versión impresa ISSN 0103-5053
J. Braz. Chem. Soc. vol.23 no.3 São Paulo marzo. 2012
Ronaldo J. VivoneI; Maria Luiza D. P. GodoyI; José Marcus Godoy*,I,II; Guaciara M. SantosIII
IInstituto de Radioproteção e Dosimetria, Comissão Nacional de Energia Nuclear, CP 37750, Barra da Tijuca, 22643-970 Rio de Janeiro-RJ, Brazil
IIDepartamento de Química, Pontifícia Universidade Católica do Rio de Janeiro, Rua Marquês de São Vicente 225, Gávea, 22453-900 Rio de Janeiro-RJ, Brazil
IIIKeck-CCAMS Facility, Earth Systems Science, B321 Croul Hall, University of California, Irvine, CA 92697-3100, USA
A 234U and 230Th determination method based on an extraction chromatographic separation followed by ICP-MS with quadrupole (ICP-QMS) was developed. For authigenic carbonates, a second separation step with ion exchange chromatography in a HNO3 solution was added. These methods were applied to seven fossil deep-water coral and two authigenic carbonate samples from the continental slope of the Campos Basin - Brazil. The ages determined for the fossil corals samples from the same sediment core ranged from 9 to 202 ky with a 1% uncertainty, consistent with the values determined by 14C dating and with those determined by 230Th/234U using flow injection coupled to an ICP-QMS. One of the authigenic carbonates analyzed presented an age of approximately 80 ky. The other sample exhibited a 230Th/234U activity ratio close to equilibrium and out of the application range of the method.
Keywords: 230Th/234U dating, ICP-MS, fossil deep-water corals, authigenic carbonates
Foi desenvolvido um método para determinação de 234U e 230Th empregando uma separação por extração cromatográfica seguida por medidas com ICP-MS quadrupolar (ICP-QMS). Para carbonatos autigênicos, uma segunda etapa de separação, por cromatografia de troca-iônica em meio nítrico, foi incluída. Esses métodos foram aplicados a sete amostras de fósseis de corais de águas profundas e duas amostras de carbonato autigênicos do talude continental da Bacia de Campos. As idades determinadas para as amostras de fósseis de corais variaram entre 9-202 kanos com uma incerteza de 1%, consistente com os valores determinados pela datação com 14C, bem como com as idades 230Th/234U determinadas por FIAS-ICP-QMS. Uma das amostras de carbonato autigênico analisada apresentou uma idade de aproximadamente 80 kanos. A outra amostra apresentou uma razão 230Th/234U próximo do equilíbrio e fora da faixa de aplicação do método.
The Brazilian continental margin contains substantial hydrocarbon reserves and thus has been the subject of research related to biostratigraphy, paleoecology and paleoceanography. Sequences of appearing and disappearing fossil deep-water corals have been found in sediments sampled from the Campos Basin continental slope.1
Deep-water fossil corals could serve as records with high temporal resolution for the reconstruction of rapid changes in past ocean circulation.2 The earth's climate changes rapidly on the geological time scale, and the bottom of the sea plays an important role as a paleoceanographic record.3,4 Some authors have suggested that the deep-water corals are well suited to the study of rapid climate change.2,5,6 Past deep circulation patterns have been reconstructed using high-resolution records of deep-water corals obtained by 14C or U/Th dating.
According to Miller,7 seabed hydrocarbon seeps have been widely used as an indicator of deep hydrocarbon accumulations. Detailed study of the geochemical, geological and geochronological aspects of seep-related features can provide important information on the origin, evolution and thermal processes of hydrocarbon migration, thus contributing to a better understanding of petroleum systems.
Over the past few years, several studies on the underwater environment and fossil records rocks have emphasized the importance of carbonate formation associated with the age of hydrocarbon seepage and microbial activity.8 Because authigenic carbonate precipitation most commonly results from anaerobic methane oxidation, its carbon source is petrogenic and thus conventional 14C dating is not applicable. However, several studies have demonstrated the applicability of the 230Th/234U method for these purposes.9-11
Depending on the sample age, the 230Th activity can be low, which causes a high uncertainty in age determination by radiochemical methods, even after long counting times. Thus, 230Th/234U in these samples (corals and authigenic carbonates) can be determined using mass spectrometry.12-15
In this work, the 230Th/234U determination method by ICP-MS was performed by the Institute for Radioprotection and Dosimetry/Brazilian Nuclear Commission (IRD/CNEN), using on-line chemical separation (FIAS) employing UTEVA cartridges (Eichrom Co.).16 The method had to be adapted for application to samples of authigenic carbonates, because low chemical yields were observed, probably due to large amounts of iron.
The aim of the present work is to develop an analytical method for 230Th/234U dating of authigenic carbonate and deep-water coral samples with improved precision and accuracy over the on-line separation (FIAS) method currently used by the IRD/CNEN. The calculated ages were compared with values obtained by other techniques, such as 14C dating by mass spectrometry coupled to a particle accelerator (AMS) by the Laboratory of the University of California (USA) and 230Th/234U dating with a flow injection system coupled to an ICP-QMS method by the Institute for Radioprotection and Dosimetry (Brazil).
The seven fossil coral and two authigenic carbonate samples, which originated from the continental slope of the Campos Basin - Brazil, were obtained from Petrobras.
To remove the U and Th fractions on the coral samples carried by the Fe-Mn coatings and clay particles, mechanical cleaning was first performed, followed by ultrasonic washing with 1 mol L-1 HNO3 for 20 min. A second chemical cleaning procedure adapted from Lomitschka and Mangini12 was then performed. One gram of cleaned coral sample was ground in an agate mortar and dissolved in 30 mL of 3 mol L-1 HNO3. When some residue remained, it was dissolved completely with an acid mixture composed by HNO3 plus HF, evaporated to eliminate HF, and the obtained residue re-dissolved with 3 mol L-1 HNO3 and reserved for its U and Th content determination by ICP-QMS. These results were used to correct the 234U and 230Th values on the carbonate phase for the detritic contribution by applying the 232Th concentration on the carbonate phase and the U/Th ratio on the residual phase. Thorium-230 and 234U concentrations on the leaching solution were determined according to the method illustrated in Figure 1. The 230Th, 234U and 238U values, corrected for the detritic contribution, were used to calculate the coral ages according to equation 3.
The external layer of the authigenic carbonate samples was removed, and the remaining part was ground in an agate mortar and sieved into four fractions < 170, 170-80, 80-42 and 42-24 mesh. One gram of each fraction was leached with 7 mol L-1 HNO3 in order to yield the carbonate phase. The remaining residual fraction was totally dissolved with HNO3 plus HF mixture, and then 238U and 232Th were determined by ICP-QMS. The 230Th, 232Th, 234U and 235U results on the carbonate phase were used to construct both Rosholt and Osmond isochrones17 and to obtain the 230Th/234U and 234U/238U activity ratios. The results obtained using these two different approaches were applied to equation 3.
The uranium and thorium elemental and isotope determinations were performed using a Perkin-Elmer ELAN 6000 ICP-QMS instrument using the conditions described in Table 1. The 233U and 229Th spikes used are traceable to BIPM and were diluted according to the Radionuclides Metrology Section of the Institute for Radioprotection and Dosimetry (SEMRA/IRD).
Mass bias was corrected by applying equation 1, where β was determined from NIST CRM U020A (1 ng mL-1) analysis and equation 2:18
To reduce the mass discrimination effect, the 234U/235U isotope ratio was measured instead of the 234U/238U isotope ratio. To calculate the 234U/238U isotope ratio, the measured 234U/235U isotope ratio was multiplied by the 235U/238U isotope ratio (0.0072526). To calculate the 230Th/234U mass concentration ratio, the obtained 230Th concentration was divided by the product of the 234U/238U isotope ratio multiplied by the 238U mass concentration. To obtain the 230Th/234U activity ratio, the calculated mass concentration ratio was multiplied by the ratio of the specific activities of the isotopes in question.
The 234U/238U and 230Th/234U activity ratios were calculated. Based on the obtained values, the ages were determined according to equation 3,19
where λ230 is the 230Th decay constant and, λ234 is the 234U decay constant.
The developed methods were tested for 234U and 230Th determinations of the NIST SRM 4357 standard reference ocean sediment material. Three 2 g aliquot samples were used for each method.
Results and Discussion
For accurate 234U and 230Th determination in the fossil coral and authigenic carbonate samples, a separation procedure was needed to eliminate the main matrix elements and to obtain a sample solution with a chemical composition similar to that of standards.20 An off-line separation procedure was applied based on extraction chromatography with 2 mL UTEVA (Eichrom Co.) columns21-26 (Figure 1). The elution curves of U and Th with 0.1 mol L-1 HCl, 0.1 mol L-1 HCl + 0.1 mol L-1 HF, and 0.1 mol L-1 HCl + 0.01 mol L-1 HF, were tested. The feed solution (40 mL of 3 mol L-1 HNO3 + 0.025 mol L-1 Al(NO3)3) was added to 1000 ng of Unat and 1000 ng of Thnat. Based on the obtained results (Figure 2), 10 mL of 0.1 mol L-1 HCl + 0.01 mol L-1 HF was used as the U and Th eluent to avoid higher HF concentrations and to obtain 100% recovery for both elements. This condition ensures that Th and U are quantitatively retained and eluted during separation from the matrix. The U and Th decontamination factors (DF) from Ca, Fe, Al, Mg and Sr, which are representative of the major matrix elements, are shown in Table 2. The obtained decontamination factors are considered satisfactory, because they are equal to or greater than 103.27
Due to the low thorium chemical yield obtained, the initial tests with authigenic carbonate samples revealed the need for a matrix separation step before chromatographic extraction with the UTEVA column (Figure 3). Based on previous work (Godoy et al.28), ion exchange chromatography with DOWEX 1X8 was tested. Using 40 mL of 7 mol L-1 HNO3 + 0.025 mol L-1 Al(NO3)3 as feed solution containing 1000 ng of U and 1000 ng of thorium, the elution curves of Th with 6 mol L-1 HCl + 0.26 mol L-1 HF, 1 mol L-1 HCl + 0.26 mol L-1 HF and 0.1 mol L-1 HCl + 0.26 mol L-1 HF were determined. Based on the obtained results (Figure 4), 25 mL of 6 mol L-1 HCL + 0.26 mol L-1 HF was chosen as the Th eluent. Thorium was strongly adsorbed, whereas uranium was not retained by the DOWEX 1X8 resin and instead was distributed between the feed and washing solutions. Both solutions were collected together for U determination, followed by a second separation step with UTEVA columns similar to that applied for the coral samples.
The obtained 234U and 230Th concentrations for the NIST SRM 4357 certified reference marine sediment sample using both methods are shown in Tables 3 and 4. The results are within the confidence interval for both isotopes. For these samples, the 230Th achieved accuracy (2.8 and 2.2%), by applying this off-line separation method are better than those obtained by Godoy et al.,16 whose values were above 10%, but within the 95% confidence interval of the certified value. The 230Th precisions (1.9 and 3.1%) are similar to that (1.5%) reported by Godoy et al.16 Although 234U does not have a certified reference value, the precision and accuracy values for this isotope using the proposed method were similar to those using on-line methods.16
For the fossil deep-water corals samples from Campos Basin, the 230Th/234U and 234U/238U activity ratios as well as the calculated ages are shown in Table 5. Taking into account the peak areas, the residual fraction contribution and the uncertainty related to the spikes, the uncertainties associated with coral were about 1%. Table 6 shows the fossil coral ages determined using the proposed method, with the previous method based on flow injection, and by 14C AMS dating at the University of California, Irvine, United States. To verify if there were statistically significant differences between the obtained results by the three methods, ANOVA tests were applied, as shown in Table 7. Because the calculated F (0.50) value was lower than the critical F (3.89) value, no significant differences were found among the three methods.
The authigenic carbonate samples contained detritic contamination not only at the surface, but throughout the entire sample. Therefore, usual cleaning procedures, such as mechanical and chemical removal of the surface layers, did not generate a final sample with a low 232Th content, as observed for the coral samples (range of ng g-1). The procedure described by Ku29 was applied with sample fractionation by size to generate independent sub-samples. Based on the obtained results, two different isochrons were constructed: 230Th/232Th vs. 234U/232Th and 234U/232Th vs. 238U/232Th for the Rosholt plots (Figure 5) and 230Th/238U vs. 232Th/238U and 234U/238U vs. 232Th/238U for the Osmond plots (Figure 6). The 230Th/234U and 234U/238U activity ratios were obtained using the slopes of the Rosholt plots and the intercepts of the Osmond plots. The obtained values were statistically equivalent: (0.542 ± 0.014) 230Th/234U and (1.1004 ± 0.0028) 234U/238U and (0.564 ± 0.046) 230Th/234U and (1.0976 ± 0.0036) 234U/238U for the Osmond and Rosholt plots, respectively. Applying these results to equation 3, the age of this authigenic carbonate sample was (83.6 ± 0.1) ky according to the Osmond plot and (88.8 ± 0.1) ky based on the Rosholt plot. According to Ludwig,17 the Osmond plot provides a more realistic value for samples with a high 232Th content. Furthermore, because 232Th and 234U each appear in both of the isochrons, the results of the regressions are correlated in a complex way. However, these correlations are never taken into account when calculating the ages or age-errors. Therefore, a simultaneous solution using all three of the ratios for the modified Osmond ratios is needed to generate a regression for an isochron line in three dimensions. The calculated age, determined using Isoplot 3.7 software,17 used to construct the 3D 232Th/238U-230Th/238U-234U/238U isochron was (78.66 ± 0.17) ky.
The other authigenic carbonate sample yielded a 230Th/234U value close to one. Consequently, the corresponding age was not calculated, indicating that it is close to or older than 300 ky.
Two 234U and 230Th determination methods based on extraction chromatographic separation followed by ICP-QMS measurements were developed. These methods were validated for fossil deep-water corals and authigenic carbonates by using certified reference marine sediment.
Samples of fossil corals from the same sediment core and from authigenic carbonates were analyzed and dated. The obtained fossil coral ages agree with other values determined using 230Th/234U flow injection separation coupled to ICP-QMS and with 14C-AMS.
The developed methods are equivalent in terms of precision and more accurate than the previously applied method based on flow injection separation coupled to an ICP-QMS. In addition, the proposed method has the following advantages: final solution free of salts and with a chemical composition similar to ICP-QMS standard solutions, less accumulation of salts in the interface of the equipment (skimmer and sampler cones) and applicability to more complex matrices, such as authigenic carbonates.
The authors are particularly grateful to Dennis Miller and Renato Kowsmann (CENPES/Petrobras) for the fossil coral and the authigenic carbonate samples and to Kenneth Ludwig (Berkeley Geochronological Center) for the Isoplot software copy.
1. Mangini, A.; Godoy, J. M.; Godoy, M. L.; Kowsmann, R. O.; Santos, G. M.; Ruckelshausen, M.; Schroeder-Ritzrau, A.; Wacker, L; Earth Planet. Sci. Lett. 2010, 293, 269. [ Links ]
2. Schroder-Ritzrau, A.; Mangini, A.; Lomitschka, M.; Earth Planet. Sci. Lett. 2003, 216, 399. [ Links ]
3. Wilson, R. C. L.; Drury, S. A.; Chapman, J. L.; J. Quaternary Sci. 2000, 15, 843. [ Links ]
4. Alley, R. B.; Marotzke, J.; Nordhaus, W. D.; Overpeck, J. T.; Peteet, D. M.; Pielke Jr., R. A.; Pierrehumbert, R. T.; Rhines, P. B.; Stocker,T. F.; Talley, L. D.; Wallace, J. M.; Science 2003, 299, 2005. [ Links ]
5. Adkins, J. F.; Cheng, H.; Boyle, E. A.; Druffel, E. R. M.; Edwards, R. L.; Science 1998, 280, 725. [ Links ]
6. Goldstein, S. J.; Lea, D. W.; Chakraborty, S.; Kashgarian, M.; Murrell, M. T.; Earth Planet. Sci. Lett. 2001, 193, 167. [ Links ]
7. Miller, D. J.; MSc Dissertation, COPPE/UFRJ, Rio de Janeiro, Brasil, 2010. (http://capesdw.capes.gov.br/capesdw/resumo.html?idtese=200679431001017028P0). [ Links ]
8. Camoin, G. F.; Sediment. Geol. 1999, 126, 1. [ Links ]
9. Aharon, P.; Schwarcz, H. P.; Roberts, H. H.; Geol. Soc. Am. Bull. 1997, 109, 568. [ Links ]
10. Watanabe, Y.; Nakai, S.; Hiruta, A.; Matsumoto, R.; Yoshida, K.; Earth Planet. Sci. Lett. 2008, 272, 89. [ Links ]
12. Lomitschka, M.; Mangini, A.; Earth Planet. Sci. Lett. 1999, 170, 391. [ Links ]
11. LaMont, S. P.; Hall, G.; J. Radioanal. Nucl. Chem. 2005, 264, 423. [ Links ]
13. Pin, C.; Joannon, S.; J. Anal. At. Spectrom. 2001, 16, 739. [ Links ]
14. Pons-Branchu, E.; Hillaire-Marcel, C.; Deschamps, P.; Ghaleb, B.; Sinclair, D. J.; Geochim. Cosmochim. Acta 2005, 69, 4865. [ Links ]
15. Watanable, Y.; Nakai, S.; Microchim. Acta 2007, 156, 289. [ Links ]
16. Godoy, M. L. D. P.; Godoy, J. M.; Kowsmann, R.; Santos, G. M.; Cruz, R. P.; J. Environ. Radioact. 2006, 88, 109. [ Links ]
17. Ludwig K. R.; Isoplot version 3.7, User's Manual, Berkeley Geochronology Center Special Publication 4, 2008. [ Links ]
18. Ingle, C. P.; Sharp, B. L.; Horstwood, M. S. A.; Parrish R. R.; Lewis, D. J.; J. Anal. At. Spectrom. 2003, 18, 219. [ Links ]
19. Ivanovich, M.; Latham, A. G.; Ku, T. L. In Uranium-series Disequilibrium; Ivanovich, M.; Harmon, R.S., eds.; Clarendon Press: Oxford, 1992, ch. 3. [ Links ]
20. Hill, S. J.; Fisher, A.; Foulkes, M. In Inductively Coupled Plasma Spectrometry and its Application; Hill, S. J., ed.; Blackwell Publishing Ltd.: Oxford, 2007, ch. 3. [ Links ]
21. Horwitz, E. P.; Chiarizia, R.; Dietz, M. L.; Diamond H.; Anal. Chim. Acta 1992, 266, 25. [ Links ]
22. Adriaens, A. G.; Fasset, J. D.; Kelly, W. R.; Simons D. S.; Adams, F. C.; Anal. Chem. 1992, 64, 2945. [ Links ]
23. Yokoyama, T.; Makishima, A.; Nakamura, E.; Anal. Chem. 1999, 71, 135. [ Links ]
24. Bool, R. A.; Schweitzer, G. K.; Garber, R. W.; J. Radioanal. Nucl. Chem. 1997, 220, 201. [ Links ]
25. Maxwell, S.L.; Nichols, S.T.; Radioact. Radiochem. 2000, 11, 46. [ Links ]
26. Ketterer, M. E.; Jordan, J. A.; Szechenyi, S. C.; Hudson, D. D.; Layman, R. R.; J. Anal. At. Spectrom. 2000, 15, 1569. [ Links ]
27. Grate, J. W.; Egorovi, O. B.; Fiskum, S. K.; Analyst 1999, 124, 1143. [ Links ]
28. Godoy, M. L. D. P.; Godoy, J. M.; Roldão, L. A.; Tauhata, L.; J. Environ. Radioact. 2009, 100, 613. [ Links ]
29. Ku, T-L. In Quaternary Geochronology: Methods and Applications; Noller, J. S.; Sowers, J. M.; Lettis, W. R., eds.; AGU Reference Shelf 4, 2000, ch.5. [ Links ]
Submitted: July 10, 2011
Published online: January 31, 2012