Nuclear Field Shift Effect in Isotope Fractionation of Mercury during

4238

J. Phys. Chem. A 2010, 114, 4238–4245

Nuclear Field Shift Effect in Isotope Fractionation of Mercury during Abiotic Reduction in the Absence of Light Wang Zheng* and Holger Hintelmann EnVironmental and Life Sciences Program, Trent UniVersity, 1600 West Bank DriVe, Peterborough, Ontario K9J 7B8, Canada ReceiVed: October 29, 2009; ReVised Manuscript ReceiVed: February 13, 2010

We investigated the abiotic reduction of inorganic Hg(II) by dissolved organic matter (DOM) and stannous(II) chloride (SnCl2) in the absence of light and quantified fractionation of Hg isotopes during these processes. The kinetics of reduction by DOM was characterized using multiple parallel pseudo-first-order reactions, implying different reactive Hg(II) species resulting from Hg-DOM complexation. Significant mass independent isotopic anomalies were observed in reduction by both reducing reagents. Isotopes with odd atomic masses (199Hg and 201Hg) showed less enrichment in reactants Hg(II) than expected for a mass dependent fractionation process. The fractionation factors (R) showed an odd-even staggering pattern that resembles the variation of nuclear charge radii. We demonstrated that these isotopic anomalies originated from nuclear field shift effect (NFS). The contribution of NFS to the measured fractionation factors was estimated and found to be as significant as the mass dependent effect. The observed ∆199Hg/∆201Hg slope was explained by NFS and determined to be between 1.5 and 1.6 in abiotic nonphotochemical reduction, which is distinguishable from slopes determined for photochemical reduction. Therefore, we first demonstrated experimentally the significance of the nuclear field shift effect during reduction of Hg(II) and showed the application of isotope fractionation to distinguish between different reduction pathways. 1. Introduction Redox reactions are the pivotal processes in the biogeochemical transformations of mercury in aquatic systems. The reduction of Hg(II) to Hg(0) increases the emission of Hg(0) into the atmosphere, and competes with the bioavailable Hg(II) pool that may otherwise be methylated. Of special interest and importance is the pathway of reduction, which can potentially reveal the particular biological and geochemical environments of Hg, where reduction took place, and thus bridge biogeochemical factors and Hg transformations. In aquatic systems, both biotic and abiotic pathways have been recognized for the reduction of Hg(II). Biotic reduction can be mediated by various microorganisms.1-3 Abiotic reduction is usually initiated by dissolved organic matter (DOM) via either photochemical4-6 or nonphotochemical pathways.7,8 Although the role of DOM as a major electron donor in natural water is well recognized,9-11 the way DOM mediates Hg(II) reduction could differ fundamentally with or without sunlight. Photochemical reduction primarily occurs through the homolysis of Hg-DOM bonds or secondary processes by reactive reductants (e.g., HO2•).12,13 Both mechanisms involve highly reactive radical intermediates as a result of photolysis of DOM or Hg-DOM complexes. Nonphotochemical reduction of Hg(II) by DOM, however, is rather poorly understood. The redox properties of DOM have been particularly attributed to quinones, which function as electron shuttle intermediates during electron transfer.9,14 A redox transfer of two electrons per quinone unit has been postulated, although it may actually take place in successive one-electron steps leading to one-electron radical intermediates known as semiquinones.15,16 Therefore, it * Correspondence author. E-mail: [email protected].

was hypothesized that quinone or semiquinone moieties in organic matter are involved in electron transfer leading to Hg(II) reduction.7 Isotope fractionation has been used as an indicator of reaction pathways.17 Of particular interest are isotopic anomalies, which open a new perspective in fundamentally understanding the pathway of transformations and causes of isotope fractionation. Isotopic anomalies refer to deviations of isotope fractionation from conventional mass dependent fractionation (MDF). The conventional theory of MDF explains both equilibrium and kinetic isotope effects arising from the differences in vibrational energy of bonds involving different isotopes, which is a function of the bond strength and masses of atoms in motion.18,19 Anomalous isotope compositions of Hg have been recently observed in various natural samples.20-26 Understanding the mechanisms of mass independent isotope fractionation (MIF) is crucial to explain these isotopic anomalies and to apply them as reaction pathway indicators. Two of the MIF theories are particularly relevant to Hg. One is the magnetic isotope effect (MIE), which selectively separates isotopes with and without unpaired nuclear spin due to hyperfine coupling (HFC) between nuclear spin and electron spin.27-29 MIE is a kinetic isotope effect, which is mostly manifested through the radical pair mechanism,30 and is observed in photochemical reactions29 and biological systems (e.g., enzymatic reactions).31 The other MIF mechanism is the nuclear field shift effect (NFS). Bigeleisen32 amended the original equilibrium fractionation factor by adding correction terms as in the following equation:

ln R ) ln R0 + ln Kanh + ln KBOELE + ln Khf + ln Kfs (1) where ln R0 is the original mass dependent fractionation factor.

10.1021/jp910353y  2010 American Chemical Society Published on Web 03/01/2010

Isotope Effects of Hg in Abiotic Dark Reduction ln Kanh, ln KBOELE, ln Khf, and ln Kfs are correction terms to anharmonic vibration, Born-Oppenheimer approximation, nuclear spin effect (hyperfine splitting), and nuclear field shift effect, respectively. ln Kanh and ln KBOELE are found to be negligible for heavy elements. The nuclear field shift is a contribution to the isotope shift widely observed in atomic spectra of heavy elements.33-35 Qualitatively, it is a shift of the electron energy states in an atom or molecule due to the perturbation resulting from the interaction between nuclear charge and electrons with a high density at the nucleus. Depending on the number of neutrons, the finite nuclear size and charge distribution changes and causes a slightly different electrostatic field to act on the electrons in the immediate vicinity of the center of attraction.32 Therefore, the NFS is a function of nuclear volume and shape. MIF in equilibrium isotope exchange has been successfully explained by NFS.36-41 Kinetic isotope effects especially during redox reactions may also be subject to NFS as a result of nuclei-electron interactions, which affect electron transfer. However, NFS in kinetic processes is still lacking experimental evidence. In this study, we investigated the kinetics and concurrent isotope fractionation of abiotic nonphotochemical reduction of Hg(II) by natural DOM and stannous(II) chloride (SnCl2). We present experimental evidence of NFS during kinetic processes and estimate the contribution of various isotope effects to the overall fractionation factor. Isotope fractionation of Hg observed in different types of reduction is also compared to demonstrate its power to distinguish between different reduction pathways. 2. Methods 2.1. Reagents. The DOM used as a reductant in the first set of abiotic reduction experiments is Nordic Reservoir natural organic matter, a commercially available DOM reference material from the International Humic Substances Society (IHSS). DOM working solutions were prepared by diluting a concentrated solution (∼400 mg of C/L) with Milli-Q water to a final concentration of 10 mg of C/L and were filtered through a 0.2 µm cellulose nitrate filter before use. Stannous(II) chloride (SnCl2) (Sigma-Aldrich, 98%) was used as a reductant in the second set of abiotic reduction experiments. SnCl2 working solutions were diluted from a concentrated solution (1.6 M SnCl2 in 1.2 M HCl), which was bubbled by Hg-free Ar overnight to remove Hg traces, to a concentration of 1.6 nM in 1.2 M HCl (reagent grade, diluted in Milli-Q water) with a pH of ∼0 to prevent formation of hydrolyzed Sn(II) complexes. Potassium permanganate (KMnO4) (Sigma-Aldrich, g99.0%, low in mercury) was used to collect the product Hg(0) of abiotic nonphotochemical reduction. The KMnO4 working solution was prepared as 1 mM KMnO4 in 0.5 M H2SO4. Bromine monochloride (BrCl) used for Hg determination was prepared according to EPA method 1631 revision E. Hg(II) used in all abiotic reduction experiments was diluted from SRM NIST 3133 Hg standard solution (9.954 mg/g Hg(NO3)2 in 10% HNO3) to a final concentration of 100 µg/L (0.5 µM). 2.2. Abiotic Nonphotochemical Reduction by DOM. Abiotic nonphotochemical reduction experiments were conducted in a 1 L quartz bottle, which was wrapped with aluminum foil and kept in a dark container to eliminate light. The reacting container was placed in a water bath to maintain the temperature between 293 and 298 K. After Hg(II) was added into the DOM working solution, the pH was adjusted to ∼6 with NaOH. Then, this solution was kept in the dark for 22 h to allow Hg(II) to equilibrate with DOM. During the equilibration period, Hg(II)

J. Phys. Chem. A, Vol. 114, No. 12, 2010 4239 may slowly bind to DOM at various binding sites.42 Finally, the solution was continuously purged with Hg-free Ar (350-400 mL/min) to transfer Hg(0), the reduction product, into KMnO4 solution. KMnO4 was used to convert Hg(0) vapor to aqueous Hg(II), which allows easier measurement of isotope ratios of the product.43 The Hg-DOM solution was incubated in the dark for up to 923 h. Subsamples (10-20 mL) were taken every few hours from the reacting solution and were immediately quenched by BrCl (1-2%). The KMnO4 solution was also renewed, when each subsample was taken. Hence, each KMnO4 solution only contained Hg(0) generated during the last sampling interval. A preliminary experiment with Hg(II) added to Milli-Q water (no DOM added, but acidified with HCl to prevent adsorption to the container) was also conducted in the dark. Reduction of Hg in this solution is negligible. 2.3. Abiotic Nonphotochemical Reduction by SnCl2. This reduction experiment was conducted under the same conditions as the experiment with DOM. The initial working solution of Hg(II) was prepared in 0.12 M HCl. The acid is necessary to prevent adsorption of Hg to the container and hydrolysis of Sn2+ once it is added.44 Small aliquots of SnCl2 (30-80 µL) were added into the Hg(II) solution. After each addition of SnCl2, the working solution was mixed and purged with Hg-free Ar (350-400 mL/min) for 30 min to transfer reduced Hg(0) into 1 L KMnO4 solution to collect the accumulated product of reduction. Subsamples of both the reactant and product (5-20 mL depending on Hg(II) concentration) were taken after each purging period and immediately quenched with 0.5% BrCl. 2.4. DOM and Hg Analysis. DOM was measured using a TOC-VCPH analyzer (Shimadzu Scientific Instruments, Inc.). Hg concentrations were determined using a Tekran 2600 CV-AFS (Tekran Instruments, Inc.). The background Hg concentration in DOM working solutions was determined to be 4.3 ng/L (∼20 pM). Hg concentrations in BrCl (1%) and KMnO4 solutions were lower than 1 ng/L (5 pM). 2.5. Isotope Ratio Analysis. Hg isotope ratios were measured by multicollector inductively coupled plasma mass spectrometry (MC-ICP/MS, Neptune, Thermo Scientific). The sample introduction system consists of a cold vapor generation system45 equipped with an additional Nafion drying tubing (PERMA PURE, USA) and Apex-Q nebulizer (Elemental Scientific Inc.). Hg vapor was generated via online reduction by 0.16 M (3% w/v) SnCl2 solution. Aerosol of Tl standard (20 ng/mL Tl in 0.12 M HCl, Aldrich) generated by the Apex was simultaneously introduced into the plasma with Hg vapor to correct for instrumental mass bias. Sample measurements were bracketed with 2 µg/L (10 nM) Hg standard solution (in 0.12 M HCl, NIST 3133). Hg concentrations in samples and bracketing standards were matched to within 5%. We measured six stable isotopes (198Hg, 199Hg, 200Hg, 201Hg, 202Hg, and 204Hg), two isotopes (203Tl, 205Tl), and one Pb isotope (208Pb). The Pb isotope was monitored to correct for isobaric interference from 204Pb. The Faraday cup configuration of the Neptune MC-ICP/MS does not allow simultaneous collection of all isotopes mentioned above. Therefore, we used two sets of cup configuration. The first set collects masses 198, 199, 201, 202, 203, 204, 205, and 208. The second set collects masses 198, 199, 200, 201, 202, 203, and 205. All samples from reduction experiments by DOM and the first trial of the SnCl2 reduction experiment were measured using the first cup configuration. All samples of the second trial of SnCl2 reduction experiment and selected samples from the other two experiments were measured using the second cup configuration.

4240

J. Phys. Chem. A, Vol. 114, No. 12, 2010

Zheng and Hintelmann

Figure 1. Kinetics of abiotic reduction of Hg(II) by DOM in the dark. The solid line is the fitted reduction curve using eq 5. Dashed lines are 95% confidence bands of this fit.

Figure 2. Rayleigh plots of δxHg (x ) 199 and 202) as a function of fR (fraction of residual reactant Hg(II)). Two replicates of SnCl2 reduction experiments are combined. Short dashed and long dashed lines are modeled Rayleigh curves based on eq 4 for δ202Hg and δ199Hg, respectively.

Hg isotope compositions are reported in δ values:

δ Hg ) x

(

x/198 Rsample

Rx/198 3133

)

- 1 · 1000‰

(2)

where x is each Hg isotope between 199 and 204. Isotopic anomaly is characterized using the “capital delta” notation (∆) and is defined as

∆xHg ) δxHg -

ln(m198 /mx) × δ202Hg ln(m198 /m202)

(3)

where mx is the atomic mass of isotope x and the mass number x ) 199, 200, 201, and 204. The kinetic fractionation factors between the product Hg(0) and reactant Hg(II) is defined as Rx/198 ) Rproduct/Rreactant, where R is the isotope ratio. The following form of the Rayleigh equation was used to evaluate kinetic fractionation factors:

ln

1000 + δ ) (Rx/198 - 1)ln fR 1000 + δ0

(4)

where δ is δxHg (defined by eq 1) of reactant Hg(II). δ0 is the initial δ value of the reactant at fR ) 1 and fR is the fraction of residual reactant Hg(II). The external reproducibility of the method is evaluated using a secondary reference materials, the UM-Almade´n Hg standard (in 1% BrCl), which was analyzed repeatedly in each analytical session. δ202Hg, ∆201Hg, and ∆199Hg of Almade´n standard measured during all analytical sessions of abiotic reduction experiments are -0.57 ( 0.07‰ (2 SD, n ) 7), -0.002 ( 0.04‰ (2 SD, n ) 7), and -0.02 ( 0.05‰ (2 SD, n ) 7), respectively. They are consistent with values reported by other laboratories.46 No isotope anomalies were determined for Almade´n standard, excluding the possibility of anomalies induced by instrumental bias. 3. Results and Discussion 3.1. Kinetics of Abiotic Reduction of Hg(II) by DOM. Reduction of Hg(II) to Hg(0) by DOM in the dark was overall a slow process, as shown in Figure 1. Approximately 50% of initial Hg(II) was reduced during 923 h. However, in the first 50-100 h, the reduction was visibly more rapid than afterward. The overall reduction can be modeled using a combination of multiple parallel pseudo-first-order reactions with two or more pools of different reactive Hg(II):

fR ) A · exp(-k1t) + (1 - A) · exp(-k2t)

(5)

where A is the fraction of one type of reactive Hg(II) and 0 e A e 1. k1 and k2 are pseudo-first-order rate constants. fR in Figure 1 is fitted using eq 5. Rate constants and A are calculated as k1 ) 0.0290 ( 0.0068 h-1 (2SE, P ) 0.0001), k2 ) 0.0005 ( 0.0001 h-1 (2SE, P ) 0.0002), and A ) 0.40 ( 0.04 (2SE, P < 0.0001). The two rate constants are significantly different from each other, supporting the idea that different types of reactive Hg(II) were present. Similar kinetics has been observed in previous studies on abiotic reduction by humic substances.9,47,48 The multiple pseudo-first-order kinetics also resemble those observed in photochemical reduction of Hg(II).6,49 The kinetics of abiotic nonphotochemical reduction of Hg(II) is likely a result of the heterogeneous complexation of Hg(II) by DOM. The reactivity of Hg is strongly affected by its preference to reduced sulfur binding sites of DOM.50-52 The molecular size, binding environment, and coordination mode within DOM also limits the reactivity of Hg both directly and through decreasing the production of electron donors such as semiquinone radicals.53 The effect of Hg-DOM interaction on the reactivity of Hg has been characterized in photochemical reduction of Hg(II).49 It has also been discovered that, in aquatic systems, newly deposited Hg tends to be more reactive than native Hg that was already in the pool, probably due to the fact that native Hg has formed highly stable complexes with DOM, while new Hg needs time to equilibrate and join the “old” Hg pool.54-56 Therefore, we believe abiotic reduction of Hg(II) by DOM is primarily controlled by how Hg(II) binds to DOM, which practically separates Hg(II) into fractions of different reactivity. As a result, the rate of electron transfer from electron donors, such as quinones and semiquinones, is limited by various factors that contributed to the reactivity of Hg(II), and hence the reduction rates. Another factor that may contribute to the abiotic reduction kinetics is the variation in the electron-transfer ability of different electron donors, which is pH dependent.57,58 In our experiments, pH is not a variable, but different quinones, semiquinones, or other electron donating species possibly coexist and may lead to different reduction rates of Hg(II). 3.2. Isotope Fractionation during Abiotic Reduction of Hg(II). 3.2.1. OWerWiew. Figure 2 shows the evolution of δ values as reduction was progressing. Reductions by different

Isotope Effects of Hg in Abiotic Dark Reduction

J. Phys. Chem. A, Vol. 114, No. 12, 2010 4241

Figure 3. Three-isotope plots for all abiotic reduction experiments. Two replicates of SnCl2 reduction experiments are combined. Dashed lines are theoretical MDF lines based on kinetic mass dependent fractionation law.67

TABLE 1: Summary of Fractionation Factors (103 ln r) and Uncertainties (2SE) Obtained from Rayleigh Plots DOM isotope pair 199/198 200/198 201/198 202/198 204/198

SnCl2 trial 1

SnCl2 trial 2

103 ln R

2SE

103 ln R

2SE

103 ln R

2SE

-0.19 -0.82 -1.02 -1.52 -2.30

0.02 0.11 0.05 0.06 0.11

-0.19 -0.88 -1.16 -1.77 -2.66

0.02 0.06 0.08 0.11 0.17

-0.22 -0.75 -1.06 -1.56 /

0.02 0.06 0.08 0.11 /

reagents yielded very similar patterns of isotope fractionation, suggesting both abiotic reduction processes undertook a similar electron transfer pathway. The evolution of δ values of both reactant and product is in good agreement with the Rayleigh fractionation model (eq 4). This observation is in accordance with previous studies on kinetic isotope fractionation, which progressively enriches heavier isotopes in reactants.59-61 However, what separates this study from aforementioned kinetic isotope fractionation processes is the discovery of significant isotopic anomalies for δ199Hg and δ201Hg (Figure 3). δ199Hg and δ201Hg are systematically lower than what would be expected for a mass dependent fractionation (∆199Hg and ∆201Hg are as high as -0.6 and -0.4‰, respectively), while all even isotope pairs δ200Hg and δ204Hg match theoretical MDF lines well (∆200Hg ) -0.03 ( 0.12 (2SD) and ∆204Hg ) -0.01 ( 0.09 (2SD)). Apparently, these anomalies cannot be accounted for by the difference between equilibrium and kinetic mass dependence laws, which can only generate a deviation that is 1-2 orders of magnitude lower than the observed ∆ values. Again, both reducing reagents (DOM and SnCl2) resulted in almost identical isotopic anomalies. Up to now, the only kinetic transformations of Hg that have been found to involve MIF are photochemical reduction20,49 and evaporation.62 We will then identify the mechanism of MIF during abiotic reduction and discuss how it is different from previously observed MIF for Hg. 3.2.2. Odd-EWen Staggering of Fractionation Factors. Fractionation factors (103 ln R) calculated using the Rayleigh equation are summarized in Table 1. As a result of MIF, plots of fractionation factors as functions of mass numbers (Figure 4) showed obvious deviation from linearity. This variation of fractionation factors intimately resembles the odd-even staggering pattern exhibited by the change of mean-square nuclear charge radii of Hg isotopes (δ〈r2〉198,i )〈r2〉i - 〈r2〉198, i )

Figure 4. Fractionation factors (103 ln R) and the change of mean square nuclear charge radii (δ〈r2〉198,i) plotted against mass numbers. δ〈r2〉198,i is defined as 〈r2〉i - 〈r2〉198. 〈r2〉 is calculated from root-meansquare (rms) charge radii in fermis (10-15 m). δ〈r2〉198,i values presented in this graph are compiled from three sources: Ulm et al.,35 Nadjakov et al.,69 and Angeli.70 Error bars are 2SE of 103 ln R. Solid lines are linear regressions across 103 ln R and δ〈r2〉 of even mass numbers.

199-204). The odd-even staggering of nuclear charge radii is a well-documented effect.63-65 Essentially, the mean square charge radius 〈r2〉 of an odd-N nucleus is always (with a few exceptions) smaller than the average of the radii of its even-N neighbors. This has caused the nonlinearity in nuclear field shift (δVFS) for different isotopes because δVFS is approximately proportional to δ〈r2〉.33,35,66 Consequently, isotope fractionation that arises from NFS exhibited a similar odd-even staggering pattern.32,36,38 Therefore, we postulate the isotopic anomalies observed in abiotic reductions originated from NFS. Bigeleisen’s32 corrected fractionation factor (eq 1) was simplified according to

ln Ri ) (hc/kT) fsi A + (1/24)(h/kT)2

(

)

1 1 B + ln Rhf mr mi (6)

where (hc/kT)fsiA ) ln Rfs represents the contribution of NFS, while (1/24)(h/kT)2(1/mr - 1/mi)B ) ln Rmd represents the contribution of MDF. ln Rhf is the contribution of hyperfine splitting caused by nuclear spin. mr and mi indicate the masses of the isotope pair with mr ) 197.9668, as 198Hg is used as the reference isotope in all isotope pairs. The expression fsi is the field shift frequency for the ith isotope. T is the temperature, k and h are the Boltzmann and Planck constants, respectively, and c is the velocity of light. A and B are adjustable constants, functioning as scaling factors for the NFS and MDF, respectively. The MDF term in eq 6 was developed for the equilibrium isotope effect. For the kinetic effect, we modified it according to Young et al.67 Since the field shift fsi is proportional to the change of mean square nuclear charge radii δ〈r2〉, at constant temperature, eq 6 can be rewritten as

( )

ln Ri ) (〈r2〉i - 〈r2〉198)a + ln

m198 b + ln Rhf mi

(7)

where a and b are new scaling factors. We can calculate a, b, and ln Rhf from the intersection point of lines plotted using eq 7 and the measured ln R of all isotope pairs. Technically, there should be only one intersection point for all these lines. However, the lines of ln R200, ln R202, and ln R204 nearly overlap because both δ〈r2〉 and ln(m198/mi) are approximately proportional to mass numbers for even isotopes. Therefore, we used ln R199, ln R201, and ln R202 to solve eq 7.

4242

J. Phys. Chem. A, Vol. 114, No. 12, 2010

Zheng and Hintelmann

TABLE 2: Scaling Factors a and ba DOM 35

Ulm et al., 1986 Nadjakov et al., 199469 Angeli, 200470 Hahn et al., 197968 b

SnCl2 trial 1

SnCl2 trial 2

a

b

a

b

a

b

-0.0048(8) -0.0048(8) -0.0040(7) -0.0007(1)

0.0291(68) 0.0293(67) 0.0397(49) 0.0435(42)

-0.0063(10) -0.0063(10) -0.0053(8) -0.0009(1)

0.0265(70) 0.0268(70) 0.0407(48) 0.0456(40)

-0.0048(9) -0.0048(9) -0.0040(7) -0.0007(1)

0.0305(60) 0.0308(60) 0.0413(40) 0.0451(33)

a The combined standard uncertainties propagated from 2SE of ln R are given in brackets. b Only the ratio δ〈r2〉i/δ〈r2〉202 is available from this reference. Therefore, the scaling factor a is calculated by substituting δ〈r2〉i with δ〈r2〉i/δ〈r2〉202 in eq 7 for this reference.

TABLE 3: Contributions of Nuclear Field Shift (NFS), Mass Dependent (MDF) and Nuclear Spin Effects (HFC: Hyperfine Coupling) to the Fractionation Factors (103 ln r) of Hg during Abiotic Nonphotochemical Reductionsa Ulm et al., 198635 isotope pair

Nadjakov et al., 199469

NFS

MDF

HFC

SUM

NFS

MDF

HFC

199/198 200/198 201/198 202/198 204/198

-0.06 -0.45 -0.59 -0.94 -1.43

-0.15 -0.29 -0.44 -0.58 -0.87

0.01 0.00 0.01 0.00 0.00

-0.19 -0.74 -1.02 -1.52 -2.29

-0.06 -0.45 -0.59 -0.94 -1.43

-0.15 -0.29 -0.44 -0.59 -0.88

0.01 0.00 0.01 0.00 0.00

199/198 200/198 201/198 202/198 204/198

-0.08 -0.59 -0.79 -1.24 -1.89

-0.13 -0.27 -0.40 -0.53 -0.79

0.02 0.00 0.02 0.00 0.00

-0.19 -0.86 -1.16 -1.77 -2.68

-0.08 -0.59 -0.78 -1.24 -1.88

-0.14 -0.27 -0.40 -0.54 -0.80

199/198 200/198 201/198 202/198 204/198

-0.06 -0.45 -0.60 -0.95 -1.43

-0.15 -0.31 -0.46 -0.61 -0.91

0.00 0.00 0.00 0.00 0.00

-0.22 -0.76 -1.06 -1.56 -2.35

-0.06 -0.45 -0.60 -0.94 -1.43

-0.16 -0.31 -0.46 -0.62 -0.92

a

Angeli, 200470

SUM

Hahn et al., 197968

NFS

MDF

HFC

SUM

NFS

MDF

HFC

SUM

-0.08 -0.36 -0.51 -0.73 -1.20

-0.20 -0.40 -0.60 -0.80 -1.19

0.09 0.00 0.09 0.00 0.00

-0.19 -0.76 -1.02 -1.52 -2.39

-0.05 -0.31 -0.45 -0.65 -0.98

-0.22 -0.44 -0.65 -0.87 -1.30

0.08 0.00 0.08 0.00 0.00

-0.19 -0.74 -1.02 -1.52 -2.28

SnCl2 Trial 1 0.02 -0.19 0.00 -0.86 0.02 -1.16 0.00 -1.77 0.00 -2.69

-0.10 -0.48 -0.67 -0.96 -1.59

-0.20 -0.41 -0.61 -0.81 -1.21

0.12 0.00 0.12 0.00 0.00

-0.19 -0.89 -1.16 -1.77 -2.80

-0.07 -0.41 -0.59 -0.86 -1.29

-0.23 -0.46 -0.69 -0.91 -1.36

0.11 0.00 0.11 0.00 0.00

-0.19 -0.86 -1.16 -1.77 -2.65

SnCl2 Trial 2 0.00 -0.22 0.00 -0.76 0.00 -1.06 0.00 -1.56 0.00 -2.35

-0.08 -0.36 -0.51 -0.73 -1.21

-0.21 -0.42 -0.62 -0.83 -1.23

0.07 0.00 0.07 0.00 0.00

-0.22 -0.78 -1.06 -1.56 -2.44

-0.05 -0.31 -0.45 -0.65 -0.98

-0.23 -0.45 -0.68 -0.90 -1.35

0.06 0.00 0.06 0.00 0.00

-0.22 -0.76 -1.06 -1.56 -2.33

DOM -0.19 -0.74 -1.02 -1.52 -2.30

SUM is the sum of the three contributions.

The nuclear spin effect ln Rhf is only a concern for odd isotopes since even isotopes of Hg have zero nuclear spin. So, we estimated the contributions of mass dependent, nuclear field shift, and nuclear spin effects to the overall fractionation factors using δ〈r2〉 values from four sources: Hahn et al.,68 Ulm et al.,35 Nadjakov et al.,69 and Angeli.70 Results are presented in Table 2 (scaling factors) and Table 3. Uncertainties of scaling factors in Table 2 are combined standard uncertainties (according to EURACHEM/CITAC guide71) estimated from the 2SE of measured ln R. In Table 3, NFS (103 ln Rfs) and MDF (103 ln Rmd) have the same negative sign, implying both effects tend to enrich isotopes with heavier or larger nuclei in the reactant Hg(II) phase. MDF originates from the difference in vibrational zero point energy (ZPE). Lighter isotopes of Hg(II) are preferentially reduced because of their higher ZPE. NFS originates from the effect of nuclear size and shape on nuclei-electron interactions. The ground state electronic energy of a light isotope lies lower than that of a heavier isotope due to the smaller size and larger surface charge density of the light isotope compared with those of a heavier isotope.32 Consequently, an electron with nonzero probability density at the nucleus will be bound more strongly to a smaller nucleus. Usually, only s-orbital electrons have a finite magnitude of electron density while p, d, and f orbitals do not.72-74 So, during reduction of Hg(II) to Hg(0), which involves transfer of electrons in the 6s orbital, lighter/smaller isotopes tend to gain electrons and be reduced, while heavier/ larger isotopes tend to remain as Hg(II) with unfilled 6s orbitals. Our calculation in Table 3 suggests the relative contributions of NFS and MDF to the total isotope fractionation are similar

at room temperature. This observation is comparable to literature data based on theoretical calculations.74 Schauble74 calculated 103 ln Rfs as -1.27 and 103 ln Rmd as -0.92 for the isotope pair 202Hg/198Hg during isotope exchange between HgCl2 and Hg0 vapor at 298 K and predicted NFS is more significant than MDF at high temperature, but their difference tends to diminish when the temperature decreases. The temperature dependence is well documented for both MDF and NFS.32,75 Our studies were carried out at room temperature at around 293-298 K. Temperature is treated as a constant in our estimation and is already incorporated into scaling factors a and b. ln R199 and ln R201 contain an additional component, which is absent in even isotope pairs. The additional term, nuclear spin effect, originates from the coupling of the nuclear and electronic spin angular momenta, which results in a hyperfine splitting in atomic spectra. The effect is not to be confused with magnetic isotope effect during radical-pair related reactions, which is a special manifestation of the effect of hyperfine interaction on isotope fractionation. As already discussed, the abiotic nonphotochemical reduction of Hg is not likely to proceed via the generation of radical pairs. In general, the nuclear spin effect is not as well characterized as mass dependent or nuclear field shift effects. Up to now, it has been reported mostly during chemical exchange reactions.32,76-78 Fujii et al.77 pointed out the center of gravity of the hyperfine structure of the odd isotope does not exist in the median between the spectra of adjacent even isotopes, which contributes to the odd-even staggering of isotope shifts. In their later study,40,76 they discussed the possibility that the nuclear spin effect is caused by a shift of minimum potential energy from the center of

Isotope Effects of Hg in Abiotic Dark Reduction

J. Phys. Chem. A, Vol. 114, No. 12, 2010 4243

gravity of hyperfine splitting energies, which also splits the vibrational energy level into hyperfine structure. Knyazev et al.78,79 evaluated the effect of the nuclear quadrupole effect, originating from the interaction between the nuclear quadrupole moment and the electrostatic field created by the valence electrons, which could be considered as a special type of nuclear spin effect for quadrupole nuclei. The contribution of the nuclear spin effect (ln Rhf) was generally found to be small or negligible, e.g., contributing approximately 10% or less to the overall fractionation in uranium chemical exchange reactions.32,78 In our estimation, ln Rhf was calculated as ln R - (ln Rfs + ln Rmd) and therefore depends on the accuracy of δ〈r2〉 and the ln R. Given the current uncertainty level associated with scaling factors (Table 2), the HFC component cannot be completely resolved. Direct calculation of ln Rhf based on hyperfine splitting data will be necessary to estimate the weight of ln Rhf with more accuracy. In general, the sums of the three effects are in quantitative agreement with the measured fractionation factors in all reduction processes (Tables 1 and 3), except for the sums of ln R204 based on Angeli,70 which showed a slight anomaly, whereas no isotopic anomalies are experimentally determined for 204Hg/ 198Hg, in either this or previous studies.20,49,62 This was also discussed by Ghosh et al.22 and Estrade et al.,62 who argued this disagreement is caused by the uncertainties of δ〈r2〉. Although the precision of our estimation is affected by the accuracy of δ〈r2〉 and measured fractionation factors, it clearly demonstrates that NFS contributed significantly to Hg isotope fractionation and is the main cause of MIF during abiotic nonphotochemical reduction of Hg(II). 3.2.3. ∆199Hg/∆201Hg Slope Resulting from the Nuclear Field Shift Effect. Isotopic anomalies were conventionally reported as deviations from the mass dependent fractionation for traditional isotopes80-82 and the resulting ∆ notation was adapted to the Hg isotope system.46 While most traditional isotope systems (e.g., O, C, and S) have only one stable odd isotope, there are two odd/magnetic isotopes of Hg (199Hg and 201Hg). One interesting discovery is that the isotopic anomalies of the two odd isotopes (∆199Hg and ∆201Hg) follow a linear relationship and the ∆199Hg/∆201Hg slope seems to vary according to the mechanism of isotope fractionation. For example, photochemical reduction yielded ∆199Hg/∆201Hg ranging from approximately 1.0-1.3 at different Hg/DOM concentration ratios.20,49 The linear relationship between ∆199Hg and ∆201Hg is also observed in abiotic nonphotochemical reductions (Figure 5). Here, we demonstrate how this slope can be quantitatively explained by considering both the nuclear field shift and nuclear spin effect. For an isotope fractionation between two phases A and B, the fractionation factor can by approximated as83,84

103 ln RA-B ≈ δA - δB

(8)

Since R is usually very close to unity, eq 8 is a good approximation for differences in δ values less than about 10‰.84 In abiotic reduction, the δ difference between product and reactant for δ199Hg and δ201Hg are typically less than 0.5 and 2‰, respectively. Considering the external uncertainties of δ values, errors caused by this approximation are negligible. Combining eq 3 and eq 8, we have i ∆Ai - ∆Bi ) 103 ln RA-B -

ln(m198 /mi) 202 × 103 ln RA-B ln(m198 /m202) (9)

where i ) 199 or 201. A and B represent product and reactant

Figure 5. ∆199Hg/∆201Hg during abiotic nonphotochemical reduction. Solid and dashed lines are the calculated ∆199Hg/∆201Hg slopes (Table 4). The overall ∆199Hg/∆201Hg is 1.61 ( 0.06 (2SE) by treating all data points as a single set, which approximately equals the average of ∆199Hg/∆201Hg slopes obtained from each experiment (Table 4).

TABLE 4: Slope ∆199Hg/∆201Hg of Linear Regression of ∆199Hg versus ∆201Hg reductant

measureda

calculatedb

DOM SnCl2 trial 1 SnCl2 trial 2

1.60 ( 0.12 1.59 ( 0.11 1.62 ( 0.08

1.55 ( 0.07 1.52 ( 0.06 1.62 ( 0.08

a Uncertainties are 2SE of the linear regression. For all slopes, r2 > 0.95, P < 0.0001. b Calculated according to eq 11. δ〈r2〉 from four data sources generated the same ∆199Hg/∆201Hg. The highest standard uncertainties generated by all data sources are shown.

phases, respectively. Incorporating eq 7 into eq 9 yields

10-3(∆Ai - ∆Bi ) )

(

a · δ〈r2〉199 -

)

ln(m198 /mi) · δ〈r2〉202 + ln Rihf ln(m198 /m202)

(10)

199 201 201 Since ∆199Hg/∆201Hg is a constant, (∆199 A - ∆B )/(∆A - ∆B ) ) ∆199Hg/∆201Hg. Therefore, ∆199Hg/∆201Hg can be expressed as

a · (δ〈r2〉199 - 0.252 · δ〈r2〉202) + ln R199 ∆199Hg hf ) (11) 201 201 2 2 ∆ Hg a · (δ〈r 〉201 - 0.752 · δ〈r 〉202) + ln Rhf The measured slopes ∆199Hg/∆201Hg and those calculated using eq 11 were tabulated in Table 4. Calculated ∆199Hg/ ∆201Hg values are statistically not different from measured values (P ) 0.31, t-test). In abiotic nonphotochemical reduction, this slope ranges approximately from 1.5 to 1.6. The variation of this slope between different sets of experiments is small, and therefore, different sets of data points were considered as a single set with an overall ∆199Hg/∆201Hg of 1.61 ( 0.06 (2SE) (Figure 5), which approximately equals the average of each single slope listed in Table 4. 3.2.4. Potential Significance of MIF in Discerning Reduction Pathways. The pattern and magnitude of MIF vary according to its mechanisms. The versatility of MIF makes it a particularly useful tool in discerning reaction pathways. With the identification of NFS, we are now able to distinguish Hg(0) originating from photochemical and nonphotochemical reductions of Hg(II) based on their different isotopic anomalies. The first distinct difference between the two types of reduction is the direction of MIF. If abiotic nonphotochemical reduction dominates the reduction of Hg(II) in aquatic systems, ∆199Hg

4244

J. Phys. Chem. A, Vol. 114, No. 12, 2010

and ∆201Hg should be positive for the product Hg(0) and negative for the reactant Hg(II) as a result of NFS, as demonstrated by Figure 3. In contrast, photochemical reduction of Hg(II) by natural DOM and sunlight yielded the opposite isotopic anomalies (positive ∆199Hg and ∆201Hg for Hg(II)) as a result of MIE.20,49 It is worth pointing out that the isotopic anomalies during photochemical reduction could be complicated by the involvement of other types of MIF, especially the reversed MIE, which depletes magnetic isotopes (e.g., 199Hg and 201Hg) from the reactant and consequently produces ∆ of the same sign as in nonphotochemical reduction.85,86 However, there is a fundamental difference between NFS and the reversed MIE. As discussed earlier, NFS does not change the sign of δ199Hg because both NFS and MDF tend to enrich heavier isotopes in the reactant. In contrast, the reversed MIE depletes 199Hg relative to 198Hg and thus would result in negative δ199Hg in the reactant if it is the dominant isotope effect.85 Therefore, using the sign of ∆ and δ values together, we are able to distinguish nonphotochemical reduction (dominated by NFS) and photochemical reduction of Hg(II) (dominated by MIE or the reversed MIE). The ∆199Hg/∆201Hg slope complements the direction of MIF in distinguishing reduction pathways and is already used to identify isotopic anomalies in natural samples originating from photochemical reduction.26 According to eq 11, ∆199Hg/∆201Hg of nonphotochemical reduction is a function of nuclear mass, volume, and to a lesser extent, nuclear spin. Considering the high reproducibility of this slope in three separate experiments, we believe the value of 1.5-1.6 is representative for this reduction pathway. Equation 11 may also apply to other isotope fractionation processes that are subject to NFS. Although the scaling factor a may vary in different reactions, it cancels out when the nuclear spin effect is negligible. Therefore, more experimental data on other transformations that are subject to NFS and a quantitative evaluation of the nuclear spin effect would help to verify the applicability of eq 11. In contrast, ∆199Hg/∆201Hg of Hg(II) photochemical reduction is substantially lower than that of nonphotochemical reduction. Bergquist and Blum20 determined a ∆199Hg/∆201Hg of ∼1.0. Zheng and Hintelmann49 later found this value to vary according to the Hg/DOC ratio, but only within a small range (1.19-1.31). As the MIF during photochemical reduction is believed to be dominated by MIE, which manifests through geminate radical pairs that are produced as intermediates of photochemical reduction,13 the origin and variation of ∆199Hg/∆201Hg could be related to the different magnetic moments of 199Hg and 201Hg, which differentiate the rates of HFC and hence the recombination rates of intermediate radical pairs centered by these two magnetic isotopes. In addition, spin-orbital coupling (SOC), which leads to spin-independent recombination of radical pairs, may compete with HFC and reduce the degree of MIE.87 The extent of SOC is affected by the constituents of radical pairs (internal and external heavy atom effects). Hence, the involvement of different degree of SOC may also contribute to the variation of ∆199Hg/∆201Hg during photochemical reduction. Therefore, both the values and fundamental causes of the ∆199Hg/∆201Hg slope vary according to reduction pathways. However, cautions should be taken when this slope is used since the current ∆199Hg/∆201Hg of MIE is still an empirical value. Also, photochemical reduction of Hg(II) is usually accompanied by concurrent nonphotochemical reduction processes, and thus the net ∆199Hg/∆201Hg observed in photochemical reduction may actually reflect a combination of MIE and, to a lesser extent, NFS. Furthermore, the accuracy of this slope is limited by the

Zheng and Hintelmann number of data and the precision of ∆ values. Thus, we suggest this slope should be used together with the sign of ∆ and δ values to discern reduction pathways. 4. Conclusions In summary, we have demonstrated the importance of NFS in abiotic nonphotochemcial reduction of Hg(II) and estimated its contribution to the overall isotope fractionation. This is the first systematic characterization of NFS in a kinetic process that is analogous to Hg transformation in natural waters. Abiotic nonphotochemcial reduction is ubiquitous, albeit usually not dominant with the presence of light, in natural waters. It provides additional pathways of Hg(0) production besides microbial and photochemical reduction. Our results suggest the isotope fractionation during different reduction pathways could be dominated by different isotope effects. With the understanding of the mechanisms and patterns of different types of MIF, these reduction pathways and their corresponding redox environments can be discerned using isotopic anomalies of either the product Hg(0) or reactant Hg(II). Therefore, this study is a further evidence for the application of isotope fractionation in tracking biogeochemical transformation of Hg. Acknowledgment. This research was supported by a Natural Sciences and Engineering research Council grant to H.H. We thank Brian Dimock for providing DOM samples, two anonymous reviewers for offering their expertise and insights to this work, and the editor Paul Wine for his suggestions. References and Notes (1) Fantozzi, L.; Ferrarac, R.; Frontini, F. P.; Dini, F. Sci. Total EnViron. 2009, 407, 917. (2) Mason, R. P.; Morel, F. M. M.; Hemond, H. F. Water, Air, Soil Pollut. 1995, 80, 775. (3) Siciliano, S. D.; O’Driscoll, N. J.; Lean, D. R. S. EnViron. Sci. Technol. 2002, 36, 3064. (4) Amyot, M.; Mierle, G.; Lean, D.; McQueen, D. J. Geochim. Cosmochim. Acta 1997, 61, 975. (5) Amyot, M.; Mierle, G.; Lean, D. R. S.; Mcqueen, D. J. EnViron. Sci. Technol. 1994, 28, 2366. (6) Xiao, Z. F.; Stromberg, D.; Lindqvist, O. Water, Air, Soil Pollut. 1995, 80, 789. (7) Alberts, J. J.; Schindler, J. E.; Miller, R. W.; D.E., N. Science 1974, 184, 895. (8) Allard, B.; Arsenie, I. Water, Air, Soil Pollut. 1991, 56, 457. (9) Bauer, M.; Heitmann, T.; Macalady, D. L.; Blodau, C. EnViron. Sci. Technol. 2007, 41, 139. (10) Chen, J.; Gu, B. H.; Royer, R. A.; Burgos, W. D. Sci. Total EnViron. 2003, 307, 167. (11) Kappler, A.; Haderlein, S. B. EnViron. Sci. Technol. 2003, 37, 2714. (12) Kunkely, H.; Horvath, O.; Vogler, A. Coord. Chem. ReV. 1997, 159, 85. (13) Zhang, H. Photochemical redox reactions of mercury. Recent DeVelopments in Mercury Science 2006, 120, 37. (14) Scott, D. T.; McKnight, D. M.; Blunt-Harris, E. L.; Kolesar, S. E.; Lovley, D. R. EnViron. Sci. Technol. 1998, 32, 2984. (15) Meisel, D.; Fessenden, R. W. J. Am. Chem. Soc. 1976, 98, 7505. (16) Rosso, K. M.; Smith, D. M. A.; Wang, Z. M.; Ainsworth, C. C.; Fredrickson, J. K. J. Phys. Chem. A 2004, 108, 3292. (17) Mancini, S. A.; Devine, C. E.; Elsner, M.; Nandi, M. E.; Ulrich, A. C.; Edwards, E. A.; Lollar, B. S. EnViron. Sci. Technol. 2008, 42, 8290. (18) Bigeleisen, J.; Mayer, M. G. J. Chem. Phys. 1947, 15, 261. (19) Urey, H. C. J. Chem. Soc. 1947, 562. (20) Bergquist, B. A.; Blum, J. D. Science 2007, 318, 417. (21) Biswas, A.; Blum, J. D.; Bergquist, B. A.; Keeler, G. J.; Xie, Z. Q. EnViron. Sci. Technol. 2008, 42, 8303. (22) Ghosh, S.; Xu, Y. F.; Humayun, M.; Odom, L. Geochem. Geophys. Geosyst. 2008, 9, Q03004. (23) Jackson, T. A.; Whittle, D. M.; Evans, M. S.; Muir, D. C. G. Appl. Geochem. 2008, 23, 547. (24) Sherman, L. S.; Blum, J. D.; Nordstrom, D. K.; McCleskey, R. B.; Barkay, T.; Vetriani, C. Earth Planet. Sci. Lett. 2009, 279, 86. (25) Zambardi, T.; Sonke, J. E.; Toutain, J. P.; Sortino, F.; Shinohara, H. Earth Planet. Sci. Lett. 2009, 277, 236.

Isotope Effects of Hg in Abiotic Dark Reduction (26) Carignan, J.; Estrade, N.; Sonke, J. E.; Donard, O. F. X. EnViron. Sci. Technol. 2009, 43, 5660. (27) Buchachenko, A. L. J. Phys. Chem. A 2001, 105, 9995. (28) Turro, N. J. Proc. Natl. Acad. Sci. U.S.A. 1983, 80, 609. (29) Turro, N. J.; Kraeutler, B. Acc. Chem. Res. 1980, 13, 369. (30) Steiner, U. E.; Ulrich, T. Chem. ReV. 1989, 89, 51. (31) Grissom, C. B. Chem. ReV. 1995, 95, 3. (32) Bigeleisen, J. J. Am. Chem. Soc. 1996, 118, 3676. (33) King, W. H. Isotope shifts in atomic spectra; Plenum Press: New York, 1984. (34) Kowalewska, D.; Bekk, K.; Goring, S.; Hanser, A.; Kalber, W.; Meisel, G.; Rebel, H. Phys. ReV. A 1991, 44, R1442. (35) Ulm, G.; Bhattacherjee, S. K.; Dabkiewicz, P.; Huber, G.; Kluge, H. J.; Kuhl, T.; Lochmann, H.; Otten, E. W.; Wendt, K.; Ahmad, S. A.; Klempt, W.; Neugart, R. Z. Phys. A: Hadrons Nucl. 1986, 325, 247. (36) Fujii, T.; Moynier, F.; Telouk, P.; Albarede, F. Astrophys. J. 2006, 647, 1506. (37) Fujii, T.; Yamamoto, T.; Inagawa, J.; Gunji, K.; Watanabe, K.; Nishizawa, K. SolVent Extr. Ion Exch. 2000, 18, 1155. (38) Moynier, F.; Fujii, T.; Telouk, P. Anal. Chim. Acta 2009, 632, 234. (39) Fujii, T.; Moynier, F.; Uehara, A.; Abe, M.; Yin, Q.-Z.; Nagai, T.; Yamana, H. J. Phys. Chem. A 2009, 113, 12225. (40) Fujii, T.; Moynier, F.; Albarede, F. Chem. Geol. 2009, 267, 139. (41) Fujii, T.; Moynier, F.; Telouk, P.; Albarede, F. Chem. Geol. 2009, 267, 157. (42) Lamborg, C. H.; Tseng, C. M.; Fitzgerald, W. F.; Balcom, P. H.; Hammerschmidt, C. R. EnViron. Sci. Technol. 2003, 37, 3316. (43) Zheng, W.; Foucher, D.; Hintelmann, H. J. Anal. At. Spectrom. 2007, 22, 1097. (44) Seby, F.; Potin-Gautier, M.; Giffaut, E.; Donard, O. F. X. Geochim. Cosmochim. Acta 2001, 65, 3041. (45) Foucher, D.; Hintelmann, H. Anal. Bioanal. Chem. 2006, 384, 1470. (46) Blum, J. D.; Bergquist, B. A. Anal. Bioanal. Chem. 2007, 388, 353. (47) Rocha, J. C.; Junior, E. S.; Zara, L. F.; Rosa, A. H.; dos Santos, A.; Burba, P. Talanta 2000, 53, 551. (48) Rocha, J. C.; Sargentini, E.; Zara, L. F.; Rosa, A. H.; dos Santos, A.; Burba, P. Talanta 2003, 61, 699. (49) Zheng, W.; Hintelmann, H. Geochim. Cosmochim. Acta 2009, 73, 6704. (50) Xia, K.; Skyllberg, U. L.; Bleam, W. F.; Bloom, P. R.; Nater, E. A.; Helmke, P. A. EnViron. Sci. Technol. 1999, 33, 257. (51) Hesterberg, D.; Chou, J. W.; Hutchison, K. J.; Sayers, D. E. EnViron. Sci. Technol. 2001, 35, 2741. (52) Drexel, R. T.; Haitzer, M.; Ryan, J. N.; Aiken, G. R.; Nagy, K. L. EnViron. Sci. Technol. 2002, 36, 4058. (53) Sutton, R.; Sposito, G. EnViron. Sci. Technol. 2005, 39, 9009. (54) Harris, R. C.; Rudd, J. W. M.; Almyot, M.; Babiarz, C. L.; Beaty, K. G.; Blanchfield, P. J.; Bodaly, R. A.; Branfireun, B. A.; Gilmour, C. C.; Graydon, J. A.; Heyes, A.; Hintelmann, H.; Hurley, J. P.; Kelly, C. A.; Krabbenhoft, D. P.; Lindberg, S. E.; Mason, R. P.; Paterson, M. J.; Podemski, C. L.; Robinson, A.; Sandilands, K. A.; Southworth, G. R.; Louis, V. L. S.; Tate, M. T. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 16586. (55) Hintelmann, H.; Harris, R.; Heyes, A.; Hurley, J. P.; Kelly, C. A.; Krabbenhoft, D. P.; Lindberg, S.; Rudd, J. W. M.; Scott, K. J.; St Louis, V. L. EnViron. Sci. Technol. 2002, 36, 5034. (56) Poulain, A. J.; Orihel, D. M.; Amyot, M.; Paterson, M. J.; Hintelmann, H.; Southworth, G. R. Chemosphere 2006, 65, 2199.

J. Phys. Chem. A, Vol. 114, No. 12, 2010 4245 (57) Curtis, G. P.; Reinhard, M. EnViron. Sci. Technol. 1994, 28, 2393. (58) Van der Zee, F. R.; Cervantes, F. J. Biotechnol. AdV. 2009, 27, 256. (59) Kritee, K.; Blum, J. D.; Barkay, T. EnViron. Sci. Technol. 2008, 42, 9171. (60) Kritee, K.; Blum, J. D.; Johnson, M. W.; Bergquist, B. A.; Barkay, T. EnViron. Sci. Technol. 2007, 41, 1889. (61) Yang, L.; Sturgeon, R. Anal. Bioanal. Chem. 2009, 393, 377. (62) Estrade, N.; Carignan, J.; Sonke, J. E.; Donard, O. F. X. Geochim. Cosmochim. Acta 2009, 73, 2693. (63) Alkhazov, G. D.; Barzakh, A. E.; Bolshakov, V. A.; Denisov, V. P.; Ivanov, V. S.; Sergeyev, Y. Y.; Chubukov, I. Y.; Tikhonov, V. I.; Letokhov, V. S.; Mishin, V. I.; Sekatsky, S. K.; Fedoseyev, V. N. Z. Phys. A: Hadrons Nucl. 1990, 337, 257. (64) Anselment, M.; Faubel, W.; Goring, S.; Hanser, A.; Meisel, G.; Rebel, H.; Schatz, G. Nucl. Phys. A 1986, 451, 471. (65) Bernard, A.; Bruggemeyer, H.; Pfeufer, V. Z. Phys. A: Hadrons Nucl. 1985, 322, 1. (66) Heilig, K. Hyperfine Interact. 1987, 38, 803. (67) Young, E. D.; Galy, A.; Nagahara, H. Geochim. Cosmochim. Acta 2002, 66, 1095. (68) Hahn, A. A.; Miller, J. P.; Powers, R. J.; Zehnder, A. Nucl. Phys. A 1979, 314, 361. (69) Nadjakov, E. G.; Marinova, K. P.; Gangrsky, Y. P. At. Data Nucl. Data Tables 1994, 56, 133. (70) Angeli, I. At. Data Nucl. Data Tables 2004, 87, 185. (71) Ellison, S. L. R.; Rosslein, M.; Williams, A. 2001, Eurachem/Citac, http://www.measurementuncertainty.org. (72) Fujii, T.; Kawashiro, F.; Yamamoto, T.; Nomura, M.; Nishizawa, K. SolVent Extr. Ion Exch. 1999, 17, 177. (73) Nishizawa, K.; Maeda, Y.; Kawashiro, F.; Fujii, T.; Yamamoto, T.; Hirata, T. Sep. Sci. Technol. 1998, 33, 2101. (74) Schauble, E. A. Geochim. Cosmochim. Acta 2007, 71, 2170. (75) Bigeleisen, J. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 9393. (76) Fujii, T.; Hirata, T.; Shibahara, Y.; Nishizawa, K. Phys. Chem. Chem. Phys. 2001, 3, 3125. (77) Fujii, T.; Yamamoto, T.; Nishizawa, K.; Inagawa, J.; Gunji, K.; Watanabe, K. SolVent Extr. Ion Exch. 1998, 16, 985. (78) Knyazev, D. A.; Semin, G. K.; Bochkarev, A. V. Polyhedron 1999, 18, 2579. (79) Knyazev, D. A.; Myasoedov, N. F. Sep. Sci. Technol. 2001, 36, 1677. (80) Thiemens, M. H. Science 1999, 283, 341. (81) Thiemens, M. H. Annu. ReV. Earth Planet. Sci. 2006, 34, 217. (82) Weston, R. E. Chem. ReV. 1999, 99, 2115. (83) Criss, R. E. Principles of stable isotope distribution; Oxford University Press: New York, 1999. (84) Hoefs, J. Stable isotope geochemistry; Springer-Verlag: Berlin, 2009. (85) Buchachenko, A. L.; Ivanov, V. L.; Roznyatovskii, V. A.; Vorob’ev, A. K.; Ustynyuk, Y. A. Dokl. Phys. Chem. 2008, 420, 85. (86) Buchachenko, A. L.; Kouznetsov, D. A.; Shishkov, A. V. J. Phys. Chem. A 2004, 108, 707. (87) Khudyakov, I. V.; Serebrennikov, Y. A.; Turro, N. J. Chem. ReV. 1993, 93, 537.

JP910353Y