatment of the samples, is equally suitable for the analysis of solid and liquid materials and has a good detection limit (LOD); moreover it offers the possibility of changing operational conditions as a function of the materials and quantity of samples [26,27]. The quality of the proposed method has been checked by analysis of certified reference materials (CRM) and its applicability to real samples has been proved by processing several organisms and sediment samples collected from the coastal sites during the austral summer from the XXI Italian Antarctic Expedition at Terra Nova Bay (Ross Sea, Northern Victoria Land).

33

Solutions employed for hydrolysis and back extraction were HCl (J.T. Baker), toluene (J.T. Baker) and l-cysteine solution (1%, v/w) that was prepared dissolving 1% l-cysteinium chloride in 12.5% anhydrous sodium sulfate and 0.775% sodium acetate. 2.3. Instrumentation The analyses were carried out with a Direct Mercury Analyser (DMA-80, Milestone srl, Italy). The sample (liquid material for MeHg analysis and solid material for T-Hg determination) is dried and then thermally decomposed by controlled heating. Decomposition products are carried to a catalyst by an oxygen flow, then sample oxidation is completed and halogens and nitrogen/sulphur oxides are trapped. The final decomposition products pass through a mercury amalgamator which collects Hg0 . The Hg amalgamator is heated to 700 ◦ C and the Hg0 is released and quantified. 2.4. Sample preparation for MeHg determination 2.4.1. Sediment sample pre-treatment (i) Approximately 1.0–2.0 g of a dry sediment (oven dried at 35 ◦ C for 48 h) was weighed out in triplicate; each sample was placed in 100 mL screw-capped polypropylene tube and hydrolyzed with 6 M HCl (10 mL). The sample was shaken for 5 min using a mechanical vertical shaker, centrifuged at 2400 rpm for 10 min and the liquid phase was discarded (no significant MeHg concentration,
2. Experimental 2.1. Materials and decontamination All material (vessels, centrifuge tubes, etc.) used was decontaminated with the following procedure: washing with a common detergent rinsing with Milli Q quality water (three times) and soaking into a clean diluted HNO3 20% (v/v) bath for 24 h at 25 ◦ C. Each soaking was followed by an intensive rising with ultra-pure water (Milli-Q). Finally, all material was dried in clean environment. 2.2. Reagents and standards All reagents used were of analytical-reagent grade unless otherwise stated. The solutions were prepared using ultra-pure water Milli-Q. The calibration standard was prepared by making two appropriate dilutions in stock water solution (1000 mg L−1 ) of CH3 Hg+ in 2% HNO3 and of Hg+ in 2% HNO3 . A blank calibration solution was also used for a zero calibration.

Fig. 1. Scheme of the proposed method for MeHg analysis in sediment and biological tissues.

34

C. Maggi et al. / Analytica Chimica Acta 641 (2009) 32–36

2.5. Methylmercury extraction 20 mL toluene was added to the organisms and sediment residue. All samples were vigorously mixed for 20 min. After centrifugation (2400 rpm for 20 min) the supernatant, containing organomercury species, was collected in falcon tubes. The whole procedure was repeated: toluene (15 mL) was added again to tube containing (i) sediment and (ii) organism tissue. The combined organic extracts were subjected twice to back extraction with 6 mL of 1% (v/w) l-cysteine aqueous solution to strip methylmercury from toluene. Then, an aliquot of l-cysteine extract was immediately analysed with mercury analyser (DMA-80). The whole procedure was optimized using different incubation times both for hydrolysis (2, 5, 10, 15 min) and for extraction (10, 15, 20, 25, 30 min). 2.6. Total mercury determination For T-Hg determination the Direct Mercury Analyser, DMA-80 (FKV) was used. The instrument calibration and analytical procedure were conducted according to EPA 7473 [27]. Measurements were carried out on ca. 0.1–0.2 g of dried (at 35 ◦ C for 48 h) solid material, without any pre-treatment. Quantification limit (LOQ) was 0.5 ng g−1 . 3. Results and discussion 3.1. Method validation

Fig. 2. Sites of sediment sampling in Terra Nova Bay.

The analytical performance of the measurements for T-Hg and MeHg was checked using overall six certified reference materials; two estuarine sediment, BCR-580 [28] and IAEA-405 [29] respectively from Community Bureau of Reference (Brussels, Belgium) and International Atomic Energy Agency (Austria); four biological tissues, two mussel tissues (marine bivalve mollusc tissue): SRM-2976 [30] and SRM-2977 [31], both from National Institute of Standard and Technology (NIST, Canada), DOLT-3 (dogfish liver) [32] and DORM-2 (dogfish muscle) [33] both from the National Research Council (NRC Canada).

All materials showed recovery greater than 80%, except DOLT-3 that showed a lower recovery (74%). This behaviour is probably due to the relatively high lipid content of this material which may affect the separation of the different phases during the organomercury species extraction by forming an emulsion [35]. The detection limit (LOD) was 0.4 ng MeHg+ which corresponded to 1.5 ng g−1 when 1.5 g of dry sample was analysed; it was calculated as the blank plus three times the standard deviation (S.D.). Quantification limit (LOQ) was 0.6 ng MeHg+ which corresponded to 2.5 ng g−1 if 1.5 g of dry sample are processed, and it was calculated as the blank plus 10 times the S.D.

3.1.1. Figures of merit The certified values of CRM, the mean values and standard deviation are presented in Table 1. The concentrations of MeHg were not statistically different from their certified ones; the z-score (standard deviation calculated with Horwitz equation) parameter ranged between −0.71 for SRM-2976 and −1.70 for DOLT-3. The precision of the whole procedure (toluene-extraction/lcysteine extraction/detection steps) was evaluated by repeating three extractions per day in three different consecutive days. Analysis of variance (ANOVA) of the results indicates the withinday and the between day precision are statistically comparable (Fcalculated < Fcritical ) for all certified reference materials (Table 2) [11,34].

3.2. Application of the methodology MeHg and T-Hg were determinated in marine sediments and organisms collected from coastal sites during the austral summer from the XXI Italian Antarctic Expedition at Terra Nova Bay (Fig. 2) in January 2006. The sediment samples were taken in eight sites near the Italian station base. The organism samples, mussels (Adamussium colbecki) and various species of Antarctic fish (Cygnodraco mawsoni, Trematomus pennelli, Trematomus bernacchii, Chionodraco hamatus, Gymnodraco acuticeps), were also collected in the same places. Organisms have been sectioned and muscle tissues anal-

Table 1 T-Hg (n1 = replicates number) and MeHg (n2 = replicates number) in certified reference materials (CRM): certified values and results of analytical determination (ng g−1 , dry weight) by DMA-80. CRM

DORM-2 DOLT-3 SRM-2976 SRM-2977 BCR-580 IAEA-405

Type of material

Dogfish muscle Dogfish liver Mussel tissue Mussel tissue Estuarine sediment Estuarine sediment

Certified value (mean ± S.D. ng g−1 )

Determined value (mean ± S.D. ng g−1 )

T-Hg

n1

T-Hg

3 3 3 3 3 3

4471 3360 60.4 92.4 116600 845.92

4640 3370 61.0 101.0 132000 810

MeHg ± ± ± ± ± ±

260 140 3.6 4.0 3000 40

4470 1590 27.8 36.2 75 5.49

± ± ± ± ± ±

320 120 1.1 1.7 4 0.53

± ± ± ± ± ±

143 57 2.9 1.4 1632 15.95

n2

MeHg

9 9 9 9 9 9

3766.8 1174.3 25.8 29.4 83.3 6.03

± ± ± ± ± ±

272.7 108.2 2.3 1.2 5.5 0.2

C. Maggi et al. / Analytica Chimica Acta 641 (2009) 32–36

35

Table 2 Analysis of variance (ANOVA): between and within-day precision. CRM

BCR-580

IAEA-405

Day 1 (ng g−1 )

Day 2 (ng g−1 )

77.46 77.10 76.61

Day 3 (ng g−1 )

78.73 79.50 79.75

6.370 6.281 6.202

78.12 84.24 83.17

5.507 5.968 5.028

6.250 6.231 6.590

Variance between days

Variance within a day

Fcalculated

Fcritical

Fcalculated

Fcritical

1.11

6.94

4.80

6.94

0.313

6.94

5.81

6.94

SRM-2976

23.63 25.56 24.83

25.07 25.55 25.23

29.08 27.32 27.08

2.05

6.94

0.82

6.94

SRM-2977

30.53 31.00 29.93

29.01 32.05 31.97

28.46 28.68 27.90

1.02

6.94

5.25

6.94

DOLT-3

1284.67 1208.28 1185.68

1253.82 1233.73 1255.97

1293.12 1266.07 1288.81

1.79

6.94

3.02

6.94

DORM-2

3946.48 3916.75 4118.93

3866.18 3867.37 3879.71

3944.56 3571.65 3751.14

1.20

6.94

2.92

6.94

ysed; all samples, organism tissues and sediment, were stored and kept at −20 ◦ C until analysis. Antarctic marine ecosystem has unique characteristics resulting from its distance from continents with high population. Anthropogenic contamination is negligible because there is no human impact due to any significant human work activities [36]. So we have found T-Hg concentrations in sediment were very low, ranging from
Table 3 T-Hg and MeHg concentration in Antarctic sediments (ng g−1 dry weight). Location

T-Hg (ng g−1 )

MeHg+ (ng g−1 )

1: P.Ta Stocchino 2: Thetys Bay 1 3: Thetys Bay 2 4: Road Bay 1 6: Road Bay 2 7: Road Bay 3 5: Molo BTN 8: Caletta

22.5 ± 11.7 ± 10.6 ± 13.0 ± 3.8 ± 3.2 ± 4.0 ±

1.2 0.1 0.4 1.2 0.1 0.2 0.2

Table 4 Hg and MeHg+ in the soft tissue of the Antarctic bivalve A. colbecki and in the muscles Antarctic fishes (C. mawsoni, T. pennelli, G. acuticeps, C. hamatus, T. bernacchii). Concentrations are expressed on the dry weight basis. T-Hg (ng g−1 ) A. colbecki A. colbecki A. colbecki A. colbecki T. bernacchii T. pennelli G. acuticeps G. acuticeps C. hamatus C. mawsoni

58.4 75.1 74.5 40.7 92.6 299.2 379.9 395.9 378.6 907.3

± ± ± ± ± ± ± ± ± ±

2.9 4.2 2.7 1.1 0.8 3.6 2.3 4.5 6.2 4.7

MeHg (ng g−1 ) 19.6 33.0 35.8 29.6 62.3 135.6 230.1 278.5 307.0 670.2

± ± ± ± ± ± ± ± ± ±

0.2 1.3 6.1 2.5 4.1 5.3 1.8 2.8 2.1 0.9

MeHg/T-Hg (%) 33.6 43.9 48.0 72.7 67.3 45.3 60.6 70.3 81.9 73.9

different feeding behaviour. In fact animal size is important in determining the rate of physiological processes influencing the uptake distribution and elimination of pollutants [42,43]. This observation is particularly true for Hg because the levels of this element in fish increase with body size: older fish generally having higher concentrations than younger fishes. The results obtained showed in addition a progressive increase of Hg concentration in organisms at different level of marine food web: T. bernacchii < T. pennelli < G. acuticeps < C. hamatus < C. mawsoni supporting natural biomagnification processes. C. mawsoni and C. hamatus are two carnivorous predator fishes also G. acuticeps is a carnivorous predator but it was smaller than C. mawsoni and C. hamatus. Instead T. pennelli and T. bernachii are a benthic fish omnivorous predators [44,45]. The proportion of T-Hg present in the methylated form in marine organisms varies widely among species and with trophic positions. The data on muscle tissue for fishes and A. colbecki in our paper fit this general picture well. 4. Conclusion Numerous procedures proposed for the analysis of MeHg in sediments and biological samples involve several steps (extraction, derivatization, preconcentration, separation and detection) and they are quite long and sources of random errors. The methodology reported in this study is simple, rapid, repeatable and accurate (good results have been obtained after analysis of certified

36

C. Maggi et al. / Analytica Chimica Acta 641 (2009) 32–36

reference materials). In addition this method shows a low detection limit and uses the same instrument for T-Hg and MeHg analysis. The data from Antarctica samples on both Hg and MeHg concentrations reported in this study confirm the principal aim of this work, that is the validation of the proposed method as suitable tool for routine analysis of organic mercury content (in addition of total content) in sediment and biota samples. References [1] N.M. Lawson, R.P. Mason, Biogeochemistry 40 (1998) 235. [2] M. Horvat, S. Covelli, J. Faganeli, M. Logar, V. Mandic, R. Rajar, A. Sirca, D. Zagar, Sci. Total Environ. 237 (1999) 43. [3] T. Dovydaitis, J. Midwifery Women’s Health 53 (2008) 325. [4] G.J. Myers, P.W. Davidson, C. Cox, C. Shamlaye, E. Cernichiari, T.W. Clarkson, Environ. Res. Sec. A 83 (2000) 275. [5] M. Sakamoto, A. Kakita, K. Wakabayashi, H. Takahashi, A. Nakano, H. Akagi, Brain Res. 949 (2002) 51. [6] G.J. Myers, P.W. Davidson, C. Cox, C. Shamlaye, D. Palumbo, E. Cernichiari, J.S. Reeves, G.E. Wilding, J. Kost, Li-Shan Huang, T.W. Clarkson, Lancet 361 (2003) 1686. [7] A. Kudo, Y. Fujikawa, S. Miyahara, J. Zheng, H. Takigami, M. Sugahara, T. Muramatsu, Wat. Sci. Technol. 8 (1998) 187. [8] C. Trombini, D. Fabbri, M. Lombardo, I. Bassura, E. Zavoli, M. Horvat, Contin. Shelf Res. 23 (2003) 1821. [9] H. Hintelmann, R.D. Wilken, Sci. Tot. Environ. 166 (1995) 1. [10] T. Tomiyasu, A. Nagano, N. Yonehara, H. Sakamoto, Rifardi, K. Oki, H. Akagi, Sci. Tot. Environ. 257 (2000) 121. [11] J.S. Landaluze, A. De diego, J.C. Raposo, J.M. Madariaga, Anal. Chim. Acta 508 (2004) 107. [12] S.M. Ullrich, T.W. Stanton, S.A. Abdrashitova, Crit. Rev. Environ. Sci. Technol. 31 (2001) 241. [13] C. Gagnon, E. Peelletier, A. Mucci, Mar. Chem. 59 (1997) 159. [14] C.R. Hammerschimidt, W.F. Fitzgerald, Environ. Sci. Technol. 38 (2004) 1487. [15] J.M. Benoit, C.C. Gilmour, R.P. Mason, A. Heyes, Environ. Sci. Technol. 33 (1999) 951. [16] V. Celo, D.R.S. Lean, S.L. Scott, Sci. Tot. Environ. 368 (2006) 126. [17] S.M. Ullrich, M.A. Ilyushchenko, G.A. Uskov, T.W. Tanton, Appl. Geochem. 22 (2007) 2706.

[18] A.M. Caricchia, G. Minervini, P. Soldati, S. Chiavarini, C. Ubaldi, R. Morabito, Microchem. J. 55 (1997). [19] M. Leermakers, W. Baeyens, P. Quevauviller, M. Horvat, Trends Anal. Chem. 24 (2005) 383. [20] R. Falter, H. Hintelmann, P. Quevauviller, Chemosphere 39 (1999) 1039. [21] H. Hintelmann, Chemosphere 39 (1999) 1093. [22] I. Rodriguez Pereiro, A. Carro Diaz, Anal. Bioanal. Chem. 372 (2002) 74. [23] J. Sanz, A. de Diego, J.C. Raposo, J.M. Madariaga, Anal. Chim. Acta 486 (2003) 255. [24] C. Maggi, M.T. Berducci, J. Bianchi, 8th European Meeting on Environmental Chemistry (EMEC8), Inverness 5–8 December 2007. [25] M.T. Berducci, C. Maggi, A. Cozzolino, M. Dattolo, S. Mariotti, F. Novelli, J. Bianchi, 4th International Conference on Trace Element Speciation in Biomedical Nutritional and Environmental Sciences, Munchen 25–29 May, 2008. [26] R. Fernandez-Martinez, M.I. Rucandio, XIII International Conference on Heavy Metals in the Environment, Brasil, 2005. [27] US EPA Method 7473: Mercury in solids and solutions by thermal decomposition, amalgamation, and atomic absorption spectrophotometry – January 1998 Epa 7443. [28] BCR-580 Report, Community Bureau of Reference, Brussels, Belgium. [29] IAEA-405 Report, International Atomic Energy Agency, Monaco. [30] SRM-2976 Report, National Institute of Standard and Technology NIST, Canada. [31] SRM-2977 Report, National Institute of Standard and Technology NIST, Canada. [32] DOLT-3 Report, National Research Council NCR, Canada. [33] DORM-2 Report, National Research Council NCR, Canada. [34] J.C. Miller, J.N. Miller, Statistical for Analytic Chemistry, 2nd ed., Ellis Horwood, London, UK, 1988. [35] R. Scerbo, C. Barchigiani, Environ. Technol. 19 (1998) 339. [36] S. Dalla Riva, M.L. Abelmoschi, E. Magi, F. Soggia, Chemosphere 56 (2004) 59. [37] L. Lambertsson, M. Nilsson, Environ. Sci. Technol. 40 (2006) 1822. [38] I. Ipolyi, P. Massanisso, S. Sposato, P. Fodor, R. Morabito, Anal. Chim. Acta 505 (2004) 145. [39] N. Mikac, Z. Kwokal, D. Martinicic, M. Branica, Sci. Total Environ. 184 (1996) 173. [40] D. Claisse, D. Cossa, J. Bretaudepu-sanjuan, G. Touchard, B. Bombled, Mar. Pollut. Bull. 42 (2001) 329. [41] H.A. Kehring, M. Costa, I. Moreira, O. Malm, Mar. Pollut. Bull. 44 (2002) 1018. [42] M.M. Storelli, R. Giacominelli-Stuffer, G.O. Marcotrigiano, Mar. Pollut. Bull. 44 (2002) 1354. [43] Z. Svobodova, L. Dusek, M. Hejtmanek, B. Vykusova, R. Smid, Ecotoxicol. Environ. Saf. 43 (1999) 231. [44] J.C. Montgomery, B.A. Foster, R.C. Milton, E. Carr, Polar Biol. 13 (1993) 429. [45] E.A. Pakhomov, Antarct. Sci. 10 (1998) 55.