Anal. Chem. 2006, 78, 304-309
Purifying Barite for Oxygen Isotope Measurement by Dissolution and Reprecipitation in a Chelating Solution Huiming Bao*
Department of Geology & Geophysics, Louisiana State University, Baton Rouge, Louisiana 70803
In the laboratory, barite precipitated from a solution with a high nitrate/sulfate ratio can have a significant amount (up to 28% by weight) of nitrate occluded in barite crystals that cannot be simply washed away. The impurity poses a serious problem for an accurate measurement of the oxygen isotope compositions for atmospheric sulfate, since atmospheric nitrate bears extremely positive ∆17O and δ18O values. Currently available methods for removing the occluded nitrate are either ineffective or not tested for oxygen isotope exchange. Here, I report a DTPA (a chelating solution) dissolution and reprecipitation (DDARP) method that is simple and effective in removing nitrate and other contaminants in barite. A series of barite dissolution and reprecipitation experiments that utilize 17O-anomalous solutions or barite crystals is conducted to examine the effect on oxygen isotopes during various treatments. It is established that no oxygen isotope exchange occurs between sulfate and water during DDARP treatment at two experimental temperatures (21 and 70 °C). Occlusion of DTPA itself in barite is negligible. Upon acidification, barite reprecipitation from a DTPA solution is quantitative (∼100%). Partially extracted barite may have slightly lower δ18O or δ34S values than the originals but no effect on ∆17O values. It is also demonstrated that heavily nitrate-contaminated barite samples are free of nitrate occlusion after two dissolution-reprecipitation cycles. One of the most commonly used methods for measuring the oxygen isotope compositions of sulfate is to first precipitate out sulfate from a solution as barite (BaSO4). This is normally achieved by adding droplets of saturated BaCl2 solution into an acidified sulfate-bearing solution. Due to the low solubility of barite, the sulfate ion is quantitatively removed from the solution. I will refer to this precipitated barite in the laboratory as synthetic barite throughout the text. There are three commonly used analytical approaches for oxygen isotope measurement: (1) off-line combustion of BaSO4 crystals with pure graphite powder, producing CO2 (minor CO produced in the process is converted to CO2 by electric discharge) for δ18O measurement;1,2 (2) combustion of BaSO4 in a graphite furnace, producing CO via a thermal conversion * Corresponding address: E-mail:
[email protected]. Phone: 225-578-3419. (1) Rafter, T. A. N. Z. J. Sci. 1967, 10, 493-510. (2) Sakai, H.; Krouse, H. R. Earth Planet. Sci. Lett. 1971, 11, 369-373.
304 Analytical Chemistry, Vol. 78, No. 1, January 1, 2006
elemental analyzer (TCEA) for δ18O measurement in a continuousflow mode;3 and (3) off-line CO2-laser fluorination of BaSO4, producing nonquantitative O2 for ∆17O measurement.4 All these methods will release oxygen from non-sulfate compounds if the barite contains oxygen-bearing impurities. Therefore, high-purity barite is critical to oxygen isotope measurement. Unfortunately, synthetic barite has been known since early 19th century5 to contain variable degrees of impurities, including both anions and cations. There were extensive studies and debates on the nature of the occlusion of foreign ions in barite in the early 20th century.6-8 Proposed mechanisms include coprecipitation, solid solution, adsorption, or simple trapping of mother solutions.7,9-11 The potential effect of the occluded impurities on barite’s oxygen isotope measurement, however, was never dealt with seriously in the stable isotope community. However, recently new interest in the oxygen isotope compositions of sulfate in atmospheric deposition and surface water, including rainwater, aerosols, desert soils, and river water, has presented situations where sulfate has to be extracted from solutions that are rich in other ions. Barite precipitated from such solutions may contain all kinds of cations and anions, among which non-sulfate oxyanions can create problems for accurate sulfate oxygen isotope analyses. Among the non-sulfate oxyanions, nitrate poses the worst problem, because (1) among all examined anions, nitrate is known to be the most easily occluded anions in barite crystals during precipitation;11 and (2) nitrate, especially that of atmospheric origin, has extremely high δ18O and ∆17O values, up to + 90 and + 33‰, respectively.12-16 Even low levels of nitrate contamination will (3) Kornexl, B. E.; Gehre, M.; Hofling, R.; Werner, R. A. Rapid Commun. Mass Spectrom. 1999, 13, 1685-1693. (4) Bao, H.; Thiemens, M. H. Anal. Chem. 2000, 72, 4029-4032. (5) Turner, E. Philos. Trans. R. Soc. London 1829, 119, 291-299. (6) Schneider, F.; Rieman, W., III. J. Am. Chem. Soc. 1937, 59, 354-357. (7) Karoglanov, Z. Z. Anal. Chem. 1936, 106, 129-146. (8) Walden, G. H., Jr.; Cohen, M. U. J. Am. Chem. Soc. 1935, 57, 2591-2597. (9) Huttig, G. F.; Menzel, E. Z. Anal. Chem. 1926, 68, 343-358. (10) Walton, G.; Walden, G. H., Jr. J. Am. Chem. Soc. 1946, 68, 1742-1750. (11) Nichols, M. L.; Smith, E. C. J.Phys. Chem. 1941, 45, 411-421. (12) Bao, H.; Marchant, D. R. AGU Fall Meeting-San Francisco 2004 2004, H52B-04. (13) Michalski, G.; Scott, Z.; Kabiling, M.; Thiemens, M. H. Geophys. Res. Lett. 2003, 30, 1870; 1810.1029/2003GL017015. (14) Michalski, G.; Bohlke, J. K.; Thiemens, M. Geochim. Cosmochim. Acta 2004, 68, 4023-4038. (15) Michalski, G.; Bockheim, J. G.; Kendall, C.; Thiemens, M. Geophys. Res. Lett. 2005, 32. (16) Kendall, C. In Isotope Tracers in Catchment Hydrology; Kendall, C., McDonnell, J. J., Eds.; Elsevier Science: New York, 1998. 10.1021/ac051568z CCC: $33.50
© 2006 American Chemical Society Published on Web 11/25/2005
impair the true sulfate oxygen isotope compositions. For example, 5% (mole fraction) oxygen from occluded nitrate can add up to +4.5 and +1.5‰ false positive δ18O and ∆17O values, respectively, to the barite. The seriousness of this nitrate occlusion problem for oxygen isotope measurement was first pointed out by Michalski, who noticed that when the nitrate/sulfate ratio is ∼1:1 in a solution, the precipitated barite contains ∼5% by weight of nitrate in barite crystals that cannot be washed away (Greg Michalski, personal communication, 2004). Several synthetic barite samples that have been examined in our laboratory contain NO3-/SO42- weight ratios up to 0.28 (see Table 8). A previous effort17 using a nitron (C20H16N4)-based method to precipitate nitrate out of solution before the precipitation of barite demonstrated that nitrate cannot be effectively removed from the solution, and the necessity to independently obtain concentration and oxygen isotope data for the occluded nitrate for a mass balance calculation introduces an unacceptable error ((0.25 to 0.35‰) for the calculated ∆17O value for pure sulfate component, considering that the actual ∆17O measurement error for pure barite can often be less than 0.05‰, and many important ∆17O variations among natural sulfates are within 0.50‰. Here, I report a simple purification method for synthetic barite samples precipitated from solutions in the laboratory. Instead of removing nitrate or other ions from a sulfate-bearing solution before BaSO4 precipitation, a repeated dissolution and reprecipitation approach is used to purify the synthetic BaSO4 after its precipitation. It is generally known that sulfate is a nonlabile oxyanion that does not readily exchange oxygen isotope with ambient water in solution under surface conditions. Diethylenetriaminepentaacetic acid (DTPA), [(HO2CCH2)2NCH2CH2]2NCH2CO2H, is an effective metal ion chelator. Theoretically, each DTPA molecule can bond with one metal ion (e.g., Ba2+) under a high-pH condition (pH >12).18 DTPA has previously been tested and shown to be effective at dissolving barite at different temperatures and concentrations.19 In our laboratory, it was found that, by simply acidifying the DTPA solution, the Ba2+ will recombine with the dissolved SO42- and new BaSO4 crystals precipitate. It is apparent that the reprecipitation of BaSO4 occurs because acid destroys the chelating ability of DTPA on Ba2+. Since the BaSO4 reprecipitated in a solution with much less nitrate content than the original solution, it should have much less occluded nitrate in its structure than before. Repeated dissolution and reprecipitation in new DTPA solutions can, therefore, remove all occluded contaminants from initial synthetic barite. The DTPA dissolution and reprecipitation (DDARP) method has not been investigated for its effect on the oxygen isotope composition of sulfate in barite. Potential effects could include (1) oxygen isotope exchange between sulfate and water, (2) the extent of the occlusion of DTPA itself, which has a C/N/O ratio of 14:3:10, in reprecipitated barite, and (3) the effect of partial dissolution or reprecipitation on oxygen isotope compositions. In this study, the effectiveness of DDARP method and its associated isotope effects are thoroughly examined in a set of (17) Bao, H. Chem. Geol. 2005, 214, 127-134. (18) Martell, A. E.; Hancock, R. D. Metal Complexes in Aqueous Solutions (Modern Inorganic Chemistry); Plenum Press: New York, 1996. (19) Putnis, A.; Juntarosso, J. L.; Hochella, M. F. Geochim. Cosmochim. Acta 1995, 59, 4623-4632.
experiments that utilize the ∆17O parameter. The ∆17O is defined as the deviation of a measured δ17O value from one predicted from the δ18O value on the basis of the terrestrial fractionation line. In the logarithmic form, it is ∆17O ) δ17O - 1000[(1 + δ18O/1000)0.52 - 1. Traditionally, oxygen isotope exchange experiments have been monitored using 18O-labeled water or minerals. The δ18O parameter will be affected not only by the fraction of oxygen that has exchanged between two oxygen-bearing species (e.g., barite and water) but also by nonequilibrium exchange processes or uncertainties of the fractionation factor in the system. The ∆17O, however, is a rather conservative parameter; it provides an accurate quantification of the fraction of oxygen that has exchanged, regardless of whether the system reaches isotopic equilibrium. I have adopted an approach for all experiments ensuring one of the phases always has a zero ∆17O value, while the other has a large positive ∆17O value. The ∆17O method has been used to calibrate isotope geothermometers in silicate-fluid systems.20,21 This study represents the first sulfate-water exchange experiment that utilizes the parameter of ∆17O. EXPERIMENTAL SECTION Samples and Materials. Three normal (with no 17O anomaly) barite samples were used in the experiments: (1) LSU-BaSO4 (CAS 7727-43-7, Fisher Scientific), (2) HP-BaSO4 (high-purity, 99.998%, Aldrich), and (3) Ohio-S, an in-house reference barite sample that has both δ18O and δ34S values significantly lower than those of LSU-BaSO4. Four 17O-anomalous barite samples were extracted from natural sulfate samples from an Oligocene ash bed in northwestern Nebraska,22 and from soils in the McMurdo Dry Valleys, Antarctica. Any nitrate contamination was removed from these samples before being used for isotope-exchange experiments. The DTPA solution used for dissolution experiments was made by dissolving DTPA (98%, CAS 67-43-6, Alfa Aesar) in a 1 M NaOH solution. All DTPA solution used is 0.05 M except for specific cases listed below. The 17O-anomalous water used in experiments was produced by mixing 17O-enriched water with measured amounts of 18-Ω (doubly deionized) water. The ∆17O and δ18O values of the labeled water were determined to be + 6.9 and + 6.5‰, respectively. DDARP Method. Fine barite crystals were mixed with a DTPA solution (0.05 M in 1 M NaOH) in a 1:500 weight ratio (i.e., ∼30 mg in 15 mL of DTPA solution). The mixture was shaken vigorously at room or slightly elevated (
