Tải bản đầy đủ - 0trang
VI. Uses of pe–pH Diagrams
R I C H M O N D J. BARTLETT AND BRUCE R. JAMES
Figure 1. A pe-pH diagram for 0, reduction reactions, including partially reduced
intermediates:superoxide (Oi), hydroxyl free radical (OH .), and hydrogen peroxide (H202).
Reduction of 0,to O2is also included for comparison with 0, reduction reactions. Activity
of oxygencontaining ionic or molecular species is 1 6 ‘ M.except that Pa is 2 1 Wa.
reduction of 0, is relatively insensitive to the 0, partial pressure in this
and the hydroxyl free radical (OH-)are the most powerful
oxidants among the oxygen species (Table I), and the latter may be formed
during stepwise, four-electron reduction of 0, to Oy, H,O,, and H,O
(Fridovich, 1978). The low and high positions of the pe-pH lines for
superoxide (0;) reduction to H,O, and for superoxide oxidation to 0,
indicate that both a powerful oxidlzing and reducing agent is formed in the
first step of reduction of 0,. The enzyme superoxide dismutase scavenges
0; in living cells using 0, as the terminal electron acceptor, but relatively
little is known about its reactivity in biological and chemical processes in
soils that may be pertinent to our understanding of the formation of highly
reduced components (e.g., soil organic matter) and highly oxidized species
(e.g., NO;) that coexist in soil at chemical equilibrium or quasi-equilibrium.
Most reduction reactions of N species (Fig. 2 and Table I) are not
reversible, and therefore are not well defined by thermodynamic pe-pH
REDOX CHEMISTRY OF SOILS
Figure 2. A pe-pH diagram for nitrogen species half-reactions, including intermediates
formed or consumed in denitrification and dinitrogen fixation. Partial pressures of gaseous
intermediates are 0.01 kPa, except N, is 78 kPa; activity for NO;, NO:, and NH:is lo-' M.
relationships. The series of half reactions composing the process of denitrification, though, is instructive in that it identifies the wide range of pe for
reduction of each of the intermediates believed to form in the sequence of
electron acceptors used by microbes:
Step 1 of the sequence occurs at pe values less than those for reduction to
0, to H,O, whereas those for steps 2, 3, and 4 are increasingly higher. The
overall reduction of NO; to N,, though, is almost identical to that for the
0 2 / H , 0 couple. The overlap of pe values for the 0, and NO; reduction
intermediates indicates that denitrification and aerobic respiration may
occur at the same time under certain conditions when organic C is used as
the electron donor. They may not be mutually exclusive, as predicted from
log K values for the overall reactions, 0,to H,O and NO; to N, .
Manganese exists in soils in the 11, 111, and IV valence states, and the
latter two are most stable as oxides or oxyhydroxides. Trivalent Mn may
R I C H M O N D J. B A R T L E T T A N D BRUCE R. JAMES
Figure 3. A pe-pH diagram for Mn oxides, trivalent Mn ions, and superoxide. Ion
activities and H 2 0 2concentrations are lo-' M,Pa is 21 Wa.
exist as a cation, especially if stabilized by ligands, such as pyrophosphate
or citrate. The pe-pH relationships of Fig. 3 predict that different valences
of Mn in Mn,O,, MnOOH, and Mn02 affect the pe at which Mn2+would
be expected to form at pH < 7, but they are all similar at pH values near 7.
The Mn3+/Mn2+line indicates that at approximately pH 4, Mnw is a
powerful oxidant similar to 0;and Mn,O, (Fig. 3) if in equilibrium with
Mn2+.At pH values near 6.5, Mnw in equilibrium with MnO, is a powerful reductant, similar to H2and 0;.
This powerful oxidizing ability of Mn3+ in equilibrium with Mn2+ may
be pertinent to anaerobic soils that are exposed to 0,, and in which Mn2+
is oxidizing to form Mn(II1, IV) oxides via Mn3+. In oxidized soils containing MnO, ,flooding and the process of becoming reduced may produce
Mn3+,which is a powerful reducing agent. The trivalent Mn species may be
ephemeral intermediates in such processes at redox interfaces, such as in
the rhizosphere of plant roots or between soil water and groundwater. As
the metal analog of superoxide in its oxidizing/reducing power and as a
free radical, Mn3+is appropriately referred to as the supermanganese ion.
Because many Mn(II1, IV) oxides are nonstoichiometric and no compound with the exact composition of MnO, is known (Arndt, 1981),
predictions of their redox properties as a function of mineralogy or valence
REDOX CHEMISTRY OF SOILS
Figure 4. A pe-pH diagram for Mn3+, MnO,, and Mn,O,; as compared with reduction
values between pH 5 and 7 for Co, Cr, Se, As, V, and Pu.Activity for ionic species is lo-' M.
in heterogeneous soils may be hard to formulate. Despite the uncertainty
of thermodynamic predictions for the redox behavior of Mn, the chemistry
of this element is pertinent to a number of processes governing speciation
and valence state of trace elements and pollutants found in soils.
The pe-pH data indicate that oxides of Mn may oxidize Pu(1II) to
Pu(IV), V(II1) to V(V), As(II1) to As(V), Se(IV) to Se(VI), N(II1) to
N( V), and Cr( 111) to Cr( VI), because the pe for each of these couples falls
below that for Mn oxides (Fig. 4 and Table I). The oxidations of Pu(III),
As( 111), Se(IV), N( 111), and Cr( 111) all have been demonstrated to occur in
soils containing Mn oxides or by synthetic Mn oxides (Amacher and
Baker, 1982; Bartlett and James, 1979; Bartlett, 1981b; Blaylock and
James, 1992; Moore ef af.,1990).
The instability of Mn3+and its ability to dismutate, as do H202and OF,
mean that kinetic constraints may be particularly important in understanding the redox behavior of Mn in soils undergoing transitions between
anaerobic and aerobic conditions. The kinetic lability of these species is
poorly understood and new knowledge could contribute significantly to
predictions of bioavailability and toxicity of numerous plant nutrients and
pollutants in a range of types of soils from rice paddies and wetlands to
well-drained agricultural and forest soils.
RICHMOND J. BARTLETT AND BRUCE R. JAMES
Figure 5. A pe-pH diagram for Fe(II1) oxides and dihydroxy species, Few/Fe2+in the
presence or absence of five organic complexing Ligands. and the HCrO;/Cr(OH), redox
couple. Activity of ionic species is IO-‘ M.
D. IRON SPECIES
Predictions of the redox behavior of Fe( 11) and Fe( 111) species indicate
that it falls below most Mn oxides species (lower pe values and less free
energy released per equivalent upon reduction), but intermediate hydrolysis products, such as Fe(OH)l, theoretically can oxidize Cr(111) to Cr( VI)
at pH <4 (Fig. 5 and Table I). In addition, complexation of Fe2+ by
organic ligands lowers the pe values at which Fe3+is converted to Fe2+and
the redox couples are similar to those of Fe( 111) oxides in the pH range of 5
to 7. This phenomenon suggests that Fe2+ becomes a more powerful
reducing agent when complexed, and may explain the ability of Fe to act as
a cofactor in enzymes involved in redox processes, such as peroxidases and
superoxide dismutases. These enzymes reduce or dismutate H20, and Oy.
The application of such concepts to abiotic redox processes in soils remains
a key area for future research.
Reduced forms of C and S are normally viewed as reductants in soils,
either in chemical or biological processes. Thermodynamic predictions
REDOX CHEMISTRY OF SOILS
- - _- _- _
- _- - _- - _
Figure 6. A pe-pH diagram for S, C, and Se species. Ion activity and molecular concentrations are lo-' M and Pa is 0.032 kPa.
support this idea for carbohydrates produced in photosynthesis, methane
from methanogenesis, and hydrogen sulfide from reduction of SO, (Fig. 6
and Table I). The reduction reaction of 0- and pquinone suggest that these
compounds may be reduced at higher pe values than are C02and SO,. The
low position of these lines, however, coincides with the Mn02/Mn3+couple at pH 7, suggesting that Mn3+ may act as a reducing agent for certain
organic species in near-neutral soils. Coupling of reduction of the organic
with oxidation of Mn may result in formation of free radical species. This
is pertinent to understanding the formation and persistence of organic
matter in high-pH soils that may contain reactive forms of Mn oxides.
Reactions of H2S and H,Se are predicted to be similar with respect to
SO, and S e O , formation (Fig. 6). Sulfidization has been studied as a
mechanism for precipitation of Fe and other heavy metals in tidal marshes
and natural or constructed wetlands (Rabenhorst and James, 1992; Rabenhorst et al., 1992; Hines et al., 1989; Kittrick et al., 1982), and selenide
formation may result in analogous products in sulfidic soils (Masscheleyn
et al., 1991).
Although SeO, and SO, are similar chemically, the oxidation of SeO, to
SeO, is predicted to occur at higher pe values than is the oxidation of H2S
to SO, (Fig. 6 and Table I). Blaylock and James (1992) observed that Mn
oxides in soils or in pure form will oxidize SeO, to SeO,, as predicted by
thermodynamics (Fig. 4). They also observed that adding reducing, phe-
RICHMOND J. BARTLETT AND BRUCE R. JAMES
nolic acids, such gallic and ascorbic acids, actually enhanced this oxidation. They hypothesized that partial reduction of MnO, in soils converted
the Mn oxide into a Mn(II1) form that was a more powerful oxidant for
SeO, than was MnO,. Such a hypothesis is supported by the relative
oxidizing power of MnO,, MnOOH, and Mn,O,, where the latter two
oxides contain Mn( 111) (Fig. 3).
VII. MEASUREMENT OF OXIDATION- REDUCTION
STATUS OF SOILS
The most common method for quantifying electron activity of soils and
natural waters is to measure the potential difference between a Pt indicator
electrode and a calomel or Ag/AgCl reference electrode, both connected to
a voltmeter of pH meter (Rowell, 1981 ;Bricker, 1982). In this method, the
Pt electrode is presumed to be inert and to not react chemically while
coming into equilibrium with electroactive species in soil solution and on
soil colloids. Recent advances in evaluations of the reliability of this potentiometric measurement have generally resulted in it being considered
unreliable for accurate assessments of redox status of soils, especially aerobic ones (Bartlett, 1981a). Other methods that employ analyses of soil
solution analytes indicative of redox status, along with thermodynamic
half-reactions, as discussed above, may prove more reliable for calculating
pe ranges for aerobic and anaerobic soil systems.
AND USEOF PLATINUM
Platinum and suitable reference electrodes are relatively easy and inexpensive to construct (Mueller et al., 1985; Farrell er al., 1991), but the
measurement technique may significantly alter measured voltages; several
aspects of electrode use and misuse with respect to the reliability of
recorded voltages for natural systems have been described (Bartlett, 198la;
Bricker, 1982; Matia et al., 1991).
B. INADEQUACIESOF PLATINUM
Assessing “electron activity” in soils relates strictly to an evaluation of
the ability of the electron to be transferred, to do thermodynamic work,
and not to its concentration in soil solution, as can be defined for H+.
Because of the nature of the electron and its differences from the H+, a
REDOX CHEMISTRY OF SOILS
number of caveats must be described and recognized when evaluating Pt
1. Dissolved Oxygen Status
A stable potential can be obtained for a Pt reference electrode pair
immersed in an oxygenated soil suspension, but it is unreliable as a measure of dissolved oxygen status (Bricker, 1982; Stumm and Morgan, 198l ).
The Pt surface may react with 0, to form ROH, which develops a potential with elemental Pt with a pe of 9.6 at pH 7 (Table I). In addition, the
measurement may not be that of the 0 2 - H 2 0 couple, but may be responding to 0, reduction intermediates, such as H,02and 0;-(Bricker,
1982). In addition, predicted pe values are relatively insensitive to changes
in dissolved O2 between 0.21 and 0.0021 atm (Table 11), the range of 0,
partial pressures in which aerobic respiration occurs (Russell, 1973). For
these reasons, Pt electrode potentials cannot be used reliably as a measure
of redox status for aerobic soils, but empirical values for pe (EMpe) may be
obtained for comparison purposes (Bartlett, 198la). Although more faith is
placed in measurements of soil pH, it also should be considered an empirical measurement because of uncertainty about the form of the hydrogen
ion in colloidal environments and about the behavior of the glass electrode
in such systems. For these reasons, both pe and pH measured with electrodes in soils may be very uncertain for accurate descriptions of the redox
status of soil environments containing air-filled pores.
2. Irreversibility of Redox Couples
Many of the important redox processes involving C, H, N, 0, and S (the
“light” elements, relative to the “heavy” metals) are irreversible in the
thermodynamic sense, and nonelectroactive gases and molecules may be
consumed or formed. As a result, potentials generated by redox couples for
these elements are difficult to obtain and interpret using a Pt electrode. In
addition, many of these reactions do not reach true chemical equilibrium,
and activities measured in soil solution may be kinetically constrained (Liu
and Narasimhan, 1989). Because the redox status of soils is often set by
“microbial potentials,” consuming or producing compounds or ions containing one or more of these elements may render Pt electrode measurements inaccurate.
3. Mixed Potentials
The goal of relating measured redox potentials to species and valence
states of various elements in soils requires that a given, singular redox