Environmental Engineering Reference
In-Depth Information
nature (Zhang and Lindberg, 1999). Hg in the environment
exists in three oxidation states, Hg(II), Hg(I), and Hg(0). In
the atmosphere, Hg exists as Hg(0), particulate Hg, rGM,
and dimethylmercury (Me
2
Hg). Hg(0) is the dominant
(
.
95%) form of all atmospheric species and is normally
transported the farthest (Mason et al., 1994). For most soils
with natural background Hg concentrations, atmospheric
Hg(0) deposition is the most important input vector (Engle
et al., 2001). Hg(0) is volatile and has a vapor pressure of
2.5×10
2
6
atm and a dimensionless Henry's law constant
of 0.32 [Hg(0)
(g)
:Hg(0)
(aq)
mass ratio] at 25°C (Schroeder
et al., 1991). Given the volatile nature of Hg(0), its dissolved
concentration in soil water is expected to influence its gas-
eous concentration, and vice versa. This behavior is consis-
tent with Henry's law for gases and Fick's law of diffusion
(Zhang and Lindberg, 1999). Hg(0) has a relatively large sta-
bility field in the environment, covering a wide range of pH
and redox potentials (Andersson, 1979; Drever, 1997). From
a thermodynamic point of view, the dominant Hg species
in oxic environments is Hg(II). The presence of Hg(0) in
oxic environments, however, indicates its kinetic stability.
Even though Hg(0) can be emitted from a variety of natural
surfaces (Mason et al., 1994), this chapter covers Hg emis-
sions from soils only. Hg emissions from other natural sur-
faces, such as vegetation and surface waters, as well as emis-
sions from the whole watersheds, are discussed by Shanley
and Bishop (this topic, chapter 8).
A large body of literature has been devoted to Hg emis-
sion from porous solid surfaces, especially uncontaminated
and contaminated soils. Some of this literature has been
reviewed by Zhang and Lindberg (1999), Schlüter (2000),
Grigal (2002), and Gustin et al. (2008). Gustin et al. (2008)
presented an update on the current estimates of global Hg
emissions. Global natural (volcanic and geothermal) and
anthropogenic emissions are in the range of ~800-3000,
and ~2000-2400 Mg yr
2
1
, respectively. Based on a review of
the published measurements, soil Hg emission flux ranges
from -2 to 13 ng m
2
2
hr
2
1
. It should be noted that most of
the published flux data are collected during the summer
and in the daytime, and as such, are not a good representa-
tive of the diel or seasonal variations (Gustin et al., 2008).
Depending on the physical and chemical soil characteris-
tics, as described below, and the ambient soil and air Hg(0)
concentrations, soils can act as sources or sinks for atmo-
spheric Hg(0) (Xin and Gustin, 2007). Soil Hg(0) reemission
is an important process to consider when estimating the Hg
budget of a watershed. In boreal forests, Grigal (2002) esti-
mates the soil Hg efflux at 11 ng m
2
2
hr
2
1
during the grow-
ing season, sufficient to reemit most of the deposited Hg.
Abbott et al. (2003) observed relatively low concentrations
of Hg in the soils close to a calciner, a point source Hg emit-
ter, and concluded that most of the deposited was reduced
to Hg(0) in and reemitted from the soil. Biester et al. (2002b)
studied Hg in soils downwind from chlor-alkali plants and
found no Hg(0) in any of the soils that received atmospheric
Hg deposition originating from the chlor-alkali plants, even
though most of the emitted Hg from these plants was in
Hg(0) form. They proposed that most of the deposited Hg(0)
would be reemitted before oxidation to Hg(II). Demers et al.
(2007) mentioned Hg(0) reemission of throughfall input
to a coniferous forest floor as a major Hg loss mechanism,
and proposed that in such forests, a large fraction of the
Hg litterfall input might indeed consist of throughfall Hg
recycled by reemission from the forest floor. Gustin et al.
(2008) proposed that only a relatively small percentage
of Hg introduced into soils via wet and dry deposition is
reemitted immediately. Instead, Hg reemission is brought
about gradually and is controlled by the physical and chem-
ical factors mentioned below (Xin and Gustin, 2007; Gustin
et al., 2008).
Zhang and Lindberg (1999) suggested that depending on
the dominance of Hg(0) or inorganic Hg(II) in soil, adsorp-
tion and desorption of Hg(0) or reduction of Hg(II) could be
the rate-limiting steps for Hg emission. Schlüter (2000) also
mentions the reduction step as the rate-limiting step in Hg
emission from soil. This, of course, assumes that physical
transport of Hg(0) from soil solution to soil air is not the
rate-limiting step. The reported soil-air Hg(0) concentra-
tions generally range between 1 and 53 ng m
2
3
(Johnson
and Lindberg, 1995), whereas reliable data on soil water
Hg(0) do not exist (Zhang and Lindberg, 1999). If such
data are available, then using the Henry's law constant,
the extent of departure from equilibrium between the two
phases can be evaluated. If the estimated equilibrium soil-
water Hg(0) concentration is significantly smaller than the
measured soil-water Hg(0) concentration, then: (a) Hg(0) is
produced in the soil environment, and (b) the rate of Hg(0)
production in soil is greater than its rate of mass transfer
into soil air. If the estimated and measured soil-water Hg(0)
are close, then the rate-determining step would be the
biotic/abiotic processes leading to Hg(0) production.
fAcToRS ThAT conTRol MeRcuRy eMiSSion
fRoM SoilS
Processes that control reduction of Hg(II) to Hg(0) in soils
can take place in soil solution or on soil surfaces and can
be biotic or abiotic. only the abiotic processes involved in
Hg(II) reduction will be reviewed in this chapter. Biotic
processes in Hg(II) reduction, which may be dominant
in some soils and sediments, are reviewed by Barkay and
Wagner-Dobler (2005) and Swartzendruber and Jaffe (this
topic, chapter 1). Factors that control abiotic Hg reduction
and emission from soils include chemical factors, such as
SoM content, Hg(II) concentration, and the presence of
oxidants, such as atmospheric ozone (Gustin et al., 1997;
Schlüter, 2000; Engle et al., 2005; Xin and Gustin, 2007;
Mauclair et al., 2008), and physical factors, such as soil
moisture content and precipitation (Gustin et al., 1999;
Gustin and Stamenkovic, 2005; Xin et al., 2007; Lindberg
et al., 1999), sunlight (Gustin et al., 1996; Carpi and Lindberg,
1997; Zhang and Lindberg, 1999; Moore and Carpi, 2005;
