Uranium (VI) Removal from aqueous solutions using elemental iron

Abstract

The objective of this study was the evaluation of elemental iron for the remediation of groundwater contaminated by uranium (VI). The study was focused on the effect of different conditions on the removal of uranium by elemental iron, as well as on the determination of the predominant reaction mechanisms. The conditions applied where either atmospheric or geochemical (presence of excess CO32-) while the presence of SO42- ions was also studied. The results indicated that under atmospheric conditions, almost 98% of U(VI) is removed when the pH is in the neutral and alkaline regions, while the acidic pH reduces the removal to about 82%. Under geochemical conditions the uranium (VI) removal is decreased while in the presence of SO42- ions it does not affected at all. The mechanism of uranium (VI) uptake by elemental iron was attributed to the precipitation of uranyl hydroxides along with the sorption of uranyl ions on the iron hydroxides.

Full text

URANIUM (VI) REMOVAL FROM AQUEOUS SOLUTIONS USING
ELEMENTAL IRON
A. Krestou, A. Xenidis, D. Panias
National Technical University of Athens- School of Mining and Metallurgical Engineering,
9, Heroon Politechniou Street, Zografos Campus, 15780 Athens, Greece
Abstract
The objective of this study was the evaluation of elemental iron for the remediation of
groundwater contaminated by uranium (VI). The study was focused on the effect of different
conditions on the removal of uranium by elemental iron, as well as on the determination of the
predominant reaction mechanisms. The conditions applied where either atmospheric or
geochemical (presence of excess CO32-) while the presence of SO42- ions was also studied. The
results indicated that under atmospheric conditions, almost 98% of U(VI) is removed when the pH
is in the neutral and alkaline regions, while the acidic pH reduces the removal to about 82%.
Under geochemical conditions the uranium (VI) removal is decreased while in the presence of
SO42- ions it does not affected at all. The mechanism of uranium (VI) uptake by elemental iron
was attributed to the precipitation of uranyl hydroxides along with the sorption of uranyl ions on
the iron hydroxides.

Introduction
Uranium is a toxic metal and one of the heaviest naturally occurring elements. However, the wide
range of utilization of uranium, i.e. as a fuel to generate electrical power, as a shielding material,
in the production of high energy X-rays as well as in the nuclear weapon industry, has resulted in
the production of high amounts of uranium all over the world, which pose hazards to the
environment and the human health. Since aqueous media are the most conceivable transport media
for radioactive wastes, uranium is usually found in aqueous streams and actually in the hexavalent
form, UO22+. Only in Europe, 14 countries face the problem of groundwater contamination caused
by uranium, which represents a particularly serious danger where drinking water resources might
be affected (PEREBAR website, [1],[2]). As a result, a large amount of work has been performed
over the last decades in order to eliminate or prevent the uranium contamination of waters and
soils.
Elemental iron, which is usually referred to as zero-valent iron (ZVI), is a leading material for the
use in cleaning up groundwater contaminants [3], because of its low cost as well as because its
secondary and ternary reaction products (Fe(II) and Fe(III), respectively), pose no harm to the
environment. It is believed [4] that when used for remediation, ZVI serves both as catalyst and as
an electron source in which elemental iron immobilizes soluble trace element contaminants by
surface reduction. However, the actual mechanism that elemental iron employs for the removal of
contaminants is still under discussion, with the reduction, precipitation and adsorption of the
contaminants on the corrosion products of ZVI being the most possible reaction paths [5].
The present study is focused on the effectiveness of ZVI in removing uranium (VI) from aqueous
solutions under variable conditions (solution pH, retention time, solid/ liquid ratio, presence of
CO32- and SO42-).
Material Characterization
The zero-valent iron sample used for the undertaking of the experiments was supplied by Gotthart
Maier, German and is characterized as cast iron grit. Chemical analysis showed that the zero-
valent iron used consists mainly of Fe (92.03% w/w), C (3.31% w/w), Si (2.04% w/w) and other
elements such as Mn, Al, S, Ni, Cr and P in concentrations lower than 1%w/w while the water
content is equal to 0.4% w/w. Mineralogical investigation carried out by X-ray diffraction and
Scanning electron microscopy (accelerated voltage: 25KV) revealed the existence of metallic iron,
graphite and iron oxide (about 10%). Thermogravimetric analysis confirmed the absence of any
degradable mineral phases. The paste pH of the iron sample was equal to 5.1-5.3, while the
specific surface area of the material was measured equal to 0.0482m2/g. The grain size of the
delivered material, as described in the Technical Data Sheet provided by the supplier, was equal to
0.35-1.20 mm. The apparent density of the sample was 2.7-2.9 g/cm3.
Experimental Procedure and Methods
The uranium solution was prepared by the Nuclear Research Centre Demokritos, Greece, by
dissolution of an appropriate quantity of uranyl nitrate hexahydrate (UO2(NO3)2·6H2O) provided
by Fluka, in a diluted solution of nitric acid so as the concentration of the resulted solution to be

equal to 1000ppb uranium. The pH of the UO2(NO3)2·6H2O solution was approximately equal to
1.2.
The uranium attenuation by zero-valent iron was studied under variable pH conditions, contact
time and pulp densities. The effect of the presence of CO32- and SO42- ions on uranium uptake by
ZVI was also studied by preparing uranium solutions containing these ions in specific
concentrations. The pH of each solution was adjusted to the desired value using a concentrated
solution of NaOH. The appropriate amount of the material was added in the U(VI) solution and
adequate stirring in an agitated glass reactor for the time required from the experimental
conditions, followed. Then the solution was filtrated and the filtrate was analyzed for uranium
with the Arsenazo-III spectrophotometric method ([1], [2]) on a HITACHI U-1100 UV-VIS
spectrophotometer. Measurements of the pH were carried out just after the addition of the reactive
material in the solution as well as at the end of each experiment.
Results
Effect of the pH Variation
The effectiveness of ZVI in uranium (VI) attenuation was studied in the pH-region between 3 and
11, a solid/ liquid ratio equal to 2g/l and a 2 hours retention time. Figure 1 presents the uranium
removal attained by elemental iron at different pH values. As shown in Figure 1, the results
indicated that zero-valent iron can effectively remove uranium (VI) from aqueous solutions,
especially near to the neutral and in the alkaline region. In the acidic region (3<pH<5) the uranium
attenuation by zero-valent iron is steadily increased from ca 82% to about 97%. A further pH raise
does not alter the uranium attenuation considerably, which remains almost constant at 97.74%.
50
55
60
65
70
75
80
85
90
95
100
234567891
Uranyl nitrate solution pH
%U (VI) removal
01112
Figure 1: Variation on the U(VI) removal in relation to the pH of the uranyl nitrate solution
([U]= 1000ppb, solid/liquid ratio=2g/l, retention time =2h, ambient temperature)
Effect of the Retention Time
The kinetic experiments were carried out for a solution pH equal to 7 and for a solid /liquid ratio
equal to 2g/l. The results of this set of experiments are illustrated in Figure 2. The results showed
that the maximum uranium (VI) attenuation corresponding to pH=7, that is ca 97%, is attained in
only one hour and then reaches a plateau.

0
10
20
30
40
50
60
70
80
90
100
0 200 400 600 800 1000
Retention time(min)
% U (VI) re moval
1200 1400 1600
Figure 2: Effect of the contact time of ZVI with the uranium nitrate solution on the uranium
(VI) attenuation ([U]= 1000ppb, pH = 7, solid/liquid ratio: 2g/l, ambient temperature)
Effect of the Solid/Liquid Ratio
In order to reveal the optimum conditions for the uranium attenuation by ZVI, it was considered
necessary to investigate the effect of different ZVI concentrations on the uranium attenuation at
pH equal to 7. The results obtained from these experiments (Figure 3) showed that the uranium
attenuation was almost independent on the available surface of the reactive material.
70
75
80
85
90
95
100
00,511,522,533,544,5
solid/ liquid ratio (g/l)
% U( VI) remova l
average
Figure 3: Effect of the solid/liquid ratio on the removal of uranium (VI)
([U]= 1000ppb, pH =7, retention time=2h, ambient temperature)
Effect of the Presence of CO3 2- and SO4 2- Ions
The effect of the presence of carbonates (400ppm) and sulfates (200ppm) on the removal of
uranium by ZVI was decided to be studied since these ions are usually present in groundwater.
The solution pH was adjusted to 7, while the solid/liquid ratio was 2g/l and the retention time 2
hours. The results of these experiments are presented in Figure 4.

0
10
20
30
40
50
60
70
80
90
100
% U(VI) removal
Pure uranium
(
VI
)
400pp m CO3 2-
200pp m SO4 2-
Figure 4: Effect of the presence of carbonates and sulfates on the uranium (VI) attenuation
by zero-valent iron ([U]= 1000ppb, pH= 7, ambient temperature, time: 2hrs solid/liquid:2g/l)
As shown in Figure 4, the presence of carbonates resulted in a 35% reduction of the uranium (VI)
attenuation while the presence of sulfates did not influence the uranium (VI) removal at all.
Discussion
In the results of the aforementioned experiments, ZVI was proven to be pretty effective in
removing aqueous uranium (VI) under various conditions. In order to interpret the experimental
results and reveal the mechanism employed by ZVI, it was considered necessary to examine the
uranium (VI) solution chemistry.
Uranium Speciation at Equilibrium with Atmospheric Carbonates
The uranium (VI) speciation under atmospheric conditions and I=0.01M is shown in Figure 5. For
the construction of this uranium (VI) speciation diagram all the known mononuclear and
polynuclear uranium species and uranium (VI) carbonato complexes were considered together
with the corresponding thermodynamic data [6], shown in Table I.
Table I: Formation constants of U(VI) complex ions at 25°C and 0.01M ionic strength
Reactions Formation
Constant
UO22+ + H2O(l) ' UO2OH+ +H+ 10-5.29
UO22+ + 2H2O(l) ' UO2(OH)2+2H+ 10-10.49
UO22+ + 3H2O(l) ' UO2(OH)3-+3H+ 10-19.20
UO22+ + 4H2O(l) ' UO2(OH)42-+4H+ 10-32.82
2UO22+ + H2O(l) ' (UO2)2OH3+ +H+ 10-2.70
2UO22+ + 2H2O(l) ' (UO2)2(OH)22+ +2H+ 10-5.71
3UO22+ + 4H2O(l) ' (UO2)3(OH)42+ +4H+ 10-12.08
3UO22+ + 5H2O(l) ' (UO2)3(OH)5+ +5H+ 10-15.82
3UO22+ + 7H2O(l) ' (UO2)3(OH)7- +7H+ 10-31.18
3UO22+ + 8H2O(l) ' (UO2)3(OH)82- +8H+ 10-37.65

3UO22+ + 10H2O(l) ' (UO2)3(OH)104- +10H+ 10-63.01
4UO22+ + 2H2O(l) ' (UO2)4(OH)26+ +2H+ 10-0.27
4UO22+ + 6H2O(l) ' (UO2)4(OH)62+ +6H+ 10-17.23
4UO22+ + 7H2O(l) ' (UO2)4(OH)7+ +7H+ 10-22.26
5UO22+ + 3H2O(l) ' (UO2)5(OH)37++3H+ 10-18.76
UO22+ + CO32- ' UO2CO3
(
a
q)
109.32
UO22+ + 2CO32- ' UO2(CO3)22- 1016.58
UO22+ + 3CO32- ' UO2(CO3)34- 1021.60
3UO22+ + 6CO32- ' (UO2)3(CO3)66- 1054.00
2UO22++CO2
(
a
q)
+4H2O
(
l
)
' (UO2)2CO3(OH)3- +5H+ 10-19.10
11UO22++6CO2
(
a
q)
+18H2O
(
l
)
' (UO2)11(CO3)6(OH)122- +24H+ 10-73.22
3UO22++CO2
(g)
+4H2O
(
l
)
' (UO2)3O(OH)2HCO3+ +5H+ 10-17.77
Figure 5: Uranium (VI) speciation at equilibrium with atmospheric carbonates at 25°C
(I=0.01M, [U]=10-5M)
0
10
20
30
40
50
60
70
80
90
100
2345678910
pH
% species
UO22+
UO2OH
+
UO2(OH)2o
(UO2)2(OH)22+
(UO2)3(OH)5+
(UO2)2CO3(OH)3-
UO2(CO3)22-
UO2(CO3)34-
As shown in Figure 5, very acidic solutions favor the formation of the uranyl ion, UO22+. In mildly
acidic solutions the prevailing uranium (VI) species is the mononuclear UO2(OH)20, while
polynuclear positively charged uranium species ((UO2)2(OH)22+ and (UO2)3(OH)5+) are also
present in much lower concentrations. The formation of uranium carbonato complexes is
preferential for neutral and alkaline solutions (Figure 5). These species are negatively charged and
as the solution pH increases the charge is moved towards to more negative values.
Uranium Speciation in the Presence of Excess Carbonates
The uranium (VI) speciation when excess carbonates are present at I=0.01M is shown in Figure 6.
For the construction of this uranium (VI) speciation diagram all the known mononuclear and
polynuclear uranium species and uranium (VI) carbonato complexes were considered together
with the corresponding thermodynamic data [6] shown in Table I.

0
10
20
30
40
50
60
70
80
90
100
2345678910111213
pH
% species
UO22+ UO2(CO3)34-
(UO2)3(OH)104-
(UO2)3(OH)82-
UO2(OH)42-
UO2(OH)3-
UO2(CO3)22-
UO2CO3o
(UO2)2CO3(OH)3-
UO2OH+
(UO2)2(OH)22+
Figure 6: Uranium speciation in the presence of 400mg/L carbonates at 25oC and I=0.31M
(total uranium concentration 5⋅10-6M).
As shown in Figure 6, the prevailing uranium (VI) species in the high acidic region is again the
positively charged uranyl species. However, carbonato complexes prevail in the rest pH region,
with their charge varying from 0 at pH =5 to 4- for pH higher than 7.5.
Mechanism Employed by ZVI
Iron corrosion in an aqueous environment is largely depended on the conditions prevailing in the
aqueous medium and is initiated according to the following reactions:
Equation 5
Equation 4
Fe0 +1/2 O2 +H2O' Fe2++ 2OH-, in oxic environments
Fe0 +2H2O'Fe2+ + H2+ 2OH-, in anoxic environments
The above reactions are always accompanied by the formation of precipitates ([4], [7]-[9]) such as
amorphous iron oxides and oxyhydroxides, goethite, aragonite, calcite, siderite etc, as described in
the following reactions:
Equation 7
Equation 6
Equation 8
2Fe2+ + 1/2O2 +3H2O ' 2FeOOH(s) + 4H+
2Fe2+ + 1/2O2 +2H2O ' Fe2O3 + 4H+
Fe2+ + CO32-' FeCO3 (s)
The precipitates shown in Equations 6, 7 and 8 present large surface area and thus can contribute
to immobilization of certain contaminants through sorption or co-precipitation.
In the case of uranium (VI) contaminated water there are three possible mechanisms that are
responsible for the uranium (VI) attenuation.

Reduction of hexavalent uranium to tetravalent uranium: The first possible mechanism implies the
reduction of the U(VI) to U(IV) on the surface of zero-valent iron and precipitation of the
sparingly soluble uraninite, according to the following reaction ([3], [8]-[10]):
Equation 9
Fe0 + UO2 2+ ' Fe2+ +UO2 (s)
In one report [8] it is mentioned that strongly reducing conditions must be attained in order the
above reactions to proceed. In another work [10], it has been reported that the above reaction is
difficult to proceed in the presence of CO2, which is a typical impurity in groundwater, due to the
shift of the boundary line between U(IV) and U(VI) towards lower Eh at high pH. Additionally, it
has been reported [3] that even if the above reaction takes place on the surface of zero valent iron,
the formed UO2(s) acts as a nucleation site for the initiation of continuous precipitation of the
amorphous U(VI) hydroxide (schoepite).
Precipitation of U(VI) as an amorphous uranium hydroxide (schoepite): The continuous corrosion
of zero-valent iron in the solution according to equations (4), (5), results in the rise of the local pH
around the zero-valent iron particles. Therefore, as the local pH in the vicinity of the Fe0 surface is
higher than the pH in the bulk of the solution, uranium (VI) hydroxide precipitation can take place
only near the surface of the zero-valent iron ([3], [4]). Therefore, uranium can be removed from
the solution through a non reductive precipitation mechanism.
This mechanism is affected by the presence of CO2 in the solution. Carbonates form very stable
complexes with uranyl ions, especially in the alkaline region, that do not form strong binds with
the iron oxides that are present on the surface of zero-valent iron. Thus, carbonates can inhibit the
growth of the amorphous uranyl hydroxides onto the Fe0 surface.
Uranyl adsorption onto iron oxides : The corrosion of iron in aqueous environments is always
accompanied by the rapid hydrolysis of ferric ions resulting in the formation of hydrous ferric
oxide or amorphous ferric hydroxide or even amorphous iron oxyhydroxide. These solid materials
can control the U(VI) migration in aqueous systems through sorption reactions. A literature review
indicates that the iron oxides and oxyhydroxides have a strong sorptive capacity for dissolved
uranyl species [5, 11] and are common accessory minerals or coating materials found in oxidized
soils and sediments where uranyl sorption reactions are more likely to be significant [12]. Two
types of sorption sites are formed on the surface of iron oxides and oxyhydroxides: The high-
affinity (Type 1) cation binding sites which are the less abundant surface sites and the low-
affinity cation binding sites (Type 2). The surface complexation constants of UO22+ for Type 1 and
Type 2 sites are 105.2 and 102.8, respectively [11]. All the above mentioned iron oxide materials
strongly adsorb dissolved uranyl species in the pH region 5 – 6. Amorphous ferric oxyhydroxide
exhibits the greatest adsorption capacity while well crystallized hematite has the least one. The
presence of calcium and magnesium cations does not significantly affect uranyl adsorption, while
in the presence of dissolved carbonates the sorption reactions are severely inhibited [12].
The reductive mechanism of uranium(VI) removal is very difficult to take place, since strongly
reducing conditions must be attained in order this mechanism to proceed. Provided that strongly
reducing conditions were not applied in this work, the other two potential mechanisms seem to be
operative. Especially, the U(VI) removal through sorption reactions onto the iron oxides seems to
be the most important one. The experimental results show that the U(VI) removal is completed at
pH higher than 5 (Fig.1) and is strongly affected by the presence of excess amount of dissolved
carbonates (Fig.4), which are the most important characteristics of this mechanism. Uranyl

removal reactions equilibrate rapidly and reach equilibrium within few hours (Fig.2), which is
more consistent with an adsorption mechanism rather than a precipitation one. Moreover, a small
amount of uranium(VI) (about 7%) can be removed from elemental iron loaded with uranium (VI)
when iron is treated with acidified and alkalized water, as shown in Figure 7, indicating that the
majority of uranium(VI) adsorbed cannot be easily desorbed from the loaded ZVI.
Dissolution of uranium in neutral and alkaline solutions
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
0 72 144 216 288 360 432 504 576 648 720 792 864 936
hours
% dissolved uranium
pH=5.5
pH=13.39
Figure 7: Dissolution of uranium from loaded ZVI in acid and alkaline solutions
Although in oxic environments and under low temperatures sorption is generally a more operative
mechanism than uranium (VI) hydroxide precipitation, the experimental data of this work cannot
exclude the latter mechanism. The speciation diagram of uranium (VI) at equilibrium with
atmospheric CO2 (Fig.5) shows that in alkaline solutions the majority of dissolved U(VI) occurs in
the form of carbonato complexes, which generally increase the uranyl hydroxide solubility.
Therefore, under these conditions, a decrease on U(VI) removal was expected, which, however,
was not observed from the experimental data. Thus, we can conclude that in alkaline region the
sorption mechanism is the only applicable mechanism. In the pH region from 5 to 7, which
comprises the core region of uranyl hydroxide precipitation, there are no data confirming or
rejecting the precipitation mechanism.
Conclusions
Elemental iron showed very high aqueous uranium (VI) removal yields under a wide range of
conditions. The only obstacle to the uranium (VI) removal was the presence of excess carbonates,
which lowered the uranium uptake by ZVI.
Three are the possible pathways responsible for the uranium (VI) attenuation from water and
groundwater using zero-valent iron. Although the reductive mechanism is feasible, it seems to be
difficult and needs careful control of the water pe conditions. On the other hand, the precipitation
of uranyl hydroxides along with the sorption of the uranyl ions on the iron oxides seems to be
more feasible under the conditions prevailing in the experiments performed in the present paper
with the majority of experimental results to confirm the sorption mechanism as the responsible
one for the U(VI) removal.

Acknowledgements
This research was funded by the European Union under the 5th Framework Programme,
PEREBAR project, Contract no. EVK1-CT-1999-00035 (see www.perebar.bam.de).
References
1. A. Krestou, A. Xenidis, D. Panias, “Mechanism of aqueous uranium (VI) uptake by natural
zeolitic tuff”, Minerals Engineering, 16, 2003, pp.1363-1370.
2. A. Krestou, A. Xenidis, D. Panias, “Mechanism of aqueous uranium (VI) uptake by
hydroxyapatite”, Minerals Engineering, 17, 2004, pp.373-381.
3. S.R. Qiu, C. Amrhein, M.L. Hunt, R. Pfeffer, B. Yakshinskiy, L. Zhang, T.E. Madey, J.A.
Yarmoff, “Characterization of Uranium Oxide Thin Films Grown from Solution onto Fe
surfaces”, Applied Surface Science 181 (2001), pp. 211-224.
4. S.R. Qiu, H.-F. Lai, M.J Robertson, M.L. Hunt, C. Amrhein, L.C. Giancarlo, G.W. Flynn,
J.A. Yarmoff, “Removal of Contaminants from Aqueous Solution by Reaction with Iron
Surfaces”, Langmuir 16 (2000), pp. 2230-2236.
5. B.Gu, L. Liang, M.J. Dickey, X. Yin, S. Dai, “Reductive Precipitation of Uranium (VI) by
Zero-Valent Iron”, Environmental Science Technology 32 (1998), pp. 3366-3373.
6. I. Grenthe, J. Fuger, et al, Chemical Thermodynamics Vol 1: Chemical Thermodynamics of
Uranium (Nuclear Energy Agency, North-Holland. 1992).
7. Y. Furukawa, J-W. Kim, J. Watkins, R.T Wilkin, “Formation of Ferrihydrite and Associated
Iron Corrosion Products in Permeable Reactive Barriers of Zero-Valent Iron”, Environmental
Science and Technology 36 (2000), pp. 5469-5475.
8. D.W. Blowes, C.J. Ptacek, S.G. Benner, Che W. T. McRae, T.A. Bennett and R.W. Puls,
“Treatment of inorganic contaminants using permeable reactive barriers”, Journal of
Contaminant Hydrology,45, Issues 1-2,(September 2000),pp. 123-137.
9. Y. Roh, S.Y. Lee, M.P. Elles, “Characterisation of Corrosion Products in Permeable Reactive
Barriers”, Environmental Geology 40, 1-2, (December 2000), pp. 184-194.
10. F-G. Simon, C. Segebade, M. Hedrich, “Behaviour of Uranium in Iron-bearing Permeable
Reactive Barriers: Investigation with 237U as a radioindicator”, The science of the Total
Environment (2002), pp. 135-139.
11. D. A. Dzombak, F. M. M. Morel, “ Surface Complaxation Modeling- Hydrous ferric oxide”,
(John Wiley & Sons Inc., 1990), pp.92, 305.
12. Ching-Kuo Daniel His, D. Langmuir “Adsorption of uranyl onto ferric oxyhydroxides:
Application of the surface complexation site-binding model”, Geochimica et Cosmochimica
Acta 49, (1985), pp.1931-1941.