Ion-Specific Isotopic Fractionation of Molybdenum during Diffusion in

Jan 20, 2007 - MS, the ratios of the diffusivities of Mo isotopes, D97Mo/. D95Mo, in aqueous solutions have been determined. Diffusion of MoO4. 2- ion...
0 downloads 0 Views 128KB Size
Environ. Sci. Technol. 2007, 41, 1596-1600

Ion-Specific Isotopic Fractionation of Molybdenum during Diffusion in Aqueous Solutions D M I T R Y M A L I N O V S K Y , * ,† DOUGLAS C. BAXTER,‡ AND ILIA RODUSHKIN‡ Division of Applied Geology, Luleå University of Technology, SE-97187 Luleå, Sweden, and Analytica AB, Aurorum 10, SE-97775 Luleå, Sweden

Experiments modeling diffusion of Mo in aqueous solutions have been performed and, using multicollector ICPMS, the ratios of the diffusivities of Mo isotopes, D97Mo/ D95Mo, in aqueous solutions have been determined. Diffusion of MoO42- ions in solution was concomitant with Mo isotopic fractionation with D97Mo/D95Mo ) 0.99988 ( 0.00004 (2σ for n ) 3). In contrast, during diffusion of Mo polyanions, such as Mo7O246- and Mo8O264-, no measurable isotope fractionation has been found with D97Mo/D95Mo ) 1.00000 ( 0.00002 (2σ for n ) 3). These results indicate the need for due consideration to Mo speciation when attempting to interpret the role of diffusive fluxes in the formation of Mo isotopic signatures in nature. They also raise the possibility that the various chemical forms of other transition metals may be characterized by species-specific isotopic fractionation effects during physicochemical reactions.

Introduction Molybdenum is an element of importance in environmental science and geochemistry. Biochemically, molybdenum is a cofactor in enzymes, responsible for the uptake of nitrogen from both nitrogen gas and nitrate in certain organisms. It also participates in important oxygen-atom transfer reactions at low redox potential (1). Geochemically, Mo is highly soluble and relatively unreactive in oxygenated waters and, hence, is a nominally conservative element in the oceans (2, 3). In contrast, Mo is readily removed from solution under euxinic conditions, so that Mo authigenic enrichments in sediments are considered an indication of strongly reducing settings. Molybdenum is an example of an element whose distribution in marine and freshwater systems is proposed to depend on diffusive transport in sediment porewaters (4-8). Removal of Mo from solution in the reducing zone of sediment leads to formation of the concentration gradient and diffusive flux of the element across the sediment-water interface. Over the course of the past few years a considerable amount of data has been reported revealing mass fractionation of Mo isotopes in environmental samples (3, 9-15). It has been shown that Mo isotopic fractionation recorded in sediments can be used efficiently in inferences on mechanisms of Mo immobilization and in reconstructions of the oxygen content of bottom waters in the past. However, these * Corresponding author phone: +46-920491384; fax: +46920491199; e-mail: [email protected]. † Luleå University of Technology. ‡ Analytica AB. 1596

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 5, 2007

results also showed a number of uncertainties in the interpretation of Mo isotope signatures. In particular, the mechanism of Mo removal and formation of Mo isotope signatures in suboxic systems is unclear and, hence, requires further study (6, 16). Anbar (3) raised the suggestion that Mo isotopic fractionation recorded in suboxic marine sediment environments might originate from differences in the relative fluxes of Mo isotopes across the sediment-water interface. However, to the best of our knowledge, no attempt to determine the extent of Mo isotope fractionation due to diffusion in water has been made. Mass-dependent kinetic isotope fractionation during chemical diffusion in aqueous solutions is theoretically modeled (17) and experimentally documented both for lighter (18-20) and heavier elements (21-24). Fractionation by diffusion results from the higher velocity of the lighter isotope relative to the heavier isotope. Recent studies have demonstrated that diffusion in aqueous solution can cause changes in isotope ratios of such elements as Fe, Zn, Li, and Cl in excess of -0.15‰ per atomic mass unit (23, 24). On the other hand, Richter et al. (24) observed no measurable effect for Mg in water in spite of the fact that codiffusing Cl in the same experiments was isotopically fractionated. This finding reveals a need to quantify isotope fractionation associated with diffusion for each element individually, rather than extrapolate the extent of the effect from known experimental data for other elements. In the aforementioned works, no attempt has been made to evaluate the isotope fractionation by diffusion for different chemical forms of the same element. This study was undertaken to evaluate Mo isotope fractionation during diffusion in aqueous solutions. The experiments were designed to allow monitoring diffusion and concomitant isotopic fractionation for distinct Mo ions. To this end, neutral (pH ∼ 7) and acidic (pH ∼ 0.5) water solutions were used in the experiments. Under these pH conditions in an oxic environment the dominant Mo species in the solutions are either molybdate oxy-anions, MoO42-, in the former case, or various Mo polyanions including heptamolybdate, Mo7O246- and octamolybdate, Mo8O264- in the latter case (25, 26). Relatively large diffusion-driven isotopic fractionation has been observed for MoO42-, whereas no effect has been found for the Mo species in acidified water. Potential implications of these data are briefly discussed.

Experimental Section Diffusion Cell and Experimental Design. The experimental approach employed in this study largely followed that used by Rodushkin et al. (23). A sketch of the diffusion cell we used to initially confine the solute on one side of a boundary is shown in Figure 1. A 0.5 mL aliquot of Na2MoO4 solution with Mo concentration of 0.40 ( 0.03 M either in water or in 0.33 M HNO3 was carefully pipetted into an acid-washed, polypropylene tube, and ∼3 mL of fine-grained pure quartz sand was added to completely cover the solution. The sand provided a mechanical barrier that prevented mixing when 8 mL of pure solvent, water or 0.33 M HNO3, was dispensed. After 68 h had elapsed since the start of the experiments, 0.5 mL layers of solvent were sampled and measured for Mo elemental and isotopic abundances. Sampling of 0.5 mL layers of the solutions has been performed using a calibrated 1 mL pipet by hand with the aid of a simple support; the latter is necessary to minimize potential manual disturbances. The pipet tip was submerged into solution at a depth of ∼1 mm ,and 0.5 mL of the solution was taken. The experiments were conducted at 20 ( 1 °C. 10.1021/es062000q CCC: $37.00

 2007 American Chemical Society Published on Web 01/20/2007

the presentation of results, δ-notation is utilized, as defined by the following relationship:

δ97/95Mo )

[

(97Mo/95Mo)sample

(97Mo/95Mo)standard

]

- 1 1000‰

(1)

where 97Mo/95Mo is the measured ratio for sample or standard, corrected for instrumental mass discrimination using Pd. The initial solution of Mo salt (Na2MoO4‚2H2O) diluted to a concentration of 1 mg l-1 was used as Mo isotope standard in this work.

Results and Discussion

FIGURE 1. Schematic diagram of the diffusion cells.

Chemicals and Reagents. Analytical grade nitric acid (65%, Merck, Darmstadt, Germany) was used throughout the study after additional purification by sub-boiling distillation in a quartz still. Dilution of samples and standards was performed using distilled Milli-Q water (Millipore MilliQ, Bedford, U.S.). The Mo anion salt used in the experiments was Na2MoO4‚2H2O (Merck). Prior to the use of Milli-Q water in the diffusion cells, µl volumes of 0.01 M NaOH were added to the water in order to adjust its pH to ∼7 in order to ensure that the Mo speciation in solution is dominated by MoO42(25, 26). Mass Spectrometry and Data Processing. Prior to Mo isotope ratio measurements, the concentrations of Mo and a suite of potentially interfering elements (Zr, Fe, Ru) in the samples were determined by single collector inductively coupled plasma sector field mass spectrometry (ICPSFMS; Element, Thermo Finnigan, Bremen, Germany). After analysis by ICP-SFMS, the Mo samples were diluted to 1 mg l-1 Mo with 0.33 M HNO3 and spiked with Pd at 0.5 mg l-1, prior to isotope ratio measurements by multiple collector ICPMS (MC-ICPMS; Neptune, Thermo Finnigan, Bremen, Germany). Typical operating conditions for this instrument and the protocol for the measurements are reported in detail elsewhere (27). Low-resolution mode was used in this study with Rpower(5,95%) ∼ 400. Samples and standards were prepared in 0.33 M HNO3 solution and introduced into the plasma through a stable introduction system (Thermo Finnigan) consisting of tandem quartz spray chamber arrangements (cyclone + Scott double pass), a micro-concentric PFA nebulizer, and a peristaltic pump (Perimax 12, SPETEC, Erding, Germany) operating at a flow rate of about 0.25 mL min-1. All analyses were made in a sequence of isotope standard, “unknown” sample, isotope standard and so on. Concentrations of samples and bracketing standards were matched to within 20%. The analyses were conducted in the static mode. Pd was used as a normalizing element to correct for instrumental mass discrimination of Mo isotopes. An exponential model for correction of instrumental mass discrimination was used as described in (27, 28). Typically, the precision of 97Mo/95Mo ratio measurements was ( 0.12‰ at the two standard deviations level. The on-line data processing included calculation of the raw ratios and filtering of outliers by a 2-σ test. Further statistical treatment of the data was performed off-line. For

The oxidation state of molybdenum in oxygenated aqueous solutions is +6 over a broad pH region (25). For the element in this oxidation state, various ionic species in aqueous solutions are formed whose presence and relative amounts depend on the pH as well as on the total molybdenum concentration (25, 29). Previous experimental work has demonstrated that, in aqueous solutions at a pH value of ∼7 and molybdate concentration 0.4 M, the element is present as the tetrahedral monomeric oxy-anion MoO42- (25, 30). This oxy-anion is protonated at low pH values and at higher concentrations (26, 29, 30). According to Cruywagen et al. (26) the protonation and condensation reactions that can occur upon acidification of molybdate are represented by the following equation:

p[MoO42-] + qH+ T [(MoO4)pHq](2p-q)-

(2)

where p and q are stoichiometric coefficients. In acidified solutions at pH ∼ 0.5 the Mo speciation in solutions is dominated by heptamolybdate (Mo7O246-) and octamolybdate (Mo8O264-) polyanions. Species such as HMo13O425-, Mo19O594- and others can also be present (25). The effective diffusion coefficient of Mo in the acidified water solution at pH ∼ 0.5 in our experiments is, therefore, represented by the average value for the various individual molybdate anionic species. The theoretical background for diffusion in liquids has been elaborated in detail and described elsewhere (31-33). Here we refer only to the necessary formulas needed in this work. Using our experimentally determined Mo concentration data, the diffusion coefficients, D, could be estimated from least-squares fitting of the solution to Fick’s second law of diffusion (33):

(

c/cf ) 1 - erf

[ ]) x x4Dt

(3)

where c and cf are the measured and final concentrations at a distance x from the initial boundary, respectively, t is time and ‘erf’ denotes the error function. Equation 3 is valid under conditions when, at the start of the experiments, all diffusing material is confined on one side of a boundary with pure solvent on the other side, i.e., c ) c0 for x < 0 at t ) 0 and c ) 0 for x > 0 at t ) 0. The experimental results for the development of Mo concentration and isotope gradients are shown in Table 1 and Figure 2. Diffusion coefficients for Mo ionic species derived from our experiments are given in Table 2 along with data reported by other workers (34, 35) for the case of infinite dilution. As seen from Table 2, the diffusion coefficient for MoO42in water measured in this study is 1.4 times that calculated from limiting ionic conductance data for infinite dilution and 2.4 times that of Mo polymolybdate in 0.33 M HNO3. The larger value of MoO42- diffusion coefficient observed in our work in comparison with the theoretical value may reflect both uncertainties in limiting ionic conductance data and VOL. 41, NO. 5, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1597

TABLE 1. Mo Concentration and Isotope Data for Diffusion Experiments in this Studya diffusion of MoO42- in water at pH 7

diffusion of Mo ionic species in 0.33 M HNO3 (pH ) 0.5)

distance from the source, cm

Mo, µM

δ97/95Mo, ‰

Mo, µM

δ97/95Mo, ‰

7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5

58 ( 11 105 ( 18 203 ( 27 346 ( 55 597 ( 83 1011 ( 83 1600 ( 139 2507 ( 134 3780 ( 245

-0.55 ( 0.06 -0.50 ( 0.05 -0.46 ( 0.03 -0.41 ( 0.05 -0.37 ( 0.09 -0.31 ( 0.06 -0.20 ( 0.06 -0.14 ( 0.06 -0.13 ( 0.05

7(7 20 ( 20 54 ( 48

0.05 ( 0.07 0.00 ( 0.02

335 ( 210 780 ( 390 1563 ( 640 2879 ( 900 4993 ( 1120 8320 ( 1200

0.04 ( 0.03 -0.04 ( 0.04 -0.03 ( 0.07 0.04 ( 0.03

a Concentration of Mo in the source solutions was 0.40 ( 0.03 M. The experimentally measured final concentration of Mo in the tubes after diffusion is completed (c×c4) was 23 500 ( 1400 µM. Uncertainties are two standard deviations for n ) 3 replicates. Temperature is 20 ( 1 °C.

FIGURE 2. Diffusion of Mo ionic species in 0.33 M HNO3 solution (2) and in H2O ([). In the experiments with 0.33 M HNO3 the uncertainty bars are two standard deviations for n ) 3 replicates at each point (the observed disproportions between upper and lower uncertainty bars are due to the logarithmical scale). In the experiments with H2O the uncertainties at the two standard deviation level (n ) 3) are smaller than the data points. the magnitude of interactions with adjacent particles of the solvent (34). The diffusion coefficient of polymolybdate ions in the acidified solution is smaller than that of MoO42-, consistent with the larger size of the diffusing species and their hydrodynamic radii. Using the Stokes-Einstein relation,

D ) kT/(6πηr)

(4)

where k is the Boltzmann constant and η the solution viscosity, the hydrodynamic radius, r, of the hydrated ions can be calculated from estimated diffusion coefficients (36). This results in hydrodynamic radii of ∼1.72‚10-10 m for MoO42- ions in water and ∼4.14‚10-10 m for the “average” Mo anions in 0.33 M HNO3. Figure 3 shows the development of the isotopic fractionation gradient between 97Mo and 95Mo during diffusion of the element in the studied solutions. Using eq 3, the evolution of the isotopic composition of Mo during diffusion can be written in terms of 97Mo/95Mo ratio as follows: 97

Mo/95Mo(x,t) )

(

1 - erf

[x

x

]) (

4D97Mot

/ 1 - erf

[x

x

])

4D95Mot

(5)

The ratio D97Mo/D95Mo ) 0.99988 ( 0.00004 has been estimated from the best fit of curves to the experimental data for MoO42for three replicates of the experiment. It is worth pointing out that the potential contribution of Mo adsorption onto the walls of the polypropylene tubes to the observed Mo 1598

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 5, 2007

isotopic fractionation is insignificant as suggested by the following line of evidence. In order to quantify the Mo adsorptive effect, control experiments have been performed in which pH neutral solutions with known Mo concentrations (100 µM) have been stored in polypropylene tubes for 3 days under constant agitation. No adsorptive loss of Mo in the tubes has been found within the limits of measurement precision (∼3% at one standard deviation level). Even assuming 3% adsorption is accompanied by Mo isotopic fraction as large as -1‰ per atomic mass unit, a simple mass balance calculation for two component mixing indicates that this hypothetical contribution would remain within the uncertainty of the isotopic measurements. Additionally, no adsorptive loss of Mo during storage of water in plastic bottles at slightly acidic to circumneutral pH for at least 2 days has been observed in previous studies (37, 38). Various models have been applied to explain the kinetic isotopic fractionation observed during diffusion. The simplest of these (39),

[ ]

D97Mo m95MoO4 ) D95Mo m97MoO4

0.5

) 0.99377

(6)

provides a theoretical limit for the case of self-diffusion in an ideal gas. For tracer diffusion through a host gas or liquid, the Chapman-Enskog theory (18, 24, 36) predicts that the ratio of isotopic diffusion coefficients will equate to the square root of the ratio of reduced masses:

[ ] m97MoO4+ mH2O

D97Mo m97MoO4× mH2O ) D95Mo m95MoO4+ mH2O

0.5

) 0.99937

(7)

m95MoO4× mH2O

Richter et al. (24) accounted for solvation effects by including variable amounts of H-bonded water molecules, which for our purposes means replacing the mxMoO4 terms with mxMoO4‚ yH2O in eq 7. Baker and Pope (34) also pointed out that the diffusion coefficients of individual Mo anions in water are dependent on the number of H-bonds per oxygen atom in the structures of these anions and that this number can vary considerably according to the structure of the anion. These authors calculated that the simple tetrahedral MoO42oxyanion can form three H-bonds per oxygen atom with adjacent water molecules, whereas large heteropolymeric species, such as Mo7O246- and Mo8O264-, have on average

TABLE 2. Mo Diffusion Coefficients (× 10-10 m2s-1) at 20 °C. medium species 2-

MoO4 Mo ionic species MoO42CoMo6O213CrMo6O213SiMo12O404-

H2O at pH 7 12.4 (

0.2b

0.33 M HNO3

infinite dilutiona

reference

8.66 5.7 ( 0.2 5.7 ( 0.2 4.8 ( 0.2

this work this work 35 34 34 34

5.2 ( 1.0b

a Infinite dilution values were recalculated to 20 °C using the Stokes-Einstein relation, eq 4. b Uncertainties are two standard deviations for n ) 3 replicates.

suggested that the transport of ions across membranes can result in much larger isotopic fractionation than takes place by diffusion in water itself (24). The results of our experiments may prove to be useful for developing a better understanding of the isotopic fractionation by marine organisms such as foraminifera. Finally, this study points to a need to characterize species-specific isotope effects concomitant with physicochemical processes, such as diffusion, for other elements.

Acknowledgments FIGURE 3. Isotopic fractionation of Mo as a function of diffusion distance in 0.33 M HNO3 solution (2) and in H2O ([). Uncertainty bars represent two standard deviations for n ) 3 replicates at each point.

less than two H-bonds per oxygen atom. Thus assuming 12 solvent molecules per molybdate ion yields D97Mo/D95Mo ) 0.99988, in perfect agreement with our determined value. In contrast to pH-neutral conditions, no Mo isotopic fractionation has been observed during diffusion of Mo in 0.33 M HNO3. As illustrated in Figure 3, variations in the isotopic composition of Mo during diffusion in 0.33 M HNO3 are within the standard deviation of the measurements. The larger molecular weights of Mo polyanions in 0.33 M HNO3 can presumably be one reason accounting for the absence of Mo isotope effect observed experimentally during Mo diffusion in the acidified solution. Another factor is that the polymeric species contain multiple metal centers, thus obliterating any distinct Mo isotopic signature. The diffusion experiments in this work have not been conducted at different temperatures, although, based on the general theory of isotope fractionation (40), temperature effects are likely to be significant. In oxygenated natural waters, the dominant chemical form of dissolved Mo is MoO42-. Hence, Mo isotopic fractionation during diffusion in the porewater of oxygenated surface sediment is to be expected and could account for minor variations in Mo isotopic signatures of sediment columns. Data obtained in our study show different degrees of Mo isotopic fractionation during diffusion for different chemical forms of the element. It is, therefore, important to consider Mo speciation when attempting to interpret the isotopic composition of Mo formed via diffusion, particularly for low pH aqueous solutions. Although Mo polyanions are more rarely encountered in nature in comparison with monomeric MoO42-, they can be found in the low pH waters of paludal areas as well as in the porewaters of Mo deposits and their mine tailings. In the context of aqueous (bio)-geochemistry, from the data presented above it might be speculated that the isotopic composition of Mo accumulated in organisms via diffusion through biological membranes may be variable depending on Mo speciation in solution. It has been

This research was supported by grants from EU’s structural fond for objective 1 Norra Norrland and Kempestiftelserna. We thank Bjo¨rn O ¨ hlander and Johan Ingri for support during this work. Christer Ponte´r and the staff of Analytica AB are gratefully acknowledged for technical assistance. We are grateful to anonymous reviewers for their constructive comments on the initial version of the manuscript.

Literature Cited (1) Williams, R. J. P.; Frau ´ sto da Silva, J. J. R. The involvement of molybdenum in life. Biochem. Biophys. Res. Commun. 2002, 292, 293-299. (2) Emerson, S. R.; Huested, S. S. Ocean anoxia and the concentrations of molybdenum and vanadium in seawater. Mar. Chem. 1991, 34, 177-196. (3) Anbar, A. D. Molybdenum stable isotopes: observations, interpretations and directions. Rev. Miner. Geochem. 2004, 55, 429-454. (4) Crusius, J.; Calvert, S.; Pedersen, T.; Sage, D. Rhenium and molybdenum enrichments in sediments as indicators of oxic, suboxic and sulphidic conditions of depositions. Earth Planet. Sci. Lett. 1996, 145, 65-78. (5) Dean, W. E.; Piper, D. Z.; Peterson, L. C. Molybdenum accumulation in Cariaco basin sediment over the past 24 k.y.: A record of water-column anoxia and climate. Geology 1999, 27, 507-510. (6) Zheng, Y.; Andersson, R. F.; van Geen, A.; Kuwabata, J. Authigenic molybdenum formation in marine sediments: a link to pore water sulphide in the Santa Barbara Basin. Geochim. Cosmochim. Acta 2000, 64, 4165-4178. (7) Morford, J. L.; Emerson, S. R.; Breckel, E. J.; Kim, S. H. Diagenesis of oxyanions (V, U, Re, and Mo) in pore waters and sediments from a continental margin. Geochim. Cosmochim. Acta 2005, 69, 5021-5032. (8) Dalai, T. K.; Nishimura, K.; Nozaki, Y. Geochemistry of molybdenum in the Chao Phraya River estuary, Thailand: Role of suboxic diagenesis and porewater transport. Chem. Geol. 2005, 218, 189-202. (9) Barling, J.; Arnold, G. L.; Anbar, A. D. Natural mass-dependent variations in the isotopic composition of molybdenum. Earth Planet. Sci. Lett. 2001, 193, 447-457. (10) Siebert, C.; Na¨gler, T. F.; Kramers, J. D. Determination of molybdenum isotope fractionation by double-spike multicollector ICPMS. Geochem. Geophys. Geosyst. 2001, 2, 2000GG000124. (11) McManus, J.; Na¨gler, T. F.; Siebert, C.; Wheat, C. G.; Hammond, D. E. Oceanic molybdenum isotope fractionation: diagenesis and hydrothermal rindge-flank alteration. Geochem. Geophys. Geosyst. 2002, 3, 2002GG000356. VOL. 41, NO. 5, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1599

(12) Siebert, C.; Na¨gler, T. F.; von Blanckenburg, F. Molybdenum isotope records as a potential new proxy for paleoceanography. Earth Planet. Sci. Lett. 2003, 211, 159-171. (13) Barling, J.; Anbar, A. D. Molybdenum isotope fractionation during adsorption by manganese oxides. Earth Planet. Sci. Lett. 2004, 217, 315-329. (14) Na¨gler, Th. F.; Siebert, C.; Lu ¨ schen, H.; Bo¨ttcher, M. F. Sedimentary Mo isotope record across the Holocene freshbrackish water transition of the Black Sea. Chem. Geol. 2005, 219, 283-295. (15) Siebert, C.; McManus, J.; Bice, A.; Poulson, R.; Berelson, W. M. Molybdenum isotope signatures in continental margin marine sediments. Earth Planet. Sci. Lett. 2006, 241, 723-733. (16) Nameroff, T. J.; Balistrieri, L. S.; Murray, J. W. Suboxic trace metal geochemistry in the eastern tropical North Pacific. Geochim. Cosmochim. Acta 2002, 66, 1139-1158. (17) Senftle, F. E.; Bracken, J. Y. Theoretical effect of diffusion on isotopic abundance ratio in rocks and associated fluids. Geochim. Cosmochim. Acta 1955, 7, 61-76. (18) O’Leary, M. H. Measurement of the isotope fractionation associated with diffusion of carbon dioxide in aqueous solution. J. Phys. Chem. 1984, 88, 823-825. (19) Fritz, S. J. Measuring the ratio of aqueous diffusion coefficients between 6Li+Cl- and 7Li+Cl- by osmometry. Geochim. Cosmochim. Acta 1992, 56, 3781-3789. (20) Appelo, C. A. J.; Postma, D. Geochemistry, Groundwater, Pollution, 2nd ed.; A. A. Balkema Publishers: Leiden, 2005; pp 86-101. (21) Desaulniers, D. E.; Kaufmann, R. S.; Cherry, J. A.; Bentley, H. W. 37Cl-35Cl variations in a diffusion-controlled groundwater system. Geochim. Cosmochim. Acta 1986, 50, 1757-1764. (22) Eggenkamp, H. G. M.; Middelburg, J. J.; Kreulen, R. Preferential diffusion of 35Cl to 37Cl in sediments of Kau Bay, Halmahera, Indonesia. Chem. Geol. 1994, 116, 317-325. (23) Rodushkin, I.; Stenberg, A.; Andre´n, H.; Malinovsky, D.; Baxter, D. C. Isotopic fractionation during diffusion of transition metal ions in solution. Anal. Chem. 2004, 76, 2148-2151. (24) Ritcher, R. A.; Mendybaev, J. N.; Christensen, I. D.; Hutcheon, R. W.; Williams, N. C.; Sturchio, Beloso, A. D. Kinetic isotopic fractionation during diffusion of ionic species in water. Geochim. Cosmochim. Acta 2006, 70, 277-289. (25) Baes, C. F.; Mesmer, R. E. Jr. The Hydrolysis of Cations, 2nd ed.; Wiley: New York, 1986; pp 253-257. (26) Cruywagen, J. J.; Draaijer, A. G.; Heyns, J. B. B.; Rohwer, E. A. Molybdenum(VI) equilibria in different ionic media. Formation constants and thermodynamic quantities. Inorg. Chim. Acta. 2002, 331, 322-329. (27) Malinovsky, D.; Rodushkin, I.; Baxter, D. C.; Ingri, J.; O ¨ hlander, B. Molybdenum isotope ratio measurements on geological

1600

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 5, 2007

(28)

(29)

(30)

(31) (32) (33) (34)

(35) (36)

(37)

(38)

(39) (40)

samples by MC-ICPMS. Int. J. Mass Spectrom. 2005, 245, 94107. Albare`de, F.; Telouk, P.; Blichert-Toft, J.; Boyet, M.; Agranier, A.; Nelson, B. Precise and accurate isotopic measurements using multiple-collector ICPMS. Geochim. Cosmochim. Acta 2004, 68, 2725-2744. Bourikas, K.; Hiemstra, T.; Van Riemsdijk, W. H. Adsorption of molybdate monomers and polymers on titania with a multisite approach. J. Phys. Chem. 2001, 105, 2393-2403. Ozeki, T.; Kihara, H. Study of equilibria in 0.03mM molybdate acidic aqueous solutions by factor analysis applied to ultraviolent spectra. Anal. Chem. 1988, 60, 2055-2059. Stearn, A. E.; Irish, E. M.; Eyring, H. A theory of diffusion in liquids. J. Phys. Chem. 1940, 44, 981-995. Ottar, B. Self-Diffusion and Fluidity in Liquids; Oslo University Press: Oslo, Norway, 1958. Noggle, J. H. Physical Chemistry, 2nd ed.; Harper-Collins: New York, 1996; pp 459-502. Baker, L. C. W.; Pope, M. T. The identical diffusion coefficients of isostructural heteropoly anions. The complete independence of D from ionic weight. J. Am. Chem. Soc. 1960, 82, 4176-4179. Li, Y.-H.; Gregory, S. Diffusion of ions in seawater and deep-sea sediments. Geochim. Cosmochim. Acta 1974, 38, 703-714. Reid, R. C.; Prausnitz, J. M.; Poling, B. E. The Properties of Gases and Liquids, 4th ed.; McGraw-Hill: New York, 1987; pp 557631. Reimann, C.; Siewers, U.; Skarphagen, H.; Banks, D. Does bottle type and acid-washing influence trace element analyses by ICPMS on water samples? A Test covering 62 elements and four bottle types: high density polyethene (HDPE), polypropene (PP), fluorinated ethane propene copolymer (FEP) and perfluoroalkoxy polymer (PFA). Sci. Total Environ. 1999, 239, 111130. Kawakubo, S.; Hashi, S.; Iwatsuki, M. Physicochemical speciation of molybdenum in rain water. Water Res. 2001, 35 (10), 24892495. Schauble, E. A. Applying stable isotope fractionation theory to new systems. Rev. Miner. Geochem. 2004, 55, 65-111. Criss, R. E. Principles of Stable Isotope Distribution; Oxford University Press: New York, 1999.

Received for review August 19, 2006. Revised manuscript received December 8, 2006. Accepted December 13, 2006. ES062000Q