International Journal of Geosciences, 2012, 3, 314-320
http://dx.doi.org/10.4236/ijg.2012.32033 Published Online May 2012 (http://www.SciRP.org/journal/ijg)
Enrichment of Phosphate on Ferrous Iron Phases during
Bio-Reduction of Ferrihydrite*
Qingman Li1, Xingxiang Wang2#, Dan Kan1, Rebecca Bartlett3,
Gilles Pinay3,4, Yu Ding1, Wei Ma5
1Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan, China
2Institute of Soil Science, Chinese Academy of Sciences, Nanjing, China
3School of Geography, Earth and Environmental Sciences, University of Birmingham, Birmingham, UK
4Ecobio-Osur, CNRS, University of Rennes 1, Rennes, France
5Clinical Department, School of Medicine, Northwest University for Nationalities, Lanzhou, China
Received November 20, 2011; revised February 8, 2012; accepted March 9, 2012
ABSTRACT
The reduction of less stable ferric hydroxides and formation of ferrous phases is critical for the fate of phosphorus in
anaerobic soils and sediments. The interaction between ferrous iron and phosphate was investigated experimentally
during the reduction of synthetic ferrihydrite with natural organic materials as carbon source. Ferrihydrite was readily
reduced by dissimilatory iron reducing bacteria (DIRB) with between 52% and 73% Fe(III) converted to Fe(II) after 31
days, higher than without DIRB. Formation of ferrous phases was linearly coupled to almost complete removal of both
aqueous and exchangeable phosphate. Simple model calculations based on the incubation data suggested ferrous phases
bound phosphate with a molar ratio of Fe(II):P between 1.14 - 2.25 or a capacity of 246 - 485 mg·P·g−1 Fe(II). XRD
analysis indicated that the ratio of Fe(II): P was responsible for the precipitation of vivianite (Fe3(PO4)2·8H2O), a domi-
nant Fe(II) phosphate mineral in incubation systems. When the ratio of Fe(II):P was more than 1.5, the precipitation of
Fe(II) phosphate was soundly crystallized to vivianite. Thus, reduction of ferric iron provides a mechanism for the fur-
ther removal of available phosphate via the production of ferrous phases, with anaerobic soils and sediments potentially
exhibiting a higher capacity to bind phosphate than some aerobic systems.
Keywords: Phosphate; Iron Reduction; Ferrihydrite; Ferrous Iron; Vivianite
1. Introduction
Phosphorus is essential for life and is increasingly the
limiting nutrient in some ecosystems, as nitrogen pollu-
tion becomes widespread [1]. In soils and freshwater se-
diments, the fate and mobility of phosphorus may be
controlled by iron geochemistry, through sorption and
desorption, co-precipitation and dissolution with both
ferrous (Fe(II)) and ferric (Fe(III)) minerals. Whilst sorp-
tion of phosphorous to ferric phases such as ferrihydrite
(Fe5O6(OH)9) tends to occur under aerobic conditions,
ferrous iron phases are among the most important com-
ponents to react with phosphate in anaerobic environ-
ments. During the development of anaerobic soil and se-
diment environments, the concentration of aqueous pho-
sphate may increase, due to the reductive dissolution of
ferri-phosphate phases [2]. However, it has been shown
that the capacity of soils and sediments to bind phosphate
is substantially increased under anaerobic conditions, ge-
nerally attributed to the formation of ferrous phases [3-7].
The disagreement regarding iron phases binding phos-
phate in complex environments, no doubt makes it im-
portant to understand how ferrous iron phases react with
phosphorus.
The production of ferrous iron phases in soils and se-
diments is complex, with both chemical and biological
controls. Reduction of ferric (hydro) oxides may be cat-
alysed by dissimilatory iron reducing bacteria (DIRB) via
electron transfer during heterotrophic metabolism [8,9].
The production of ferrous iron and new ferrous minerals
however, is dependent on desorption from the surface of
the original ferric iron mineral [10-14]. Ferric (hydro)
oxide reduction then is dependent on redox, organic car-
bon supply, and the amount and reactivity of ferric pha-
ses; the new ferrous phases that form will further be de-
pendent on the surrounding chemistry, including the pre-
sence or absence of phosphate [15,16]. It can be pre-
dicted that products from ferric (hydro) oxide reduction
in the environment should be a mixture of ferrous phases
*This work was supported by the National Natural Science Foundation
of China (No. 40730528 and 40873061).
#Corresponding authors.
C
opyright © 2012 SciRes. IJG

Q. M. LI ET AL. 315
with different proportion.
Ferrous iron may act to decrease the concentration of
aqueous phosphate by sorption. Under non-sulfidogenic
anaerobic conditions, vivianite (Fe3(PO4)2·8H2O, Ksp 10 -
35.8), a stable ferrous mineral may be formed with pho-
sphate incorporated in 1.5:1 molar ratio of Fe(II):P [17,
18]. However, field observations have shown that the
reduction of soils and sediments is coupled with raised
aqueous phosphate, suggesting vivianite formation may
be subject to other controls. Other ferrous minerals may
also bind phosphate, including siderite (FeCO3) [19], and
mixed valence iron phases such as green rust

x
2
x2 2
AyHO





x
II III
6x x12
Fe Fe OH



 and magnetite
(Fe3O4) [20].
The potential for ferrous iron phases produced under
reducing conditions to bind phosphate is poorly defined.
This work describes laboratory experiments that simulate
the anaerobic environment in order to study the fate of
phosphate during microbial reduction of ferrihydrite, and
creates a simple model of phosphate binding. Ferrihy-
drite was used as a model of less stable ferric hydroxides
to act as electron acceptor for DIRB under controlled
conditions. Ferrihydrite readily interacts with phosphate
either by surface adsorption or by co-precipitation with
reported sorption maxima for phosphate of 0.6 - 2.5
mmol·g−1 [19,21-23], greatly larger than those of other
crystalline ferric oxides [24-27]. Importantly, ferrihydrite
is considered ubiquitous and highly reactive in soil and
sediment environments, and may be preferentially re-
duced by bacteria to form a range of ferrous phases [28].
2. Materials and Methods
2.1. Preparation of Materials
Ferrihydrite (Fe5O6(OH)9) was prepared by titrating 0.5
M NaOH into a FeCl3 solution until a final pH appro-
aching 7.0, followed by dialysis as described by Atkin-
son et al. [29]. Analysis of transmission electron micros-
copy (TEM) and X-ray diffraction confirmed the precipi-
tate as ferrihydrite, which was kept in suspension until
use in the experiment.
Nutrient solutions used for the enrichment of DIRB
were prepared according to an adaptation of Lovley and
Philips [30]. Two nutrient solutions were prepared: 1)
composed of (g·L−1): CaCl2·2H2O, 0.1; KCl, 0.1; NH4Cl,
1.5; NaH2PO4·H2O, 0.6; NaCl, 0.1; MgCl2·6H2O, 0.1;
MgSO4·7H2O, 0.0937; MnSO4·H2O, 0.0043; (NH4)6Mo7O24,
0.0008; yeast juice, 0.05; NaOOCCH3·3 H2O, 4.48; and 2)
composed of (g·L−1): CaCl2·2H2O, 0.1; KCl, 0.1; NH4Cl,
1.5; NaH2PO4·H2O, 0.6; suspended organic material (see
below).
Three natural organic materials were chosen as carbon
source for DIRB: Lemna trisulca (L. trisulca); Microcys-
tis flos-aquae (M. flos-aquae); and Vallisneria natans (V.
natans). These were sampled from Yuehu Lake and Di-
anchi Lake, China, rinsed with deionised water and dried
and ground to fine a powder (<50 m). A stock of sus-
pended organic material was prepared by adding 3 g of
dry powder to nutrient solution 2 until all organic par-
ticles had sunk to the bottle bottom, and diluted to 1 L.
The composition of the suspended organic material is
given in Table 1 (carbon content is equal for all species).
Fresh anaerobic sediment (sampled from the Yuehu
Lake, Wuhan, China) was transferred into a brown bottle,
diluted with culture solution 1) and anaerobically incu-
bated in the dark at 28˚C ± 0.5˚C for 30 d with occa-
sional stirring. A sub-sample of the anaerobic sediment
suspension was used for enrichment of DIRB after sepa-
ration by centrifugation. 50 mL of the supernatant was
diluted with culture solution 1) containing ferrihydrite (to
a final concentration of ~15 mmol·L−1 Fe(III)), and fur-
ther incubated in the dark at 28˚C ± 0.5˚C for 30 d. This
purification was repeated 16 times in order to generate
the DIRB suspension. Before its use in experiments, the
DIRB suspension was adjusted to neutral pH using
NaOH, and sparged with N2 for 1 h.
To protect the DIRB suspension from infection, all
equipment and solutions used in its preparation were
sterilized at 120˚C for 30 min and handled using aseptic
technique. Using this approach, the final suspension was
enriched in DIRB, but was not a pure culture and would
also have contained other microbial groups.
2.2. Experiment Design
Before the experiment began, the suspensions of organic
material were mixed with ferrihydrite to obtain a culture
medium, and left for 24 h to allow the sorptive reaction
of ferrihydrite with phosphate to reach equilibrium. To
inoculate the experiments, 5 mL DIRB suspension was
added to 1 L culture medium and incubated in the dark at
28.0˚C ± 0.5˚C. In order to avoid overpressure in the
incubation bottles (from CO2 production), a fine plastic
tube was attached to the bottle mouth and fed into oxygen-
free water. Each experiment was conducted in triplicate.
The procedure for experimental controls was the same;
the suspensions of organic material were mixed with
ferrihydrite as above, but not inoculated with DIRB.
Incubations were sampled at regular intervals via nee-
Table 1. Composition of organic material used in culture so-
lutions.
P N Fe Ca
2+
Organic C source mg·g−1 DW
L. trisulca 4.96 8.44 0.21 0.70
V. natans 3.23 5.3 0.27 0.46
M. flos-aquae 1.25 14.8 0.38 0.67

Q. M. LI ET AL. 316
dle and syringe, and analysed for pH, phosphate fractions
(aqueous, exchangeable, incorporated) and iron fractions
(Fe(II), Fe(III)).
2.3. Analyses
Phosphate was operationally fractionated into 3 phases:
aqueous phosphate (4), exchangeable phosphate
(loosely sorbed P) and total bound phosphate (total P in
solid phases). Aqueous phosphate was determined after
filtration (0.45 m membrane). Exchangeable phosphate
(Pex) was extracted in 0.5 M KCl for 30 min and filtered
(0.45 m membrane); the phosphate in the filtrate is re-
garded as the sum of exchangeable and aqueous phos-
phates. Total bound phosphate (TPB) was obtained through
subtracting the sum of aqueous phosphates from total
phosphorus. Total phosphate was determined after H2SO4
+ H2O2 digestion.
3
PO 
3
PO 
3
PO
Filtered phosphate samples were determined by the
molybdenum blue method with ascorbic acid as reducing
agent.
Total iron was determined after hot HCl extraction by
spectrometry (722, Shanghai Analytical Instrument) us-
ing 10% hydroxylamine HCl as reductant and 2% 2,2’-
dipyridine as chromogenic reagent. Fe(II) was deter-
mined by elimination of the reduction step, and addition
of ammonium fluoride to prevent Fe(III) interference.
At the end of the experiment, the solid phases were
characterized by X-ray diffraction (XRD) analysis of N2-
dried samples using a Philips PW1050 X-ray diffracto-
meter (using CuK radiation, with scans taken from 4˚ to
64˚ at a scan rate of 2˚/min).
The pH values of incubation suspensions were meas-
ured using glass electrode with a calomel electrode as re-
ference electrode (pHS-3C meter, Xinkong Medical Ap-
paratus Co., Ltd., Jiangyan, China).
Data analyses (average, standard deviation) and statis-
tical analyses (correlation coefficients) were conducted
in this study. The regression diagnostics were checked by
F-test, and a p < 0.05 was considered to indicate signifi-
cance.
3. Results
Chemical data from the incubation experiments are shown
in Figure 1. The production of Fe(II) in all the incubation
experiments showed that conditions remained anaerobic
and reducing throughout.
In the DIRB experiments, Fe(II) production was rela-
tively rapid in the first 20 days (increase of >480 mg·g−1
Fe). The constant rate of Fe(II) production was consistent
among all live experiments (26.12 mg·g−1·Fe·d−1 for L.
trisculca, 23.36 mg·g−1·Fe·d−1 for M. flos-aquae and
22.54 mg·g−1·Fe·d−1 for V. natans), before approaching
equilib- rium between 20 and 31 days. Fe(II) production
was de- pendent on carbon source with the order of L.
trisculca > M. flos-aquae > V. natans equivalent to 73%,
53% and 52% of total Fe respectively. In the control ex-
periments Fe(II) production was much less (<250
mg·g−1·Fe), respectively equivalent to 31%, 21% and
14% of total Fe, with the order relative to carbon source
the same as for the live experiments. Reduction of Fe(III)
(ferrihydrite) to Fe(II) was clearly enhanced by the pres-
ence of live DIRB during the experiments, although some
chemical reduction may also have taken place (as indi-
cated by controls) [31].
Aqueous phosphate (4) and exchangeable phos-
phate (Pex) decreased rapidly over the first 20 days (from
~70 mg·L−1·4

and ~80 mg·g−1·Fe Pex), and were
almost completely removed by 31 days in all DIRB incu-
bations. The rate of removal was remarkably similar be-
tween experiments, with some difference in initial P con-
centrations dependent on the P content of the original
organic material (Table 1). There was some removal of
aqueous and exchangeable P in the control experiments,
although this was approximately half that of the live ex-
periments (removal of <20 mg·L−1·4 and <40
mg·g−1·Fe Pex) and did not approach zero. As both P frac-
tions decreased during the incubations, it was clear that
aqueous P was not being removed by sorption (and trans-
formed to Pex), but was bound as mineral P. This was true
for both DIRB and control experiments, albeit at a lesser
rate in the absence of DIRB.
3
PO 
Aqueous Fe tended to increase over incubation time. In
the DIRB experiments, aqueous Fe slowly increased in
the initial 10 day incubation, but abruptly rose after that,
with the highest concentration in a range between 20 - 40
mg·L−1. In the control experiment, aqueous Fe tended to
increase continuously, with the final concentration largely
dependent on the type of organic materials. To combine
the decrease of aqueous P at the later period of experi-
ments, the raise of aqueous Fe should be a result of aque-
ous P consumption in the DIRB experiments.
4. Discussion
The inverse relationship between ferrous iron production
and aqueous and exchangeable P removal in the incuba-
tion experiments suggested a single control on Fe and P
geochemistry. Reduction of ferric iron was coincident
with the production of ferrous iron phases and precipita-
tion of phosphorous. This was enhanced in the presence
of DIRB. While the proportion of Fe(III) was high (at the
start of the experiment), <75% of total phosphorous was
either aqueous or exchangeable, but under reducing con-
ditions, the production of Fe(II) induced precipitation of
nearly all aqueous and exchangeable phosphorous, pre-
sumably as a ferrous iron phase. This was contrary to
some literatures [4], which suggests that ferric iron re-

Q. M. LI ET AL. 317
Figure 1. Production of ferrous iron (Fe(II)) and removal of aqueous phosphate (3
4
PO

and exchangeable phosphorous (Pex)
during anaerobic incubation of ferrihydrite in the presence of phosphorous and organic material. Initial total Fe(III) in in-
cubation systems was 680 mg·L−1; initial phosphate in incubation systems was 172 mg·L−1 (L. trisulca), 140 mg·L−1 (M. flos-
aquae) and 157 mg·L−1 (V. natans); the data are mean values of triplicate incubations.
duction should be coupled to the release of sorbed phos-
phorous and increase of aqueous phosphate. Indeed, it
was important to note that this closed experimental sys-
tem might behave differently to the natural environment;
however, it was clear that the production of ferrous
phases might be more important in P geochemistry than
previously realised.
The reduction of ferric (hydro) oxides to ferrous iron is
thought to be limited by the accumulation of ferrous iron
at the surface of the original (hydro) oxides [14]. In these
incubation experiments, 52% - 73% Fe(III) was reduced
to Fe(II), suggesting the minimal influence of accumu-
lated Fe(II) at the surface of ferrihydrite. It was possible
that the presence of organic ligands might have aided
complexation and removal of ferrous iron from the min-
eral surface [31]. However, it was further likely that the
removal of ferrous iron by precipitation with aqueous
and exchangeable P provided a mechanism by which iron
reduction could continue unchecked. Indeed, the reduc-
tion of Fe(III) (and accumulation of Fe(II)) effectively
ended once exchangeable and aqueous P had been com-
pletely removed (change in rate of Fe(II) accumulation
and P removal after 20 days, Figure 1).
The reduction of ferrihydrite to ferrous iron in the
presence of aqueous and exchangeable phosphorous ap-
peared to have promoted the precipitation of ferrous iron
phosphate. To describe the mode of ferrous phases to
bind phosphate, we proposed a simple model. This re-
quired the following assumptions: 1) the interaction of
ferrous iron phases with phosphate was independent of
the presence of ferrihydrite, and vice versa; 2) the distri-
bution of phosphate in both ferrihydrite and ferrous iron
phases was homogeneous; 3) there was sufficient phos-
phate to interact with iron phases. Then, the following

Q. M. LI ET AL. 318
relationships are given:
PBFe(III)  mFe(III) (1)
PBFe(II)  mFe(II) (2)
where PBFe(III) and PBFe(II) represent phosphate bound to
ferrihydrite and ferrous iron phases respectively, and
mFe(III) and mFe(II) represent the quantities of ferrihydrite
and ferrous iron phases in the incubations.
The total phosphate bound (TPB) is expressed as:
TPB = PBFe(III) + PBFe(II) (3)
We operationally define that:
PBFe(III) = KFe(III)*mFe(III) (4)
PBFe(II) = KFe(II)*mFe(II) (5)
in which the constants of KFe(III) and KFe(II) refer to the
unit capacity of ferrihydrite and ferrous iron phases to
enrich solid phase phosphate respectively. Substituting
Equations (4) and (5) into (3) gives:
TPB = KFe(III)*mFe(III) + KFe(II)*mFe(II) (6)
If the total iron in a incubation system is given as m and
the produced Fe(II) is mFe(II), Fe(III) (mFe(III)) can be ob-
tained by subtracting:
mFe(III) = m − mFe(II) (7)
Combining Equation (6) with Equation (7) gives:
TPB = (KFe(II) − KFe(III))* mFe(II) + KFe(III)*m (8)
Equation (8) showed a linear relationship between TPB
and mFe(II). The constants of KFe(II) and KFe(III) could then
be calculated through the slope and intercept of a linear
curve.
As the simple model prediction, our experimental data
described a linear dependence of bound P on Fe(II) in all
incubations (Figure 2, r2 ~ 0.95 given in Table 2) and
showed that iron reduction could enrich P in the solid
phase. Table 2 showed the capacity of ferrihydrite and
ferrous iron to bind P, the latter having a greater potential
Figure 2. Dependence of phosphorous binding (TPB) on production of Fe(II) (mFe (II )) during incubations. Data in (a) and (b) is
respectively from DIRB and control experiments.
Table 2. Model derived P-binding capacities for Fe(II) and Fe(III) solid phases in incubations.
Organic C source a Fitted graph mFe(II) v TPB R2
bK Fe(III)
mg·g −1
cFe(III):P
bK Fe(II)
mg·g −1
cFe(II):P
DIRB TPB = 0.187 mFe(II) + 87.15 0.95* 87.2 6.35 274 2.02
L. trisulca
CK TPB = 0.303 mFe(II) + 77.09 0.98* 77.1 7.18 380 1.46
DIRB TPB = 0.183 mFe(II) + 90.84 0.95* 90.8 6.10 280 2.00
V. natans
CK TPB = 0.408 mFe(II) + 77.70 0.97* 77.7 7.12 485 1.14
DIRB TPB = 0.163 mFe(II) +85.27 0.94* 85.3 6.49 257 2.15
M. flos-aquae
CK TPB = 0.157 mFe(II)+ 89.20 0.93* 89.2 6.21 246 2.25
aIncubation data (Figure 2). TPB = total bound phosphorous; mFe(II) = produced Fe(II); bModel derived constant (Equation (6)); cMolar ratio; *Significance level
P < 0.05.

Q. M. LI ET AL. 319
Figure 3. XRD trace of products of ferrihydrite reduction in the presence of phosphate. (a) and (c) samples taken from V.
natans DIRB and control experiments after 31 days; (b) sample taken from L. trisulca control experiment after 31 days.
(77.1 - 90.8 mg·P·g−1 Fe(III) compared to 246 - 485
mg·P·g−1 Fe(II)). The binding capacity of ferrihydrite in
these experiments was higher than those reported in other
works [19,23], most likely due to dialyzation during fer-
rihydrite preparation and precipitation of Ca-P on the
surface of ferrihydrity, yet this was still exceeded by the
binding capacity for ferrous iron. This difference was
probably owed to the mechanism by which P is bound to
the ferrihydrite and ferrous iron. According to the calcu-
lated constants KFe(III) and KFe(II), the Fe:P molar ratio was
6.10 - 7.18 for ferrihydrite and 1.14 - 2.25 for ferrous
phases. This supported the assumption that ferrihydrite
binds phosphorous by sorption or co-precipitation [21,
22].
The molar ratio for Fe(II):P of ~2, was slightly higher
than the stoichiometry for the ferrous iron phosphate
mineral, vivianite (Fe3(PO4)28H2O); XRD analysis con-
firmed vivianite was perfectly crystallized (Figure 3(a)).
The slightly higher Fe(II):P value for the incubations
suggested not all produced Fe(II) formed vivianite. This
might be due to sorption of Fe(II) on ferrihydrite surfaces,
or the formation of mixed phase intermediates during
iron reduction, such as magnetite. The XRD trace also
suggested small amounts of other ferrous iron minerals
(not identified) were present, despite a predominance of
vivianite. For Fe(II):P approximate to vivianite, product
analysis indicated a part of vivianite began forming (Fi-
gure 3(b)), but crystalline degree was obviously lower
than Fe(II):P of ~2, suggesting the formation of vivianite
needs sufficient Fe(II). This inference could be supported
by the XRD trace of the ~1 Fe(II):P products, in which
vivianite was not detected (Figure 3(c)). The result also
suggested that not all Fe(II) interacting with phosphate
produced vivianite even in the presence of high concen-
tration P.
5. Conclusion
This work showed the potential for anaerobic soils and
sediments to exhibit a higher capacity to bind phosphate
than aerobic soils and sediments because of the produc-
tion of ferrous phases, with vivianite a dominant product
of iron reduction in the production of sufficient Fe(II). It
was likely that in the natural environment, local geo-
chemistry would further influence the stability of ferrous
iron-phosphate phases. Production of organic ligands or
sulphides in some systems for instance might lead to
Fe(II) complexation or iron sulphide precipitation, thus
limiting ferrous iron-phosphate, and potentially increas-
ing aqueous phosphate. Further work was needed to de-
termine the importance of ferrous iron-phosphate in Fe
and P cycles, but it was clear that interactions between
Fe(II) and P have a powerful influence on the in situ re-
gulation of phosphorous bio-availability.
REFERENCES
[1] N. Gruber and J. N. Galloway, “An Earth-System Per-
spective of the Global Nitrogen Cycle,” Nature, Vol. 45,
No. 451, 2008, pp. 293-296. doi:10.1038/nature06592
[2] W. H. Patrick Jr., S. Gotoh and B. G. Williams, “Strengite
Dissolution in Flooded Soils and Sediments,” Science,
Vol. 179, No. 4073, 1973, pp. 564-565.
doi:10.1126/science.179.4073.564
[3] F. N. Ponnamperuma, “The Chemistry of Submerged
Soils,” Advanced in Agronology, Vol. 24, 1972, pp. 29-96.
doi:10.1016/S0065-2113(08)60633-1
[4] W. H. Patrick Jr. and R. A. Khalid, “Phosphate Release
and Sorption by Soils and Sediments: Effect of Aerobic
and Anaerobic Conditions,” Science, Vol. 186, No. 4158,
1974, pp. 53-55. doi:10.1126/science.186.4158.53
[5] G. J. D. Kirk, “The Biogeochemistry of Submerged
Soils,” John Wiley & Sons, London, 2004.
doi:10.1002/047086303X
[6] A. J. Miller, E. A. G. Schuur and O. A. Chadwick, “Re-
dox Control of Phosphorus Pools in Hawaiian Montane
Forest Soils,” Geoderma, Vol. 102, No. 3-4, 2001, pp.
219-237. doi:10.1016/S0016-7061(01)00016-7
[7] N. Chacon, W. L. Silver, E. A. Dubinsky and D. F. Cu-
sack, “Iron Reduction and Soil Phosphorus Solubilization
in Humid Tropical Forests Soils: The Roles of Labile
Carbon Pools and an Electron Shuttle Compound,” Bio-
geochemistry, Vol. 78, No. 1, 2006, pp. 67-84.

Q. M. LI ET AL. 320
doi:10.1007/s10533-005-2343-3
[8] R. A. Royer, W. D. Burgos, A. S. Fisher, B. H. Jeon, R. F.
Unz and B. A. Dempsey, “Enhancement of Hematite Bio-
reduction by Natural Organic Matter,” Environmental
Science & Technology, Vol. 36, No. 13, 2002, pp. 2897-
2904. doi:10.1021/es015735y
[9] T. S. Peretyazhko, J. M. Zachara, D. W. Kennedy, J. K.
Fredrickson, B. W. Arey, J. P. McKinley, C. M. Wang, A.
C. Dohnalkova and Y. Xia, “Ferrous Phosphate Surface
Precipitates Resulting from the Reduction of Intragrain
6-Line Ferrihydrite by Shewanella Oneidensis MR-1,”
Geochimica et Cosmochimica Acta, Vol. 74, No. 13, 2010,
pp. 3751-3767. doi:10.1016/j.gca.2010.04.008
[10] D. R. Lovley, E. J. P. Phillips and D. J. Lonergan, “En-
zymatic versus Nonenzymatic Mechanisms for Fe(III)
Reduction in Aquatic Sediments,” Environmental Science
& Technology, Vol. 25, No. 6, 1991, pp. 1062-1067.
doi:10.1021/es00018a007
[11] R. A. Royer, W. D. Burgos, A. S. Fisher, R. F. Unz and B.
A. Dempsey, “Enhancement of Biological Reduction of
Hematite by Electron Shuttling and Fe(II) Complexa-
tion,” Environmental Science & Technology, Vol. 36, No.
9, 2002, pp. 1939-1946. doi:10.1021/es011139s
[12] R. A. Doong and B. Schink, “Cysteine-Mediated Reduc-
tive Dissolution of Poorly Crystalline Iron(III) Oxides by
Geobacter Sulfurreducens,” Environmental Science &
Technology, Vol. 36, No. 13, 2002, pp. 2939-2945.
doi:10.1021/es0102235
[13] J. .K Fredrickson, S. Kota, R. K. Kukkadapu, C. Liu and J.
M. Zachara, “Influence of Electron Donor/Acceptor Con-
centrations on Hydrous Ferric Oxide (HFO) Bioreduc-
tion,” Biodegradation, Vol. 14, No. 2, 2003, pp. 91-103.
doi:10.1023/A:1024001207574
[14] R. A. Royer, B. A. Dempsey, B. H. Jeon and W. D. Bur-
gos, “Inhibition of Biological Reductive Dissolution of
Hematite by Ferrous Iron,” Environmental Science &
Technology, Vol. 38, No. 1, 2004, pp. 187-193.
doi:10.1021/es026466u
[15] E. J. O’Loughlin, “Effects of Electron Transfer Mediators
on the Bioreduction of Lepidocrocite (Gamma-FeOOH)
by Shewanella Putrefaciens CN32,” Environmental Sci-
ence & Technology, Vol. 42, No. 18, 2008, pp. 6876-
6882. doi:10.1021/es800686d
[16] S. Rakshit, M. Uchimiya and G. Sposito, “Iron(III) Bio-
reduction in Soil in the Presence of Added Humic Sub-
stances,” Soil Science Society of America Journal, Vol.
73, No. 1, 2009, pp. 65-71. doi:10.2136/sssaj2007.0418
[17] P. P. Hearn and D. L. E. C. Parkhurst, “Authigenic Vivi-
anite in Potomac River Sediments: Control by Ferric
Oxyhydroxides,” Journal of Sediment Research, Vol. 53,
No. 1, 1983, pp. 165-177.
[18] R. L. Frost, W. Martens, P. A. Williams and J. T. Klo-
progge, “Raman and Infrared Spectroscopic Study of the
Vivianite-Group Phosphates Vivianite, Baricite and Bo-
bierrite,” Mineralogical Magazine, Vol. 66, No. 6, 2002,
pp. 1063-1073. doi:10.1180/0026461026660077
[19] R. K. Kukkadapu, J. M. Zachara, J. K. Fredrickson and D.
W. Kennedy, “Biotransformation of Two-Line Silica-Fer-
rihydrite by a Dissimilatory Fe(III)-Reducing Bacterium:
Formation of Carbonate Green Rust in the Presence of
Phosphate,” Geochimica Cosmochimica Acta, Vol. 68,
No. 13, 2004, pp. 2799-2814.
doi:10.1016/j.gca.2003.12.024
[20] N. Parmar, Y. A. Gorby, T. J. Beveridge and F. G. Ferris,
“Formation of Green Rust and Immobilization of Nickel
in Response to Bacterial Reduction of Hydrous Ferric
Oxide,” Geomicrobiology of Journal, Vol. 18, No. 4,
2001, pp. 375-385. doi:10.1080/014904501753210549
[21] U. Schwertmann, D. G. Schulze and E. Murad, “Identifi-
cation of Ferrihydrite in Soils by Dissolution Kinetics,
Differential X-Ray Diffraction and Mössbauer Spectros-
copy,” Soil Science Society of America Journal, Vol. 46,
No. 4, 1982, pp. 869-875.
doi:10.2136/sssaj1982.03615995004600040040x
[22] J. .L Jambor and J. E. Dutrizac, “Occurrence and Consti-
tution of Natural and Synthetic Ferrihydrite, a Wide-
spread Iron Oxyhydroxide,” Chemical Reviews, Vol. 98,
No. 7, 1998, pp. 2549-2585. doi:10.1021/cr970105t
[23] J. C. Ryden, J. R. McLaughlin and J. K. Syers, “Mecha-
nisms of Phosphate Sorption by Soils and Hydrous Ferric
Oxide Gels,” Journal of Soil Science, Vol. 28, No. 1,
1977, pp. 72-92. doi:10.1111/j.1365-2389.1977.tb02297.x
[24] S. Kuo and E. G. Lotse, “Kinetics of Phosphate Adsorp-
tion and Desorption by Hematite and Gibbsite,” Soil Sci-
ence, Vol. 116, No. 6, 1973, pp. 400-406.
doi:10.1097/00010694-197312000-00002
[25] R. L. Parfitt and R. J. Atkinson, “Phosphate Adsorption
on Goethite (α-FeOOH),” Nature, Vol. 264, No. 5588,
1976, pp. 740-742. doi:10.1038/264740a0
[26] J. Torrent, V. Barrón and U. Schwertmann, “Phosphate
Adsorption and Desorption by Goethites Differing in
Crystal Morphology,” Soil Science Society of America
Journal, Vol. 549, No. 4, 1990, pp. 1007-1012.
doi:10.2136/sssaj1990.03615995005400040012x
[27] L. Madrid and P. D. Arambarri, “Adsorption of Phos-
phate by Two Iron Oxides in Relation to Their Porosity,”
Journal of Soil Science, Vol. 36, No. 4, 1985, pp. 523-
530. doi:10.1111/j.1365-2389.1985.tb00355.x
[28] W. R. Fischer, “Standard Potentials (E0) of Iron(III) Ox-
ides under Reducing Conditions,” Zeitschrift fuer Pflan-
zenernaehrung und Bodenkunde, Vol. 150, No. 5, 2007,
pp. 286-289. doi:10.1002/jpln.19871500504
[29] R. J. Atkinson, A. M. Posner and J. P. Quirk, “Crystal
Nucleation in Fe(III) Solutions and Hydroxide Gels,”
Journal of Inorganic and Nuclear Chemistry, Vol. 30, No.
9, 1968, pp. 2371-238.
doi:10.1016/0022-1902(68)80247-7
[30] D. R. Lovley and E. J. P. Phillips, “Novel Mode of Mi-
crobial Energy Metabolism: Organic Carbon Oxidation
Coupled to Dissimilatory Reduction of Iron or Manga-
nese,” Applied and Environment Microbiology, Vol. 54,
No. 6, 1988, pp. 1472-1480.
[31] W. Zhang, Q. M. Li, X. X. Wang, Y. Ding and J. Sun,
“Reducing Organic Substances from Anaerobic Decom-
position of Hydrophytes,” Biogeochemistry, Vol. 94, No.
1, 2009, pp. 1-11. doi:10.1007/s10533-009-9295-y