114
F. J. Millero
Field measurements made on the oxidation of H2S in the Black Sea (Millero 1991b),
the Framvaren Fjord (Millero 1991b; Yao and Millero 1995b), the Chesapeake Bay
(Millero 1991c), and the Cariaco Trench (Zhang and Millero 1993b) yielded half times
that were much faster than determined in the laboratory on Gulf Stream sea water.
The rates of oxidation of H2S in surface waters (with added NaHS), deep waters and
mixtures of surface and deep waters in the Cariaco Trench are shown in Fig. 4.21 along
with the half times for these runs. As will be discussed later, the cause of the faster
rates of oxidation of H2S in these natural waters is due to the high concentrations of
Fe2+ and Mn2+. The intermediates SO;- and S20;- were determined, along with the
disappearance of H2S, during the course of the oxidation of sulfide in the waters of
the Cariaco Trench (Zhang and Millero 1993b). The results are shown in Fig. 4.22. The
decrease of H 2 S and resultant increase of SO;-, S20;- and soi- occurring during the
oxidation is similar to laboratory studies and in anoxic basins.
To determine if this increase was due to trace metals, we have measured the rates
of oxidation ofH2S in seawater with added transition metals (Vazquez et al.1989). These
studies have shown that at concentrations below 300 nM, the rates are only affected
by Fe 2 +, Cu 2 + and Pb 2 +. At higher metal concentrations, the rates of oxidation of H 2S
increase for all the metals except Zn 2 + (Fig. 4.23). The order of the increase in the rates
at higher concentrations for these metals is Fe 2 + > Pb 2 + > Cu 2 + > Fe3+ > Cd 2 + >
Ni2+ > C0 2 + > Mn2+.
Only Fe2+ and Mn2+ have levels in anoxic basins high enough to affect the oxidation ofH 2 S. The effect of metals on the oxidation ofH2S with oxygen can be estimated
from (Fig. 4.24)
log(k I ko) = a + blog [M]
where
• a = 6.55, b = 0.820
• a = 5.18, b = 0.717
• a = 1.68, b = 0.284
for Fe(II)
for Fe(I1I)
for Mn(II)
These equations are valid, respectively, from 10- 8 to 10- 5 .3,10- 7 . 2 to 10- 3 . 3 and 10- 5 . 9
to 10- 3 . 3 M. At the maximum levels of Fe 2 + in the Cariaco Trench, one would expect the
rates of oxidation to be 17 times faster due to Fe2+. These estimates are the same orders as found in our direct measurements. The calculated half times (t1l2 = 17.2,2.7, and
1.5 h) respectively, for the surface, mixed and deep waters are in good agreement with
the measured values (tll2 = 17.2,3.0, and 1.6 h).
The increase in the rates by Fe(II) at low concentrations is truly a catalytic effect.
This is caused by the oxidation of Fe(II) (Millero et al. 1987b):
Fe(OHh + O2 ~ Fe(OH); + 02"
The oxidation products 02" and Fe( OR); may also oxidize H2S. The reaction of dissolved or particulate Fe(III) with H 2 S in the anoxic can regenerate Fe(II) to complete
the catalytic cycle. The overall reaction is given by
F. J. Millero
Field measurements made on the oxidation of H2S in the Black Sea (Millero 1991b),
the Framvaren Fjord (Millero 1991b; Yao and Millero 1995b), the Chesapeake Bay
(Millero 1991c), and the Cariaco Trench (Zhang and Millero 1993b) yielded half times
that were much faster than determined in the laboratory on Gulf Stream sea water.
The rates of oxidation of H2S in surface waters (with added NaHS), deep waters and
mixtures of surface and deep waters in the Cariaco Trench are shown in Fig. 4.21 along
with the half times for these runs. As will be discussed later, the cause of the faster
rates of oxidation of H2S in these natural waters is due to the high concentrations of
Fe2+ and Mn2+. The intermediates SO;- and S20;- were determined, along with the
disappearance of H2S, during the course of the oxidation of sulfide in the waters of
the Cariaco Trench (Zhang and Millero 1993b). The results are shown in Fig. 4.22. The
decrease of H 2 S and resultant increase of SO;-, S20;- and soi- occurring during the
oxidation is similar to laboratory studies and in anoxic basins.
To determine if this increase was due to trace metals, we have measured the rates
of oxidation ofH2S in seawater with added transition metals (Vazquez et al.1989). These
studies have shown that at concentrations below 300 nM, the rates are only affected
by Fe 2 +, Cu 2 + and Pb 2 +. At higher metal concentrations, the rates of oxidation of H 2S
increase for all the metals except Zn 2 + (Fig. 4.23). The order of the increase in the rates
at higher concentrations for these metals is Fe 2 + > Pb 2 + > Cu 2 + > Fe3+ > Cd 2 + >
Ni2+ > C0 2 + > Mn2+.
Only Fe2+ and Mn2+ have levels in anoxic basins high enough to affect the oxidation ofH 2 S. The effect of metals on the oxidation ofH2S with oxygen can be estimated
from (Fig. 4.24)
log(k I ko) = a + blog [M]
where
• a = 6.55, b = 0.820
• a = 5.18, b = 0.717
• a = 1.68, b = 0.284
for Fe(II)
for Fe(I1I)
for Mn(II)
These equations are valid, respectively, from 10- 8 to 10- 5 .3,10- 7 . 2 to 10- 3 . 3 and 10- 5 . 9
to 10- 3 . 3 M. At the maximum levels of Fe 2 + in the Cariaco Trench, one would expect the
rates of oxidation to be 17 times faster due to Fe2+. These estimates are the same orders as found in our direct measurements. The calculated half times (t1l2 = 17.2,2.7, and
1.5 h) respectively, for the surface, mixed and deep waters are in good agreement with
the measured values (tll2 = 17.2,3.0, and 1.6 h).
The increase in the rates by Fe(II) at low concentrations is truly a catalytic effect.
This is caused by the oxidation of Fe(II) (Millero et al. 1987b):
Fe(OHh + O2 ~ Fe(OH); + 02"
The oxidation products 02" and Fe( OR); may also oxidize H2S. The reaction of dissolved or particulate Fe(III) with H 2 S in the anoxic can regenerate Fe(II) to complete
the catalytic cycle. The overall reaction is given by