The Sulphur-mercury (II) System in Natural Waters

THE SULPHUR-MERCURY(H) SYSTEM IN NATURAL WATERS

DAVID DYRSSEN AND MARGARETA WEDBORG Department of Analytical and Marine Chemistry University of G6teborg and Chalmers University of Technology S-412 96 G6teborg, Sweden

ABSTRACT. Sulphur is an essential element for aquatic biosystems, the life processes of which lead to the formation of low moelcular weight S compounds in the water. The results of our calculations indicate a pronounced tendency for Hg(II) to form HgS (or HgOHSH) and Hg(SR)2 complexes in the presence of HzS and thiols. Likewise, MeHg will form CH3HgSH and CH3HgSR complexes, but in this case the chloride complex will dominate at low concentrations of H2S and thiols. In acidic low salinity water, CH3HgC1 is the dominant MeHg species at the lowest concentration of sulphide/thiols (0.1 nM), whereas a hundredfold increase of the sulphide/thiol concentration, or an increase of the pH to neutral or slightly alkaline conditions, will result in a total dominance for CH3HgSH and CH3HgSR.

1. Introduction

Aquatic biosystems require a series of S-containing compounds. These compounds are biosynthesized by assimilation and reduction of sulphate ions (Andreae, 1990; Dyrssen, 1989). Upon degradation of the organic matter, several low molecular weight S compounds are formed (Cutter and Krahforst, 1988). Some of them, such as H2S and thiols, form strong complexes with Hg 2+ and CH3Hg +. Previously, we have calculated the state of trace sulphide in seawater (Dyrssen and Wedborg, 1989). In this paper we shall examine the state of Hg(II) (0.01 nM) and MeHg (0.001 nM) as functions of pH (4 to 9), pC1 (0.3 to 3.7), and total concentrations of HzS and thiols within the range 0.1 to 10 nM.

2. Materials and Methods

2.1. ESTIMATION OF STABILITY CONSTANTS

The following constants have been experimentally determined and can be found in various tables of stability constants (Sill6n and Martell, 1964, 1971; Smith and Martell, 1976; Martell and Smith, 1982; Smith and Martell, 1989). For our estimations, the ionic strength dependence may be neglected.

Hg 2+ + 2SH- -~- Hg(SH)2 ; log K = 37.72

H + + HgS2 2- .~- HgS2H- ; log K = 8.30

H + + HgS2H- .~- Hg(SH)2 ; log K = 6.19

Water, Air, and Soil Pollution 56:507-519, 1991. © 1991 Kluwer Academic Publishers. Printed in the Netherlands.

5O8

H + + SH- ~ H2S

Hg 2+ + 2C1- ~--- HgC12

Hg 2+ + C1- -~- HgC1 +

Hg 2+ + 2OH- -~- Hg(OH)2

Hg 2+ + OH- ~- HgOH +

D. DYRSSEN AND M. WEDBORG

; l o g K = 6.88

; log [}2 = 13.22

; log K 1 = 6.74

; log [32 = 22.23

; log K 1 = 10.67.

The solubility constant was calculated by Dyrssen and Kremling (1990):

HgS(s) + H + ~ Hg 2+ + SH- ; log K s = -38.9.

The following constants were found to fit the titration of thiols (Dyrssen and Wedborg, 1986):

Hg 2+ + 2RS" ~ Hg(SR)2 ; log K = 41.6

H + + RS- ~ RSH ; log K = 9.34.

40

.30

2 0

10-

log#2 /.#~#S-

CN H_@ .I~S-

OH-Q I - ~

log# I 1 J I

5 10 15 2O

Fig. 1 Stability constant for the formation of HgL 2 complexes plotted v s the constants for HgL +. The straight line of log [32 = 1.87 log ~1 + 0.3 was used to estimate log [31 for the complexes HgSH + and HgSR +.

THE SULPHUR-MERCURY(II) SYSTEM IN NATURAL WATERS

For MeHg the following constants Schellenberg (1965):

CH3Hg + + C1- -~- CH3HgC1

CH3Hg ÷ + OH- ~ CH3HgOH

CH3Hg + + RS- .~- CH3HgSR

were determined by Schwarzenbach

; log K 1 = 5.25

; log K 1 = 9.37

; log K 1 = 16.12.

509

and

From a plot of log [32 vs log K 1 (= log [31) for the ligands CI-, Br-, I-, SCN-, OH-, and CN-, the following linear relationship was found for Hg 2+ (see Figure 1):

log [32 = 1.87 log K 1 + 0.3.

This enables the estimation of the following constants (see Figure 1):

Hg 2+ + SH- ~ H g S H + ; log K 1 = 20.0

Hg 2+ + RS- ~ HgSR + ; log K 1 = 22.1.

We have previously shown (Anf~ilt et al., 1968) that mixed complexes MAmB n may be calculated by the statistical relation

log [311 = log 2 + 0.5 (log A[32 + log B[~2) (m = n = 1).

F rom this relation the following mixed equilibrium constants were calculated:

Hg 2+ + C1- + SH- ~- HgC1SH ; log K = 25.77

Hg 2+ + OH- + SR- ~ H g O H S R ; log K = 32.2

Hg 2+ + OH- + SH- ~ H g O H S H ; log K = 30.3.

The latter constant, together with K s and the ionic product of water (pK w = 13.7), gives the solubility of HgS in water:

HgS(s) + H 2 0 ~ H g O H S H ; log K -- -22.3.

The mixed complex H g O H S H cannot be distinguished f rom the dissolved complex HgS (aq) 'by equilibrium measurements since

HgS(aq) + H 2 0 ~ H g O H S H ; log K is undetermined.

There is also a straight line relationship between the second consecutive stability constant for Hg 2+

H g L ÷ + L- ~ H g L a

and the first constant for CH3Hg +

CH3Hg ÷ + L- ~-~ CH3HgL

; log K2

; log K 1.

510 D. DYRSSEN AND M. WEDBORG

From known values for the consecutive K 2 constants for the Hg(II) complexation with the ligands C1-, Br-, I-, OH-, SCN-, and CN- we obtained (Figure 2):

log K 2 = 1.2 log K 1 + 0.3.

With this equation we estimated (Figure 2):

CH3Hg + + SH- ~- CH3HgSH ; log K t = 14.5

and

Hg 2+ + RS- ~- HgSR + ; log K = 41.6 - 19.6 = 22.0.

20 log K 2 /

CN-

15

10'

SCNt~ Br-

5 / c~- log K1

I I I

5 10 15

Fig. 2 Consecutive stability constants K2= [HgL2]/[HgL+][L -] plotted v s the constant K 1 = [CH3HgL]/[CH3Hg+][L-]. The straight line log K 2 = 1.2 log K 1 + 0.3 was used to estimate log K2for the ligands RS- and HS-.

Previously (Dyrssen and Wedborg, 1974) we have estimated all of the mixed stability constants for HgClmBr n, but since the bromide concentration is low in low salinity waters we need to take into consideration only the species with m = 3 and 4 and n = O, i .e .

Hg 2+ + 3C1- ~ HgC13- ; log ~3 = 14.07

THE SULPHUR-MERCURY(II) SYSTEM IN NATURAL WATERS 511

Hg 2+ + 4C1- ~- HgC142- ; log [~4 = 15.07.

In seawater, mixed complexes with Br- (mainly HgC12Br and HgCI3Br 2-) may be included in the calculation, but since HgC142- is the dominant Hg-halogenide form this should not be of importance for the result (see Dyrssen and Wedborg, 1974).

2.2. A CONTRADICTION

The stability constants for HgOHSH in equilibrium with solid HgS(s) which we calculated above has a very low value (log K = -22.3), especially if we compare to experimental results for ZnS and CdS by Gtibeli and Ste-Marie (1967) and Ste-Marie et al. (1964). See also Dyrssen (1985, 1988) and Dyrssen and Wedborg (1989). The mixed stability constant for ZnOHSH is estimated to be log [~11 = 11.71 from the statistical relation given above. This result, together with the solubility constant for the equilibrium

ZnS(s) + H + ~ Zn 2+ + SH- ; log K s = -10.89

and the ionic product of H20 gives log Ksl(Calc ) for the equilibrium

ZnS(s) + H20 ~ ZnOHSH ; log Ksffcalc ) = -12.95.

The experimental value of Gtibeli and Ste-Marie (1967) is log Ksl = -5.87 for ZnS(aq) in equilibrium with ZnS(s). The corresponding values for Cd are log Ksffcalc ) = -16.04 (from log [311, log K s, and pKw) and log Ksl = -6.85 from experiments (Ste-Marie et al., 1964; Dyrssen, 1988). The simple relation

log Ksl(Calc ) = 2.26 log Ksffexp)

gives log Ks1 = -10 for HgS. This value was used on our calculation of the state of H2S in seawater (Dyrssen and Wedborg, 1989). Thus we are left with a choice of one "experimental" constant

HgS(s) -~- HgS(aq)

and one calculated constant

; log Ksl = -10

HgS(s) + H20 ~ HgOHSH ; log Ksl = -22.3.

From each of these constants we may calculate the H + dissociation constant of HgSH +

HgSH + ~ HgS(aq) + H + ; pK a = -8.9

HgSH + + H20 ~ HgOHSH + H + ; pK a = +3.4.

If these two values are compared with the constants for the following hydrolysis reactions

HgOH + + H20 ~- Hg(OH)2 + H +

HgC1 + + H20 ~ HgC1OH + H +

; pK a = 2.6

; pK a = 3.1


only the latter value of 3.4 is acceptable. One is therefore left with the impression that the experimental values for ZnS and

CdS represent some colloidal state, which, however, may correspond to the real world. From equilibrium studies it is not possible to make a distinction between HgS(aq) and HgOHSH, since OH- + SH- is equivalent to S 2- + H20. The accepted H + dissociation constant for the HS- ion is around pK a = 17, and therefore, since the constant for

OH- + SH- ~ S 2- + H20

is very small (0.001), OH- and SH- may coexist in the same complex.

3. Results

3.1. Hg 2+ - SH- IN SEAWATER

We included the following species in the model calculations: HgSH ÷, Hg(SH) 2, HgS2H-, HgS22-, HgC1 ÷, HgC12, HgC13-, HgC142-, HgOH ÷, Hg(OH)2, HgC1SH, HgOHSH, HgOHC1, and HgS(s). The conditions were: pH = 8, pC1 = 0.25, [SH-]tot = 0.1 to 10 nM, [Hg(II)]to t = 10 pM (2 ng L-I). For log Ksa = -22.3 practically all of the Hg was found as solid HgS. The most important dissolved species was HgS2H-. For log Ks1 = -10 all of the Hg was in the form of dissolved HgS (HgOHSH).

Table I. The speciation of CH3Hg + in seawater at various concentrations of sulphide and a total concentration of MeHg of 1 pM.

Species [SH-]~o t, nM 0.1 1 10

COnC. CO~C. CO~C.

fM % fM % /M %

CH3Hg + negl. negl. negl. -

CH3HgSH 204 20 719 72 962 96

CH3HgCI 694 69 245 24 33 3

CH3HgOH 103 10 36 4 5 0.5

THE SULPHUR-MERCURY(II) SYSTEM IN NATURAL WATERS

3.2. Hg 2+ - SH- IN LOW SALINITY WATER

513

The same species as above were taken into account. The pH was varied between 4 and 9 and pC1 was 3.7 (0.0002 M). [HS-]tot was varied between 0.1 and 10 nM with [Hg]tot = 10 pM. Again,,,practically all of the Hg was in the form of HgS(s) with Hg(SH)2, HgS2H- and HgS2 z- as the most important dissolved species, depending on the pH.

3.3. Hg 2+ - RS- IN SEAWATER AND LOW SALINITY WATER

In this case we included the species Hg 2+, HgSR +, Hg(SR) 2, HgOHSR, HgC1 +, HgC12, HgCl3-, HgCl42-, HgOH +, Hg(OH)2, and HgOHCl. As before, pH = 8 and pC1 = 0.25 was used for seawater, while for low salinity water the pH was varied between 4 and 9 with pC1 = 3.7. The total concentration of thiols was varied between 0.1 and 10 nM with [Hg]to t = 10 pM. The result was that practically all of the Hg should be present in the form of dissolved Hg(SR) 2, and that the mixed complex HgOHSR should be the second most important species (about 10 -4 times the concentration of Hg(SR)2).

3.4. CH3Hg + - SH- IN SEAWATER

In this case the species CH3HgSH, CH3HgC1, and CH3HgOH were taken into account. The total concentration of MeHg was 1 pM while the total sulphide concentration was varied between 0.1 and 10 nM, pH = 8, and pC1 = 0.25. Table I shows the dependence of the distribution on the sulphide concentration. For [HS-]to t = 0.1 nM, CH3HgC1 is the dominant species, and for [HS-]tot = 1 and 10 nM, CH3HgSH is most important.

Z 100 -

8 0 -

6 0 - 0.1 40-

2 0 -

I 4 ~ 6

nM' i i

7 8 pH

%

100

CH3HgSH 80--

6 0 - 1

4 0 -

2 0 - CH3HgCI

I 4 5

n M

I -~ 8 9 pH

Fig. 3 MeHg distribution in low salinity water with 0.0002 M chloride and low concentrations of H2S (0.1 and 1 nM). The two figures show the pH dependence of the speciation. For 10 nM sulphide, practically all of the MeHg is in the form of CH3HgSH.


The stability constant for the dinuclear complex

CH3Hg + + CH3HgS- ~ (CH3Hg)2S ; log K = 16.3

(Schwarzenbach and Schellenberg, 1965) is 1.8 log units larger than the constant for the corresponding reaction of MeHg + with SH- (log K 1 = 14.5, estimated from Figure 2 and defined above). However, at trace concentrations, (MeHg)2S is still negligible compared to MeHgSH.

If the constant for the equilibrium

CH3Hg + + SH- ~ CH3HgS- + H + ; log K = 7.0

(Schwarzenbach and Schellenberg, 1965) is combined with the log K a = 14.5 from Figure 2, the following acid dissociation constant can be estimated:

CH3HgSH ~ CH3HgS- + H + ; pK a = 7.5.

Consequently, the dissociation of MeHgSH can be expected to increase the dominance of the S complexes for pH's above 7.5.

3.5. CH3Hg + - SH- IN LOW SALINIITY WATER

The results for 0.1 and 1 nM total sulphide are shown in Figure 3. As a consquence of the increasing concentration of S H up to pH = 7.5 (pK a = 6.88), the concentration of the species CH3HgSH increases from about 50 % at pH 4 for 0.1 nM sulphide to 100 % at pH 5.5 to 7.5. At pH 7, the concentration of CH3HgOH starts to increase. For a total sulphide concentration of 10 nM the species CH3HgSH predominates within the whole pH range 4 to 9.

Table lI. The speciation of CH3Hg + in seawater at different total concentrations of thiols and a total concentration of MeHg of 1 pM.

Species [RS-]tot, nM 0.1 1 10

COnC. COLIC. C012C.

fM % fM % fM %

CH3Hg + negl. negl. negl. -

CH3HgSR 333 33 834 83 980 98

CH3HgC1 581 58 145 15 17 2

CH3HgOH 86 9 21 2 3 0.3


%

1 O0

8 0 -

60 -

40 -

2 0 -

g C l ~ 3HgRS

1 nM

CH3HgOH I I I I

4 5 6 7 8 9 pH

%

I O0

80-

60-

40-

2 0 - ~

I

4 5

1 nM

t I I

6 7 8 9 pH

%

100 "'-""'~CH~ Hg RS 8 0 -

6 0 -

4 0 -

2 0 -

~ C H 3 HgCl I

4 5

10 nM

I I I

6 7 8 9 pH

Figure 4. MeHg distribution in low salinity water with 0.0002 M chloride and low concentrations of thiols (0.1, 1, and 10 nM). The figures show the pH dependence of the speciation.


3.6. CH3Hg + - RS- IN S E A W A T E R

For the complexes CH3HgSR, CH3HgC1, and CH3HgOH the calculat ions show that the chlor ide complex predominates at low thiol concentrations (Table II). The hydoxide complex is always of minor importance.

3.7. CH3Hg + - RS- IN L O W SALINITY W A T E R

The results of the calculation on water with a low chloride concentration (0.0002 M) are shown in Figure 4. The hydroxide complex is again of minor importance and the chloride complex predominates only in an acidic environment and for low concentrations o f thiols.

3.8. C O M P A R I S O N W I T H H Y D R O G E N SELENIDE

Se(IV) and Se(VI) are the principal oxidation states of Se in natural waters. However , Se(-II) exists within particles, and it can be expected that biogenic reduction to the -II oxidat ion state occurs (Cutter, 1985; Cowan, 1988). Since Se is a bioact ive element which is chemical ly related to S, a comparison of the Hg(II)-Se(-II) and Hg(II)-S(-II) complexat ion may be justified.

Mehra and Gfibeli (1971) determined some constants for the Hg(II)-Se(-II) system:

H + + HSe ~ H2Se

H + + Se 2- ~ HSe-

H20 ~ H + + OH-

HgSe(s) ~ Hg 2+ + Se 2-

Hg 2+ + OH- + HSe- ~ HgOHSeH

Hg 2+ + OH- + 2HSe- ~- HgOH(SeH) 2-

Hg 2+ + 2OH + 2HSe ~--- Hg(OH)2(SeH)22-

HgSe(s) -~- HgSe(aq)

For comparison with the constants for the equi l ibr ium constants are calculated:

HgSe(s) + H + ~ Hg 2+ + HSe

HgSe(s) + H20 ~- HgOHSeH

Hg 2+ + 2HSe- + H20 ~- HgOH(SeH)2- + H ÷

Hg 2÷ + 2HSe- + 2H20 ~ Hg(OH)2(SeH)22- + 2H ÷

; log K = 3.48

; log K = 11.60

; pK w = 14.0

; log Ks0 = -56.6 + 0.2

; log K -- 51.2_+ 0.2

; log K = 52.8 + 0.3

; log K = 61.0 _+ 0.3

; log Ks1 = -7.83.

sulphide system, !the

; log K = -45.0

; log K = -7.8

; log K = 38.8

; log K = 33.0.

following

F rom Table III it is evident that the values of the stabili ty constants for the selenide complexes are higher than those for the corresponding sulphide complexes, with the exception that the intrinsic solubil i ty of HgSe(s) is higher than that for HgS(s). This


may, however, be an effect of the high value for the formation of HgOHSeH.

Table III. Comparison between stability constants for Hg(II) sulphide and Hg(II) selenide complexes.

E q u i l i b r i u m L = S 2- L = Se 2-

HgL(s) + H + -~- Hg 2+ + HL- -38.9 -45.0

HgL(s) -~- HgL(aq) - 10 (-22.6) -7.83

Hg 2+ + OH- + HL- ~ HgOHLH 30.3 51.2

Formation of HgOH(LH)2- or HgLzH- 31.53 38.8

Formation of Hg(OH)2(LH)22- or HgL22 23.23 33.0

4. Concluding remarks

The stability constants used and estimated in our present work are associated with uncertainties, caused by the neglection of temperature and ionic medium effects as well as the use of statistical estimates. However, adjustment of the values for the constants in the order of 0.3 to 0.7 log units will not alter the conclusions as to the dominant species, since the S complexes are very much stronger than the halogenide and hydroxide complexes (see Figure 1).

In a recent Ph.D. thesis (StrOmberg, 1990; see also StrOmberg et al. , 1991), relativistic calculations on some Hg compounds were presented. Of special interest in connection with our present work is Str6mberg's conclusion no. 10: "Hg(SH)2(aq) and HgSSH-(aq) will probably be photolyzed by sunlight". This process may lead to the formation of elemental Hg °. This finding, as well as our results, is in accordance with the idea that the formation of Hg-S complexes in natural waters with living organisms present may play a key role in the global cycle of Hg. Also, the results of our calculations may have some bearing on the effect of pH and lake liming on the Hg concentration, the binding of Hg to humic substances, and the biological availability of Hg.

518

Acknowledgements

D. DYRSSEN AND M. WEDBORG

The authors wish to thank Professor Cyrill Brosset, Dr. ~ke Iverfeldt, and Dr. Dan Str6mberg for some interesting discussions in relation to their work and our calculations on the Hg-S complexation. We also wish to thank the reviewers for helpful comments and the Swedish Natural Science Research Council for their support of our work on marine sulphur systems.

References

Andreae, M.O.: 1990, Mar. Chem. 30, 1.

Anf~ilt, T., Dyrssen, D., Ivanova, E., and Jagner, D.: 1968, Sv. Kern. Tidskrift 80, 340.

Cowan, C.E.: Review of Selenium Thermodynamic Data, EPRI Report EA-5655, prepared by Battelle, Pacific Northwest Laboratories, Richland, Wash., U.S.A.

Cutter, G.A.: 1985, Anal. Chem. 57, 2951.

Cutter, G.A. and Krahforst, C.F.: 1988, Geophys. Res. Lett. 15, 1393.

Dyrssen, D.: 1985, Mar. Chem. 15, 285.



Dyrssen, D. and Kremling, K.: 1990, Mar. Chem. 30, 193.

Dyrssen, D. and Wedborg, M.: 1974, Ch. 5 in The Sea, Vol. 5 (E.D. Goldberg ed.), Wiley-Interscience.

Dyrssen, D. and Wedborg, M.: 1986, Anal. Chim. Acta 180, 473.

Dyrssen, D. and Wedborg, M.: 1989, Mar. Chem. 26, 289.

Gtibeli A.O. and Ste-Marie, J.: 1967, Can. J. Chem. 45, 2101.

Martell A.E. and Smith, R.M.; 1982, Critical Stability Constants, Vol. 5: First supplement. Plenum Press, New York and London.

Mehra, M.C. and Giibeli A.O.: 1971, J. Less-Common Metals 25, 221.

Scwarzenbach G. and Schellenberg, M.: 1965, Helv. Chim. Acta 48, 28.

Sill6n, L.G. and Martell, A.E.: 1964, Stability Constants of Metal-Ion Complexes. Chemical Society, London, Special Publ. No. 17.


Sill6n, L.G. and Martell, A.E.: 1971, Stability Constants of Metal-Ion Complexes, Supplement No. 1. Chemical Society, London, Special Publ. No. 25.

Smith, R.M.. and Martell, A.E.: 1976, Critical Stability Constants, Vol. 4: Inorganic Complexes. Plenum Press, New York and London.

Smith, R.M.. and Martell, A.E.: 1989, Critical Stability Constants, Vol. 6: Second supplement. Plenum Press, New York and London.

Ste-Marie, J., Torma, A.E., and Giibeli A.O.: 1964, Can. J. Chem. 42, 662.

Str6mberg, F.: 1990, Some Mercury Compounds Studied b_2 Relativistic Quantum Chemical Methods. Ph.D. thesis, Department of Inorganic Chemistry, University of G6teborg, Sweden.

S tr6mberg, D, S tr6mberg, A., and Wahlgren, U.: 1991, this volume of WASP.

The Sulphur-mercury (II) System in Natural Waters

Documents

Transcript of The Sulphur-mercury (II) System in Natural Waters