Tải bản đầy đủ - 0 (trang)
A. Effect of Organic Ligands on Iron Oxide Dissolution Rates

A. Effect of Organic Ligands on Iron Oxide Dissolution Rates

Tải bản đầy đủ - 0trang

S. M. KRAEMER ET AL.



26



dissolution in the presence of each ligand Ln, and f(DG) is a function of the

solution saturation state (expressed as Gibbs free energy of the dissolution

reaction). The f(DG) is 1 at strong undersaturation, decreases as equilibrium

is approached, and becomes 0 at equilibrium (Kraemer and Hering, 1997).

Therefore, at strong undersaturation Eq. (11) reduces to the more familiar

rate law (Furrer and Stumm, 1986).

Rnet ẳ SkLn ẵLn ads



12ị



Based on Eq. (12), a linear correlation between adsorbed ligand concentrations and dissolution rates is diagnostic for a ligand‐controlled dissolution mechanism. For example, goethite dissolution rates are linearly related

to adsorbed DMA concentrations at pH 8, resulting in a Langmuir type

relationship between soluble DMA concentrations and dissolution rates

(Reichard et al., 2005). Rate coeYcients for steady state iron oxide dissolution in the presence of natural and synthetic organic ligands including

microbial and plant siderophores have been published.

As seen in Table III, the steady state rate coeYcients for goethite dissolution in the presence of the phytosiderophore DMA have the same order of

magnitude as those of oxalate and EDTA and tend to increase with increasing

pH. A common explanation for the influence of pH on dissolution rate

coeYcients are changes in the surface speciation of the ligand (Nowack

and Sigg, 1997) or deprotonation of adjacent ligating groups at the mineral

surface (Kraemer et al., 1998). The eVect of increasing rate coeYcients is

compensated by decreasing adsorbed DMA concentrations with increasing

pH as discussed earlier, resulting in maximum dissolution rates at pH 6.

Observed net dissolution rates are lower than those predicted by Eq. (12) due

to readsorption of Fe–DMA complexes, particularly at pH < 7 as discussed

earlier (Reichard et al., 2005). In several studies, dissolution of iron oxides

in the presence of MA over a fixed period of time has been observed.

Table III

Rate CoeYcients of Ligand‐Promoted Dissolution of Goethite in the Presence of Various

Siderophore and Nonsiderophore Ligands

Ligand



pH



kL [hÀ1]



References



DFO‐B

DFO‐D1

Oxalate

EDTA

DMA

DMA

DMA

DMA



6.5

6.5

6

5.3

5

6

7

8



0.05

0.06

0.001

0.0016

0.0008

0.003

0.003

0.004



Cocozza et al., 2002; Kraemer et al., 1999

Cocozza et al., 2002; Kraemer et al., 1999

Reichard et al., submitted

Nowack and Sigg, 1997

Reichard et al., 2005

Reichard et al., 2005

Reichard et al., 2005

Reichard et al., 2005



PHYTOSIDEROPHORE‐PROMOTED PLANT IRON ACQUISITION



27



Inoue et al. (1993) chose reaction times so that ferrihydrite reached solubility

equilibrium (4 h). Other iron oxide phases may or may not have reached

equilibrium in the same time. Therefore, reported dissolved concentrations

(after 4 h) may be controlled by equilibrium thermodynamics, others by

dissolution kinetics and cannot easily be compared. However, observed

soluble iron concentrations after 4 h reaction time in the presence of MA

decreased in the order ferrihydrite >> lepidocrocite ! goethite ¼ hematite.

Observed goethite dissolution rate coeYcients in the presence of microbial

siderophores (DFO‐B and DFO‐B) are higher than those of DMA‐promoted

dissolution. However, observed surface concentrations of the microbial siderophores are low, even at high soluble concentrations above 100 mM resulting in net goethite dissolution rates in the same range as DMA‐promoted

dissolution (Cheah et al., 2003; Cocozza et al., 2002; Kraemer et al., 1999).

For a detailed discussion of ligand‐controlled dissolution in the presence of

microbial siderophores, see Kraemer (2004).



1.



Inhibitory Effects of Inorganic Ligands and Humic Substances

on Ligand‐Controlled Iron Oxide Dissolution



The model of ligand‐controlled dissolution presented here assumes that

parallel dissolution mechanisms are independent in the sense that a rate

coeYcient kLn for ligand‐controlled dissolution in the presence of the

corresponding ligand Ln is independent of the presence of other adsorbed

or soluble ligands. However, the adsorbed ligand concentrations (Ln)ads are

clearly not independent of other ligands competing for adsorption. For

example, phosphate has a high aYnity for adsorption at iron oxide surfaces,

but it does not promote iron oxide dissolution. Due to its high aYnity for

adsorption, it can competitively displace other ligands (that promote iron

oxide dissolution) in a ligand‐exchange reaction. Therefore, phosphate can

have an inhibitory eVect on ligand‐controlled dissolution (Bondietti et al.,

1993; Jones et al., 1996). Strong inhibition of goethite and hematite dissolution in the presence of MA was observed at high concentration (0.1 M) of

phosphate and sulfate (Hiradate and Inoue, 1998a,b). Under the same conditions, lepidocrocite and ferrihydrite dissolution was partially inhibited by

phosphate, while sulfate inhibited the readsorption of Fe–MA complexes,

which led to an increase of soluble Fe–MA concentrations below pH 7 relative

to similar experiments in the absence of sulfate (Hiradate and Inoue, 1998b).

Silicate inhibited the dissolution of ferrihydrite above pH 8 and at silicate

concentrations above 0.2 mM (Watanabe and Matsumoto, 1994).

Humic substances can have similar inhibitory eVects on oxide dissolution.

Ochs et al. (1993) found that humic substances increased dissolution rates of

g‐Al2O3 at pH 3, but inhibited dissolution at higher pH. They attributed the



28



S. M. KRAEMER ET AL.



inhibitory eVect to the formation of unreactive polynuclear surface complexes. Also, the concomitant decrease or even reversal (Tipping and Cooke,

1982) of the usually positive surface charge of iron oxides (in the acidic

to slightly alkaline pH range) tends to decrease adsorption of negatively

charged ligands.



2. Synergistic Effects of Organic Acids on

Phytosiderophore‐Controlled Iron Oxide Dissolution

An obvious prerequisite for dissolution is that the solution is undersaturated with respect to the dissolving mineral. In a nonequilibrium system,

organic ligands can decrease the solution saturation state by iron complexation in solution. The eVect of the solution saturation state (i.e., the Gibbs

free energy change of the dissolution reaction DG) is quantitatively expressed

in the rate law Eq. (11) as f(DG). At a given soluble iron concentration,

the eVect of a ligand on DG depends on the ligand concentration and on

its aYnity for iron. For example, Jones et al. (1996) estimated that citrate

does not have a suYcient aYnity for iron to increase the solubility of

ferrihydrite in the rhizosphere at pH ! 7 to a level that supports plant

growth. In contrast, microbial and plant siderophores have a strong eVect

on the solution saturation state even at small free concentrations due to

their extremely high aYnity for iron. Therefore, siderophores can indirectly

facilitate dissolution by increasing the iron solubility without direct participation in mineral surface reactions (Cheah et al., 2003; Reichard et al.,

submitted, b). For example, Cheah et al. (2003) observed no dissolution

of goethite due to its low solubility at pH 5 at oxalate concentrations

<100 mM. However, oxalate did accelerate goethite dissolution in the presence of the microbial siderophore DFO‐B. This resulted in a synergistic eVect,

where the presence of DFO‐B facilitated a ligand (i.e., oxalate)‐controlled

dissolution mechanism. Similar synergistic eVects have been observed in the

presence of malonate, fumarate, and succinate in combination with DFO‐B

(Reichard et al., submitted, b). It is intriguing to speculate that the coexudation of phytosiderophores and organic acids by iron‐limited strategy II

plants (Fan et al., 1997, 2001) may be related to their synergistic eVect on

iron oxide dissolution.

Phytosiderophore are exuded diurnally for a few hours with maximum

release rates 6 h after sunrise (Marschner, 1995). This strategy minimizes

bacterial phytosiderophore degradation (Crowley and Gries, 1994), but it

also reduces the time during which iron can be mobilized, impairing iron

acquisition from crystalline iron oxide minerals that dissolve via slow

surface‐controlled dissolution mechanisms. However, the diurnal phytosiderophores release creates a strong nonsteady state with respect to the



PHYTOSIDEROPHORE‐PROMOTED PLANT IRON ACQUISITION



29



solution saturation state and surface chemistry of iron‐bearing mineral

phases. These nonsteady state conditions trigger fast iron oxide dissolution

reactions if other ligands, such as oxalate, citrate, or malate, are also present

(Reichard et al., 2005, submitted, b). Such organic ligands are common in the

rhizosphere, as discussed earlier, and are coexuded by iron‐limited grasses

(Fan et al., 1997). The rapid iron release in response to transient phytosiderophore exudation partly compensates for decreased iron release rates during

the rest of the day thus satisfying nutritional demands (Reichard et al.,

submitted, a). A mechanism of nonsteady state iron oxide dissolution was

proposed, where a labilizing ligand (e.g., oxalate or malonate) adsorbs at the

mineral surface by a ligand‐exchange mechanism. The adsorbed ligands

catalyze a surface reaction leading to the formation of kinetically labile iron

surface centers in pseudoequilibrium with a low concentration of soluble iron

complexes. On release of phytosiderophores, the labilized iron can be

eYciently dissolved (Reichard et al., submitted, b).



B.



THERMODYNAMICS AND KINETICS OF LIGAND‐EXCHANGE REACTIONS

WITH PHYTOSIDEROPHORES AS RECEIVING LIGANDS



It has been observed that variations of the uptake rate of iron from

organic iron complexes by strategy II plants follow the same diurnal pattern

as release rates of phytosiderophores by the plant roots and that external

supply of phytosiderophores to nutrient solutions increase the uptake rates

(Cesco et al., 2002; Yehuda et al., 1996). From these observations, it has been

concluded that iron acquisition from organic complexes by strategy II plants

involves a ligand‐exchange reaction that results in the release of iron from the

organic ligands and the subsequent formation of iron–phytosiderophore

complexes.



1.



Ligand Exchange Equilibria



As discussed earlier, it has been established that strategy II plants acquire

iron from organic complexes by ligand‐exchange reactions with phytosiderophores as receiving ligands. Figure 8A shows equilibrium complexation of

1 mM Fe(III) by DMA in the presence of competing ligands as a function of

total DMA concentrations. EDTA at micromolar concentrations can only

compete with DMA in the absence of calcium. In the presence of calcium,

EDTA inhibits precipitation of ferrihydrite (at low DMA concentrations) but

DMA eYciently scavenges iron from EDTA. High concentrations of citrate

(1 mM) compete only at DMA concentrations below 10 mM and in the

absence of calcium. Micromolar concentrations of the microbial siderophore



S. M. KRAEMER ET AL.



30



DFO‐B compete strongly with high concentrations of DMA for iron complexation independent of the presence or absence of calcium. The much higher

aYnity of DFO‐B for iron compared to phytosiderophores was experimentally verified in a ligand‐exchange experiment where 100 mM epi‐HMA did

not bind significant concentrations of iron in the presence of 200 mM Fe

DFOB at pH 6 (Hoărdt et al., 2000). This is consistent with equilibrium

calculations using the set of equilibrium constants listed in Tables IV and V.

At millimolar phytosiderophore concentrations that may be realized in the

apoplast or very close to the root surface, ligand exchange may lead to the

formation of low concentrations of iron–phytosiderophore complexes in a

range that may partially satisfy iron requirements. However, slow ligand

exchange rates may further reduce the eYciency of this reaction for iron

acquisition as discussed later.

Due to large variations in the stability of Fe–siderophore complexes, it

is diYcult to draw generalizations on their availability for plant uptake.

For example, DMA is able to eYciently sequester iron from the iron–

rhodotorulic acid (Fe2RA3) complex (Fig. 8B). Rhodotorulic acid is a tetradentate siderophore forming binuclear complexes with somewhat lower

stability compared to the other siderophore complexes shown here. Zhang

et al. (1991a) found higher iron uptake via an apoplastic pathway in the

presence of the rhodotorulic acid complex compared to the DFO‐B complex.

However, they did not report if rhodotorulic acid was present in (molar)

excess over iron. Due to the stoichiometry of the dominant complex, the



Table IV

pKa of Organic and Inorganic Ligands, Corrected to Zero Ionic Strength

with the Davies Equation T ¼ 298.15 K

L



pKa1



pKa2



pKa3



pKa4



pKa5



DMA

MA

Epi‐HMA

DFO‐Bd

Ferrichromed

Coprogend

Rhodotorulic acidd

Citrated

EDTAd

Carbonated



2.13a

2.17c

2.13a

8.32

8.33

7.85

8.71

3.13

1.5

6.35



2.74a

2.76c

2.74a

9.06

9.44

9.3

9.88

4.76

2.22

10.33



3.4b

3.45b

3.45b

9.73

10.49

9.82



6.4

3.13





8.69b

8.38b

7.54b

11.06









6.27





10.66b

10.51b

10.28b











10.95





a



von Wire´n et al. (2000).

Murakami et al. (1989).

c

Sugiura et al. (1981).

d

Martell et al. (2001).

b



Tài liệu bạn tìm kiếm đã sẵn sàng tải về

A. Effect of Organic Ligands on Iron Oxide Dissolution Rates

Tải bản đầy đủ ngay(0 tr)

×