Impact of MnO2 on the efficiency of metallic iron for the removal of dissolved CrVI, CuII, 
Mo
1 
2 
3 
4 
5 
6 
VI, SbV, UVI and ZnII
Noubactep C.*(a,c), Btatkeu K.B.D.(b), Tchatchueng J.B.(b)
(a) Angewandte Geologie, Universität Göttingen, Goldschmidtstraße 3, D - 37077 Göttingen, Germany; 
(b) ENSAI/University of Ngaoundere, BP 455 Ngaoundere, Cameroon;  
(c) Kultur und Nachhaltige Entwicklung CDD e.V., Postfach 1502, D - 37005 Göttingen, Germany. 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
(*) e-mail: cnoubac@gwdg.de; Tel. +49 551 39 3191, Fax: +49 551 399379 
Abstract   
The idea that manganese oxide (MnO2) sustains the reactivity of metallic iron (Fe0) is 
investigated in this study. A multi-elemental aqueous system containing CrVI, CuII, MoVI, SbV, 
UVI, and ZnII (each about 100 μM) was used as model solution. Non-disturbed batch 
experiments were performed at initial pH values 4.0 and 6.0 for one month. Three different 
systems were investigated: (i) MnO2 alone, (ii) “Fe0 + sand”, and (iii) “Fe0 + MnO2”. The 
experimental vessels contained either: (i) no material (blank), (ii) up to 9.0 g/L of MnO2, or 
(iii) 5 g/L Fe0 and 0 to 9.0 g/L MnO2 or sand. Results clearly revealed quantitative 
contaminant removal (> 70 %) confirming the suitability of Fe0 as a highly efficient reactive 
material for the removal of the 6 tested metallic ions over a pH range applicable to 
environmental waters. Results also corroborated the suitability of MnO2 to sustain the long-
term Fe0 reactivity. Further studies in dynamic systems (column studies) are necessary to fine-
tune the use of MnO2 in Fe0 filtration systems. 
Keywords: Drinking water, Heavy metals, Iron filters, Manganese oxides, Zerovalent iron. 
  
 1 
1 Introduction 23 
24 
25 
26 
27 
28 
29 
30 
31 
32 
33 
34 
35 
36 
37 
38 
39 
40 
41 
42 
43 
44 
45 
46 
The use of metallic iron (Fe0) for the treatment of contaminated groundwater is already a 
standard remediation approach [1-3]. This approach has the great advantage that many classes 
of contaminants are removed in a single filtration operation [4]. This observation has 
motivated the suggestion of Fe0 as reactive agent for decentralized safe drinking water 
provision [5-8] in general and for household filters in particular [9,10]. 
A packed Fe0 bed is regarded as a filtration system in which contaminants are removed during 
aqueous iron corrosion [5-8]. In such a system, the main mechanisms of contaminant removal 
are: (i) adsorption onto iron corrosion products (iron oxides and hydroxides), (ii) enmeshment 
with precipitating iron oxides/hydroxides (co-precipitation), and (iii) adsorptive size-
exclusion (straining). Adsorptive size-exclusion is improved during the service life of a filter 
by the in-situ formation of volumetric expansive corrosion products [11,12]. In fact, the 
volume of any iron corrosion product (e.g. FeO, Fe(OH)2, Fe(OH)3, Fe3O4, Fe2O3, FeOOH) is 
larger than that of the original metal (Fe0). The ratio between the volume of expansive 
corrosion product (Vox) and the volume of iron consumed in the corrosion process (VFe) is 
called ‘‘rust expansion coefficient’’ (η) and takes values between 2.08 and 6.40 [11]. 
η = Vox/VFe      (1) 
The idea of using Fe0 for household filters is not new [13-16]. However, conventional 
household Fe0 filters were found very efficient but not sustainable as they were clogged after 
some weeks of operation [17]. Recent theoretical studies [9,10,18] have re-vived research on 
household Fe0 filters. It was shown that reducing the proportion of Fe0 in a filter (admixture 
with a non expansive material) is the prerequisite for long-term efficiency. Furthermore, tools 
to sustain the long-term reactivity were discussed. These tools included the use of bimetallic 
systems (e.g. Fe0/Ni0, Fe0/Pd0) and the use of MnO2 admixture [19]. 
 2 
The use of MnO2 to sustain iron reactivity has already been discussed in the literature for the 
removal of methylene blue [20], clofibric acid [21], diclofenac [22,23], radium [24], and 
uranium [24,25]. The idea behind using MnO
47 
48 
49 
50 
51 
52 
53 
54 
55 
56 
57 
58 
59 
60 
61 
62 
63 
64 
65 
66 
67 
68 
69 
70 
71 
2 to sustain Fe0 reactivity is that, FeII species 
from Fe0 oxidation (Eq. 2) are used for the reductive dissolution of MnO2 (Eq. 3) [26,27]. For 
the sake of clarity, the oxidation of Fe0 by water (H+) and MnO2 are given by Eq. 4 and 5. 
Fe0(s) + 2 e- ⇔ FeII(aq)        (2) 
MnO2 (s) + 2 FeII(aq) + 2 H+(aq) ⇔ MnII(aq) + 2 FeIII(aq) + 2 H2O(l)  (3) 
Fe0(s) + 2 H+(aq) ⇔ FeII(aq) + 2 H2(g)      (4) 
Fe0(s) + MnO2(s) + 4 H+(aq) ⇔ MnII(aq) + FeII(aq) + 2 H2O(l)   (5) 
Chemical and electrochemical reactions likely to occur in a Fe0/MnO2/H2O systems are 
discussed in details in ref. [27]. For the present work, it is sufficient to consider that: (i) 
reaction 5 is more favourable than reaction 4 (MnO2 is a stronger oxidizing agent than H2O), 
and (ii) FeII consumption (via oxidation by MnO2 to FeIII) will result in an increase in Fe0 
oxidation after the Le Chatelier's principle. 
The chemical reaction between FeII and MnO2 necessarily takes place at the surface of MnO2. 
In other words, FeII species are transported away from the vicinity of the Fe0 surface and are 
not available to form the oxide-film. The formation of the oxide-film is responsible for Fe0 
passivation (reactivity loss) [1,4]. Sustained Fe0 reactivity can be intuitively coupled with 
long-term contaminant removal. It is essential to notice that FeIII species formed at the 
vicinity of MnO2 are not “free” to co-precipitate contaminants. Rather, they can only remove 
contaminant by adsorption or by improving adsorptive size-exclusion. This is the reason why 
a delay of contaminant removal has been reported in the presence of MnO2 in short term batch 
experiments [20,23,25]. 
Available results [20-25] univocally showed the capability of MnO2 to sustain contaminant 
removal. However, apart from Burghardt and Kassahun [24] who investigated the binary 
 3 
Ra/U system, all available data are related to single-contaminant systems. Such systems 
typically fail as environmental analogues. Therefore, there is a high need for multi-elemental 
studies for Fe
72 
73 
74 
75 
76 
77 
78 
79 
80 
81 
82 
83 
84 
85 
86 
87 
88 
89 
90 
91 
92 
93 
94 
95 
96 
0 remediation to advance the design of treatment infrastructures [28]. 
The objective of the current study is to investigate the suitability of MnO2 to sustain Fe0 
reactivity using a multi-elemental system as model solution. Tested contaminants are: CrVI, 
CuII, MoVI, SbV, UVI, and ZnII (each about 100 μM). These elements are known for their 
different affinity to Fe hydroxides and their different redox properties (Table 1) [29]. The 
experiments were performed under non-disturbed conditions at pH 4.0 and 6.0 for up to 60 
days in three different systems: “MnO2 alone”, Fe0 + MnO2”, Fe0 + sand”. The results are 
comparatively discussed. 
2 Materials and methods 
2.1 Chemicals 
All chemicals (K2Cr2O7, CuSO4.5H2O, Na2MoO4.2H2O, K(SbO)C4H4O6, UO2(CH3COO)2, 
ZnSO4) used in this study were of analytical grade. All solutions were prepared using a spring 
water. The used spring water was from the Lausebrunnen in Krebeck (administrative district 
of Göttingen). Spring water was used as proxy for natural water. Its average composition (in 
mg/L) was: Cl–: 9.4; NO3–: 9.5; SO42-: 70.9; HCO3-: 95.1; Na+: 8.4; K+: 1.0; Mg2+: 5.7; Ca2+: 
110.1; and pH 7.8. pH adjustment to values of 4.0 and 6.0 was performed with diluted NaOH 
and HNO3 solutions. These initial pH values were selected to uncover the pH value of natural 
waters [29].  
Table 1 summarizes some characteristics of the six tested metals which are important in 
discussing their removal from the aqueous solution [30]. The chemicals were weighed to yield 
an initial concentration of 0.10 mM (100 μM) corresponding to concentrations varying 
between 5.2 and 23.8 mg/L (Tab. 1). The operational initial concentration was determined 
from the so-called blank experiment (72 to 99 μM). Deriving initial concentration from the 
 4 
blank experiments enabled the consideration of all possible factors affecting the decrease of 
metal concentration. These factors include adsorption onto the walls of the essay tubes, 
common ion effect and precipitation. The initial concentration (100 μM) was selected to ease 
discussion on the molar basis. The resulting weight concentrations (Tab. 1) uncover the 
concentration range tested for environmental remediation [28,29,31].  
97 
98 
99 
100 
101 
102 
103 
104 
105 
106 
107 
108 
109 
110 
111 
112 
113 
114 
115 
116 
117 
118 
119 
2.2 Solid materials 
Fe0 material: The used Fe0 material is a readily available scrap iron. Its elemental 
composition was determined by X-Ray Fluorescence Analysis and was found to be: C: 
3.52%; Si: 2.12%; Mn: 0.93%; Cr: 0.66%. The material was fractionated by sieving. The 
fraction 1.6 - 2.5 mm was used. The sieved Fe0 was used without any further pre-treatment. 
Manganese oxide: A natural manganese nodule was used as source of MnO2. The sample 
was collected from the deep sea, crushed and sieved. An average particle size of 1.5 mm was 
used. Its elemental composition was determined by X-Ray Fluorescence Analysis and was 
found to be: Mn: 41.8%; Fe: 2.40%; Si: 2.41%; Ni: 0.74%; Zn: 0.22%; Ca: 1.39%; Cu: 
0.36%. These manganese nodules originated from the pacific ocean (Guatemala basin: 06°30 
N, 92°54 W and 3670 m deep). The target chemically active component is MnO2, which 
occurs naturally mainly as birnessite and todorokite [32-34]. Generally, natural manganese 
oxides exhibit marked variability of key structural parameters (e.g. porosity, degree of 
hydration, average manganese oxidation state) that influence their chemical reactivity. The 
used MnO2 was proven reactive in previous works [25]. 
Sand: The used sand was a commercial material for aviculture (“Papagaiensand” from RUT – 
Lehrte/Germany). Papagaiensand was used as received without any further pre-treatment nor 
characterization. This sand was the operational reference non-adsorbing material. 
 5 
Materials selected for study were known to be effective for adsorbing metallic ions (Fe0, 
MnO
120 
121 
122 
123 
124 
125 
126 
127 
128 
129 
130 
131 
132 
133 
134 
135 
136 
137 
138 
139 
140 
141 
142 
143 
144 
2, sand), delaying the availability of iron corrosion products in Fe0/H2O systems (MnO2) 
[20-25], or as admixing agent (sand).  
2.3 Experimental methodology 
Batch experiments without shaking were conducted in essay tubes containing 22.0 mL of the 
model solution (about 100 μM CrVI, CuII, MoVI, SbV, UVI and ZnII). Two sets of experiments 
were performed: Experiment 1 (≤ 33 days) and experiment 2 (60 days).  
Experiment 1: The essay tubes containing weighted solid materials were left further 
undisturbed for 1 to 33 d. At pre-selected times, 22 ml of the model solution was added to 
three tubes  to yield the wished experimental duration one day after the last solution addition. 
The time “one day after the last solution addition” (end of the experiment) was the date of the 
measurement of the pH value and the preparation of solutions (dilutions) for metal analysis. 
The batches consisted of 0 to 9.0 g L-1 of MnO2 or 5 g L-1 Fe0 and 0 to 9.0 g L-1 MnO2 or 
sand. The extent of metal removal and the pH value in each system was characterized at the 
end of the experiment. At this date, up to 200 μL of the supernatant solutions were carefully 
retrieved (no filtration) and diluted for concentration measurements. 
Experiment 2:  The essay tubes containing the model solution and (i) 0 to 9.0 g L-1 of MnO2 
or  (ii) 5 g L-1 Fe0 and 0 to 9.0 g L-1 MnO2 or sand were left undisturbed for 60 d, 200 μL of 
the supernatant solutions were retrieved and diluted (no filtration) for concentration 
measurements and the pH value of the remaining solution was measured. 
2.4 Analytical methods 
Analysis for Cr, Cu, Fe, Mn, Mo, Sb, U and Zn was performed by inductively coupled plasma 
mass spectrometry (ICP-MS) at the Department of Geochemistry (Centre of Geosciences, 
University of Göttingen). All chemicals used for experiments and analysis were of analytical 
grade. The pH value was measured by combination glass electrodes (WTW Co., Germany). 
 6 
The electrodes were calibrated with five standards following a multi-point calibration 
protocol [35] and in agreement with the new IUPAC recommendation [36]. 
145 
146 
147 
148 
149 
150 
151 
152 
153 
154 
155 
156 
157 
158 
159 
160 
161 
162 
163 
164 
165 
166 
167 
168 
169 
Each experiment was performed in triplicate and averaged results are presented. 
3 Results and Discussion 
3.1 pH variation 
Figure 1a depicts the time-dependant evolution of the pH value for the experiment with an 
initial pH of 4.0. MnO2 has no impact on the pH of the system. The pH increases in both 
systems containing Fe0 this is attributed to iron corrosion (Eq. 2) [37,38]. Three days after the 
start of the experiments, the pH in both Fe0 systems (Eq. 2 and Eq. 5) was larger than 5.0 
suggesting that quantitative contaminant removal by adsorption and co-precipitation was 
likely to occur [39,40]. In fact, metal removal mostly occurs by adsorption and co-
precipitation. Even chemically transformed metal species (e.g. CrIII, MoIII, UIV) must be 
removed by one of these mechanisms which are all coupled with iron oxide precipitation. 
Quantitative iron precipitation take place only when the pH value is larger than 4.0 to 4.5 
[37,38,41]. For pH < 4.0, the solubility of iron is high and iron precipitation is not 
quantitative. A classical example is that of Cr [42-44]. Soluble CrVI is reduced at pH 2.0-3.5 
to soluble CrIII and the pH is raised to a value above 6.0 where CrIII is essentially less soluble. 
Clearly, quantitative aqueous contaminant removal by Fe0 is only expected at pH > 4.5. 
Metals are then fixed selectively according to factors like their atomic radii [30], their charges 
at a given pH value, their oxidation state (Tab. 1).  
It is interesting to notice that the pH value was levelled to a value of 6.0 in the system “Fe0 + 
sand” after 10 days. The pH further increases to 9.0 in the system “Fe0 + MnO2” after 21 
days. The greater pH increase in the presence of MnO2 is a strong experimental evidence for 
the suitability of MnO2 to sustain the reactivity of Fe0 at the long-term ([19] and ref. cited 
therein). In fact, iron corrosion leading to elevated pH values is sustained by MnO2 (Eq. 5). 
 7 
It should be kept in mind that slow kinetics of the involved multi-steps heterogeneous 
reactions provides a continuing source of powerfully hydrous ferric oxides for contaminant 
removal and is thus advantageous as such systems are designed to function for years [1,45].  
170 
171 
172 
173 
174 
175 
176 
177 
178 
179 
180 
181 
182 
183 
184 
185 
186 
187 
188 
189 
190 
191 
192 
193 
194 
Figure 1b compares the extent of pH increase in both systems with Fe0 for the experiment at 
initial pH 6.0 (t ≤ 33 d). A clear difference between systems “Fe0 + sand” and “Fe0 + MnO2” 
is observed at this time. This difference is a reflect of the discussed sustained Fe0 reactivity by 
MnO2. According to Eq. 4 and Eq. 5, this difference in at first glance the reflect of the higher 
oxidative capacity of MnO2 compared to H2O. However, it should be kept in mind that Eq. 5 
is not likely to quantitatively occur in a single step, even under acidic conditions [27]. 
3.2 Iron release 
Figure 2 summarizes the time-dependant evolution the iron concentration for the investigated 
systems at pH 6.0 (initial pH). It is seen that the iron concentration first increased from 0 μM 
at t = 0 to about 150 μM at t = 8 d and then monotonically decreased to a value of 20 to 30 
μM in the system  “Fe0 + sand”. This iron profile is compatible with the process of oxide-film 
formation and the accompanying decreased corrosion kinetics [37,38,46]. 
Figure 2 also shows that, in system  “Fe0 + MnO2” dissolved Fe is maximal at t = 3 days and 
(about 35 μM) decreased to 7 μM at t = 30 d. This observation is compatible with recent 
findings from Pan and van Duin [47,48] who reported on a three-stages iron oxidation based 
on the generated species and oxidation speed. Accordingly, early iron oxides are rather mixed 
and instable, whereas the later oxides are more organized and stable. The oxidation speed is 
significantly reduced. Moreover generated FeII species are partly adsorbed onto crystallized 
iron oxides and are not present in the aqueous phase. Therefore, despite sustained reactivity of 
Fe0, Fe is not quantitatively released into the solution. Fe concentration remained very low but 
never decreased to undetectable levels. As discussed in section 3.1 the impact of MnO2 on pH 
increased was still measured two months after the start of the experiments. 
 8 
3.3 Manganese release 195 
196 
197 
198 
199 
200 
201 
202 
203 
204 
205 
206 
207 
208 
209 
210 
211 
212 
213 
214 
215 
216 
217 
218 
219 
Figure 3 compares the extent of Mn release in the three systems at pH 6.0 (initial pH). It is 
seen that the “Fe0 + MnO2” is the sole system releasing Mn. The Mn concentration first 
increased from 0 μM at t = 0 to about 120 μM at t = 7 d. The concentration is then levelled to 
about 80 μM through the end of the experiment. This levelled Mn concentration ([Mn] ≠ 0 
μM) corroborated sustained Fe0 reactivity (Eq. 4). It should be noticed that in the system 
“MnO2 alone”, Mn concentration was constantly lower than 20 μM. It is known than MnO2 is 
very stable in water under ambient conditions [32-35,49]. 
3.4 Metal removal 
3.4.1 Initial pH = 4.0 
Figure 4a summarizes the residual metal concentration at day 33 as a function of the MnO2 
loading in system “Fe0 + MnO2”. It is clearly shown that the extent of contaminant removal 
was negligible although the final pH was larger that 5.0 (section 3.1).  Fig. 4b shows that 
systems with lower MnO2 loadings exhibited higher Fe concentration. This observation is 
consistent with the above discussed sustainability of Fe0 reactivity by MnO2 (Eq. 4). In fact, 
larger MnO2 loadings are coupled with greater extent of Fe0 corrosion, yielding larger pH 
values and decreased Fe solubility [38]. 
The main feature from Fig. 4 is that Fe0 is not suitable for water treatment when the pH is 
lower than 5.0 (e.g. acid mine drainage). For such situations alternatives should be used or the 
pH will be first increased. Another important feature from Fig. 4a is that the concentration of 
Cu and Cr were partly higher than that of the working initial solution. This result is 
compatible with the fact that these elements were leached from MnO2 [33]. In fact, while the 
Cr content of the used MnO2 was not measured, the Cu was 0.36 % as indicated in section 
2.2. 
3.4.2 Initial pH = 6.0 
 9 
Figure 5 summarizes the evolution of the residual metal concentration (μM) for the three 
systems at initial pH 6.0. It is shown from Fig. 5a that metal removal by MnO
220 
221 
222 
223 
224 
225 
226 
227 
228 
229 
230 
231 
232 
233 
234 
235 
236 
237 
238 
239 
240 
241 
242 
243 
2 is minimal. In 
the presence of Fe0, the removal of CrVI, CuII, SbV, UVI and ZnII is  quantitative (> 90 %) for 
33 d contact time. However, the removal extent of theses species is significantly influenced 
by the presence of MnO2 (“Fe0 + sand” vs. “Fe0 + MnO2” systems). MoVI is the sole ion 
which removal extent has never reached 90 % under tested experimental conditions (Tab. 2).  
Table 3 comparatively quantifies the extent of metal removal in the “Fe0 + sand” and “Fe0 + 
MnO2” systems. Negative δ1 and δ2 values clearly demonstrated the delay of metal removal 
due to the presence of MnO2. Positive Δ’ values indicate an overall progression of the process 
of metal removal despite delay due to the presence of MnO2. The ability of MnO2 to sustain 
Fe0 reactivity is delineated. 
In system “Fe0 + sand”, MoVI removal is not quantitative (Tab. 2, Fig. 5b). In system “Fe0 + 
MnO2”, SbV and MoVI removal are not quantitative (Tab. 2, Fig. 5c). This result corroborates 
the view that metals are removed by “free” iron corrosion products [21,22,23,25] since both 
elements exhibit the lowest affinity for iron oxides ([28,29] and references cited therein). It is 
very important to notice that the surface charge of the adsorbent alone is not sufficient for the 
discussion of the removal behaviour. In fact, Tab. 1 shows that only Zn and Cu are present as 
positively charged species (cations) which are readily adsorbed by negatively charged iron 
oxides [29]. However, U and Cr were also available as negatively charged species (anions). 
Furthermore, in Fe0/H2O systems, CrVI can be readily reduced (e.g. by FeII species) to less 
soluble CrIII while reduction of soluble UVI to less soluble UIV is less favourable. 
Nevertheless, Cr and U exhibited very similar removal behaviours. It should be kept in mind 
that iron oxides and hydroxides are in-situ generated and are in permanent transformation 
such that contaminant removal is not performed by a well-defined adsorbing agent [39,40]. 
 10 
The experiments at initial pH 6.0 were duplicated for 60 d (Experiment 2) to access the 
evolution of the “Fe
244 
245 
246 
247 
248 
249 
250 
251 
252 
253 
254 
255 
256 
257 
258 
259 
260 
261 
262 
263 
264 
265 
266 
267 
0 + MnO2” system. The results are presented in Tab. 2 and Tab. 3. Results 
showed a continuous decrease of the concentration for all elements. For example, the removal 
extent of MoVI increased from 70.0 % after 1 month to 83.7 %  after 2 months in the “Fe0 + 
sand” system (Δ1,2 value = 13.7 %). In the “Fe0 + MnO2” system, the extent of MoVI removal 
increased from 60.3 to 74.7 % (Δ’1,2 value = 14.4 %).  Tab. 2 shows the following variations 
for Δ1,2 and Δ’1,2 values: 0.4 ≤ Δ1,2 ≤ 13.7 ; 1.3 ≤ Δ’1,2  ≤ 16.3. The highest Δ values 
correspond to elements (Mo, Sb and Zn) with the lowest affinity to iron oxides in the 
investigated pH range (pH > 6.0).  Note that, in the “Fe0 + sand” system, further contaminant 
removal beyond 33 d was only significant (> 3 %) for Mo (13.7 %) and Zn (3.8 %) (Tab. 2). 
Table 3 shows that the impact of MnO2 on the extent of metal removal was higher after 1 
month (δ1 > δ2). The increasing order of δ1 value was: Cr < Mo < Cu < Zn < Sb < U. The 
increasing order of element covalent radius (Tab. 1) is: Cu = Zn < Cr < Mo = Sb < U. The 
increasing order of element atomic mass (Tab. 1) is: Cr < Cu < Zn < Mo < Sb < U. A survey 
of these three series suggests that, similar to the charge of the species, none of the criteria is 
really relevant to rationalize the interaction of tested element with in-situ generated iron 
corrosion products. Tested metals ions have permanent or variable oxidation state (Tab. 1) 
and possibly participate in redox processes. All these properties determine their removal from 
the aqueous phase [29] . 
4 Concluding remarks 
The suitability of Fe0 for the removal of dissolved CrVI, CuII, MoVI, SbV, UVI, and ZnII is 
accessed in this communication. Results corroborated the view that tested contaminants are 
removed by in-situ generated iron corrosion products. A non-reducible species (ZnII), two less 
adsorbable species (MoVI, SbV) and three other elements with slightly different affinity to iron 
 11 
corrosion products are all quantitatively removed for sufficient long experimental durations (> 
1 month). 
268 
269 
270 
271 
272 
273 
274 
275 
276 
277 
278 
279 
280 
281 
282 
283 
284 
285 
286 
287 
288 
289 
290 
The suitability of MnO2 to sustain Fe0 reactivity for the contaminant removal is also 
confirmed. The importance of this latter aspect has been recently presented in theoretical 
works on the suitability of MnO2 to sustain the efficiency of household water filters [19]. 
While MnO2 has been proved to delay the kinetics of contaminant removal in short term batch 
experiments, it has been postulated that the main benefit of MnO2 is to sustain Fe0 corrosion. 
In fact, in Fe0 filters, sustained Fe0 corrosion produces iron (hydr)oxides at different depths 
for quantitative contaminant removal. Future works should investigate this aspect for a proper 
design of sustainable Fe0 filtration systems at all scales (household filters, subsurface reactive 
walls). A recently presented tool for the design of laboratory column experiments for better 
results comparability [50] could support these efforts. 
Acknowledgments 
Thoughtful comments provided by Angelika Schöner (FSU Jena - Germany) on the revised 
manuscript are gratefully acknowledged. The used scrap iron was kindly purchased by the 
branch of the Metallaufbereitung Zwickau, (MAZ) in Freiberg. The manuscript was improved 
by the insightful comments of anonymous reviewers from the Chemical Engineering Journal. 
References 
[1] R.W. Gillham, Development of the granular iron permeable reactive barrier technology 
(good science or good fortune). In "Advances in environmental geotechnics : 
proceedings of the International Symposium on Geoenvironmental Engineering in 
Hangzhou, China, September 8-10, 2009"; Y. Chen, X. Tang, L. Zhan (Eds);  Springer 
Berlin/London (2010) 5–15. 
 12 
[2] S. Comba, A. Di Molfetta, R. Sethi, A Comparison between field applications of nano-, 
micro-, and millimetric zero-valent iron for the remediation of contaminated aquifers, 
Water Air Soil Pollut. 215 (2011) 595–607. 
291 
292 
293 
294 
295 
296 
297 
298 
299 
300 
301 
302 
303 
304 
[3] M. Gheju, hexavalent chromium reduction with zero-valent iron (ZVI) in aquatic systems, 
Water Air Soil Pollut. (2011) doi 10.1007/s11270-011-0812-y. 
[4] A.D. Henderson, A.H. Demond, Long-term performance of zero-valent iron permeable 
reactive barriers: a critical review, Environ. Eng. Sci. 24 (2007) 401–423. 
[5] C. Noubactep, A. Schöner, P. Woafo, Metallic iron filters for universal access to safe 
drinking water, Clean: Soil, Air, Water 37 (2009) 930–937. 
[6] C. Noubactep, Metallic iron for safe drinking water worldwide, Chem. Eng. J. 165 (2010) 
740–749. 
[7] C. Noubactep, A. Schöner, Metallic iron: dawn of a new era of drinking water treatment 
research? Fresen. Environ. Bull. 19 (2010) 1661–1668. 
[8] C. Noubactep, Metallic iron for safe drinking water production, Freiberg Online Geology, 
vol. 27 (2011) 38 pp. ISSN 1434-7512. (www.geo.tu-freiberg.de/fog). 305 
306 
307 
308 
309 
310 
311 
312 
313 
314 
315 
[9] C. Noubactep, S. Caré, Enhancing sustainability of household water filters by mixing 
metallic iron with porous materials, Chem. Eng. J. 162 (2010) 635–642. 
[10] C. Noubactep, S. Caré, F. Togue-Kamga, A. Schöner, P. Woafo, Extending service life 
of household water filters by mixing metallic iron with sand, Clean – Soil, Air, Water 38 
(2010) 951–959. 
[11] C. Anstice, C. Alonso, F.J. Molina, Cover cracking as a function of bar corrosion: part I-
experimental test, Materials and structures 26 (1993) 453–464. 
[12] S. Caré, Q.T. Nguyen, V. L’Hostis, Y. Berthaud, Mechanical properties of the rust layer 
induced by impressed current method in reinforced mortar, Cement and Concrete 
Research 38 (2008) 1079–1091. 
 13 
[13] A.H. Khan, S.B. Rasul, A.K.M. Munir, M. Habibuddowla, M. Alauddin, S.S. Newaz, A. 
Hussam, Appraisal of a simple arsenic removal method for groundwater of Bangladesh, 
J. Environ. Sci. Health A 35 (2000) 1021–1041. 
316 
317 
318 
319 
320 
321 
322 
323 
324 
325 
326 
327 
328 
329 
330 
331 
332 
333 
334 
335 
336 
337 
338 
339 
340 
[14] A. Hussam, A.K.M. Munir, A simple and effective arsenic filter based on composite iron 
matrix: Development and deployment studies for groundwater of Bangladesh, J. 
Environ. Sci. Health A 42 (2007) 1869–1878. 
[15] T.K.K. Ngai, R.R. Shrestha, B. Dangol, M. Maharjan, S.E. Murcott, Design for 
sustainable development – Household drinking water filter for arsenic and pathogen 
treatment in Nepal, J. Environ. Sci. Health A 42 (2007) 1879–1888. 
[16] H. Chiew, M.L. Sampson, S. Huch, S. Ken, B.C. Bostick, Effect of groundwater iron and 
phosphate on the efficacy of arsenic removal by iron-amended biosand filters, Environ. 
Sci. Technol. 43 (2009) 6295–6300. 
[17] A. Hussam, Contending with a Development Disaster: SONO Filters Remove Arsenic 
from Well Water in Bangladesh, Innovations 4  (2009) 89–102. 
[18] C. Noubactep, S. Caré, Dimensioning metallic iron beds for efficient contaminant 
removal, Chem. Eng. J. 163 (2010) 454–460. 
[19] C. Noubactep, S. Caré, K.B.D. Btatkeu, C.P. Nanseu-Njiki, Enhancing the sustainability 
of household Fe0/sand filters by using bimetallics and MnO2, Clean – Soil, Air, Water 
38 (2011) xy–zt. 
[20] C. Noubactep, Characterizing the discoloration of methylene blue in Fe0/H2O systems, J. 
Hazard. Mater. 166 (2009) 79–87. 
[21] A. Ghauch, H. Abou Assi, A. Tuqan, Investigating the mechanism of clofibric acid 
removal in Fe0/H2O systems, J. Hazard. Mater. 176 (2010) 48–55. 
[22] A. Ghauch, H. Abou Assi, S. Bdeir, Aqueous removal of diclofenac by plated elemental 
iron: Bimetallic systems, J. Hazard. Mater. 182 (2010) 64–74. 
 14 
[23] A. Ghauch, H. Abou Assi, H. Baydoun, A.M. Tuqan, A. Bejjani, Fe0-based trimetallic 
systems for the removal of aqueous diclofenac: Mechanism and kinetics, Chem. Eng. J. 
(2011) 1033–1044. 
341 
342 
343 
344 
345 
346 
347 
348 
349 
350 
351 
352 
353 
354 
355 
356 
357 
358 
359 
360 
361 
362 
363 
364 
[24] D. Burghardt, A. Kassahun, Development of a reactive zone technology for simultaneous 
in situ immobilisation of radium and uranium, Environ. Geol. 49 (2005) 314–320. 
[25] C. Noubactep, G. Meinrath, J.B. Merkel, Investigating the mechanism of uranium 
removal by zerovalent iron materials, Environ. Chem. 2 (2005), 235–242. 
[26]  D.F.A. Koch, Kinetics of the reaction between manganese dioxide and ferrous ion, Aust. 
J. Chem 10 (1957) 150–159. 
[27] M.Sh. Bafghi, A. Zakeri, Z. Ghasemi, M. Adeli, Reductive dissolution of manganese ore 
in sulfuric acid in the presence of iron metal. Hydrometallurgy 90  (2008) 207–212. 
[28] T.B. Scott, I.C. Popescu, R.A. Crane, C. Noubactep, Nano-scale metallic iron for the 
treatment of solutions containing multiple inorganic contaminants, J. Hazard. Mater. 
186 (2011) 280–287. 
[29] Y.N. Vodyanitskii, The role of iron in the fixation of heavy metals and metalloids in 
soils: a review of publications, Eurasian Soil Sci. 43 (2010) 519 –532. 
[30] J.C. Slater, Atomic Radii in Crystals. J. Chem. Phys. 41 (1964) 3199 –3205. 
[31] K.J. Cantrell, D.I. Kaplan, T.W. Wietsma, Zero-valent iron for the in situ remediation of 
selected metals in groundwater, J. Hazard. Mater. 42 (1995) 201–212. 
[32] J.E. Post, Manganese oxide minerals: Crystal structures and economic and environmental 
significance, Proc. Natl. Acad. Sci. USA 96 (1999), 3447–3454. 
[33] A. Mukherjee, A.M. Raichur, K.A. Natarajan, J.M. Modak, Recent developments in 
processing ocean manganese nodules – A critical review, Mineral Processing & 
Extractive Metall. Rev. 25 (2004) 91–127. 
 15 
[34] I. Saratovsky, P.G. Wightman, P.A. Pasten, J.F. Gaillard, K.R. Poeppelmeier, Manganese 
oxides: Parallels between abiotic and biotic structure, J. Am. Chem. Soc. 128 (2006) 
11188–11198. 
365 
366 
367 
368 
369 
370 
371 
372 
373 
374 
375 
376 
377 
378 
379 
380 
381 
382 
383 
384 
385 
386 
387 
[35] G. Meinrath, P. Spitzer, Uncertainties in determination of pH, Mikrochem. Acta 135 
(2000) 155–168. 
[36] R.P. Buck, S. Rondinini, A.K. Covington, F.G.K. Baucke, C.M.A. Brett, M.F. Camoes, 
M.J.T. Milton, T. Mussini, R. Naumann, K.W. Pratt, P. Spitzer, G.S. Wilson, 
Measurement of pH. Definition, standards, and procedures (IUPAC Recommendations 
2002), Pure Appl. Chem. 74 (2002) 2169–2200. 
[37] A.Y. Aleksanyan, A.N. Podobaev, I.I. Reformatskaya, Steady-state anodic dissolution of 
iron in neutral and close-to-neutral media, Protection of Metals 43 (2007) 66–69. 
[38] S. Nesic, Key issues related to modelling of internal corrosion of oil and gas pipelines – 
A review, Corros. Sci. 49 (2007) 4308–4338. 
[39] C. Noubactep, The fundamental mechanism of aqueous contaminant removal by metallic 
iron, Water SA 36 (2010) 663–670. 
[40] C. Noubactep, Aqueous contaminant removal by metallic iron: Is the paradigm shifting? 
Water SA 37 (2011) 419–426. 
[41] G.W. Whitman, R.P. Russel, V.J. Altieri, Effect of hydrogen-ion concentration on the 
submerged corrosion of steel, Indust. Eng. Chem. 16 (1924) 665–670. 
[42] L.-Y. Chang, Alternative chromium reduction and heavy metal precipitation methods for 
industrial wastewater, Environ. Prog. 22 (2003) 174–182. 
[43] L.-Y. Chang, Chromate reduction in wastewater at different pH levels using thin iron 
wires - A laboratory study. Environ. Prog. 24 (2005) 305–316. 
 16 
[44] M. Gheju, I. Balcu, Removal of chromium from Cr(VI) polluted wastewaters by 
reduction with scrap iron and subsequent precipitation of resulted cations. J. Hazard. 
Mater. (2011), doi:10.1016/j.jhazmat.2011.09.002. 
388 
389 
390 
391 
392 
393 
394 
395 
396 
397 
398 
399 
400 
401 
402 
403 
404 
405 
406 
[45] D.E. Giles, M. Mohapatra, T.B. Issa, S. Anand, P. Singh, Iron and aluminium based 
adsorption strategies for removing arsenic fromwater, J. Environ. Manage. 92 (2011) 
3011–3022. 
 [46] P. Schmuki, From Bacon to barriers: a review on the passivity of metals and alloys, J. 
Solid State Electrochem. 6 (2002) 145–164. 
[47] T. Pan, A.C.T. van Duin, Passivation of steel surface: An atomistic modeling approach 
aided with X-ray analyses, Mater. Lett. 65  (2011) 3223–3226. 
[48] T. Pan, Quantum chemistry-based study of iron oxidation at the iron–water interface: An 
X-ray analysis aided study, Chem. Phys. Lett. 511  (2011) 315–321. 
[49] D. Postma, C.A.J. Appelo, Reduction of Mn-oxides by ferrous iron in a flow system: 
column experiment and reactive transport modelling, Geochim. Cosmochim. Acta 64 
(2000) 1237–1247. 
[50] C. Noubactep, S. Caré, Designing laboratory metallic iron columns for better result 
comparability, J. Hazard. Mater. 189 (2011) 809–813. 
 
 17 
Table 1: Some characteristics of iron, manganese and the six tested metals.  R is the 
empirically element covalent radius (in picometers - pm) after Slater [30]. M (g/mol) is the 
element atomic mass. DO is the degree of oxidation. The used DO is bold-marked and 
underlined. C
406 
407 
408 
409 
410 
411 
412 
413 
0 is the element initial concentration in μM and mg/L (ppm). It is seen that for 
the same molar concentration (100 μM) the mass concentration varies from 5.2 ppm for Cr to 
23.8 ppm for U. C0,eff is the operational initial concentration for the experiment at pH 6.0 (see 
text). The most likely species of tested element at pH 6.0 to 9.0 is given [29]. 
 
X R M DO C0 C0 C0,eff Speciation 
 (pm) (g/mol) (-) (μM) (mg/L) (μM) (-) 
Cr 140 51.996 III, VI 100 5.2 98.9 HCrO4-
Cu 135 63.546 II 100 6.4 86.8 [Cu(H2O)6]2+
Fe 140 55.847 0, II, III 0.0 0.0 0.0 - 
Mn 140 54.938 II, III, VI 0.0 0.0 0.0 - 
Mo 145 95.94 IV, VI 100 9.6 98.5 MoO42-
Sb 145 121.75 III, V 100 12.2 72.1 Sb(OH)6-
U 175 238.029 IV, VI 100 23.8 79.2 [UO2(CO3)3]4-
Zn 135 65.38 II 100 6.5 87.3 [Zn(H2O)6]2+
 414 
415 
 18 
Table 2:  Comparison of the extent of metal removal (in %) in the “Fe0 + sand” and “Fe0 + 
MnO
415 
416 
417 
418 
419 
2” systems after 1 and 2 months. C0 is the operational initial concentration (see text). Δ is 
the different of the removal extent after 1 month and 2 months. 
 
 
Element C0 Fe0 + sand Fe0 + MnO2
 (μM) 1 month 2 months Δ1,2 1 month 2 months Δ’1,2
Cr 98.9 98.5 98.9 0.4 98.1 99.4 1.3 
Cu 86.8 93.7 96.4 2.7 83.1 91.3 8.3 
Mo 98.5 70.0 83.7 13.7 60.3 74.7 14.4 
Sb 72.1 92.8 95.2 2.4 77.1 93.4 16.3 
U 79.2 96.2 97.1 0.9 72.2 83.2 11.0 
Zn 87.3 92.9 96.7 3.8 79.8 94.3 14.5 
420 
421 
 
 19 
Table 3: Impact of MnO2 on the extent of metal removal by Fe0. δi is the difference between 
the extent of metal removal in system “Fe
421 
422 
423 
424 
0 + MnO2” (P’i) and “Fe0 + sand” (Pi) after i months 
(δI = P’i - Pi). Δ’ = δ2 - δ1. 
 
Element Fe0 + sand Fe0 + MnO2 δ1 δ2 Δ' 
 P1 P2 P1’ P2’ (%) (%) (%) 
Cr 98.5 98.9 98.1 99.4 -0.4 0.4 0.8 
Cu 93.7 96.4 83.1 91.3 -10.6 -5.1 5.5 
Mo 70.0 83.7 60.3 74.7 -9.7 -9.0 0.7 
Sb 92.8 95.2 77.1 93.4 -15.8 -1.8 13.9 
U 96.2 97.1 72.2 83.2 -24.0 -13.9 10.1 
Zn 92.9 96.7 79.8 94.3 -13.1 -2.4 10.7 
425 
426 
427 
 
 
 20 
Figure 1 427 
428 
429 
 
 
0 6 12 18 24 30 36
4
5
6
7
8
9
10 (a) Fe
0 + MnO2
 Fe0 + sand
 MnO2
pH
 v
al
ue
elapsed time / [days]
 430 
431  
0 2 4 6 8 10
7
8
9
10
11
(b)
 sand
 MnO2
fin
al
 p
H
 v
al
ue
additive dosage / [g/L]
 432 
433 
434 
435 
436 
 
 
 
 21 
Figure 2 436 
437  
0 6 12 18 24 30 36
0
25
50
75
100
125
150
175
pH = 6.0
 MnO2
 Fe0 + MnO2
 Fe0 + sand
iro
n 
/ [
μM
]
elapsed time / [days]
 438 
439 
 22 
439 
440 
Figure 3 
 
0 6 12 18 24 30 36
0
25
50
75
100
125
 MnO2
 Fe0 + MnO2
 Fe0 + sand
m
an
ga
ne
se
 / 
[μM
]
elapsed time / [days]
 441 
442 
 23 
Figure 4 442 
443  
0 4 8 12 16 20
60
75
90
105
120 pH = 4.0
Fe0 + MnO2
 Cr
 Cu
 Zn
 Mo
 Sb
 U
el
em
en
t /
 [μ
M
]
MnO2 / [g/L]
 444 
445  
0 3 6 9 12 15
0
90
180
270
360
450 pH = 4.0
Fe0 + MnO2
 Mn
 Fe
el
em
en
t /
 [μ
M
]
MnO2 / [g/L]
 446 
447 
448 
 
 24 
Figure 5 448 
0 5 10 15 20 25 30 35
0
10
20
30
40
50
60
70
80
90
(a)
MnO2 alone
pH = 6.0
 Cr
 Cu
 Zn
 Mo
 Sb
 Uel
em
en
t /
 [μ
M
]
elapsed time / [days]
 449 
0 5 10 15 20 25 30 35
0
10
20
30
40
50
60
70
80
90
(b)
pH = 6.0
Fe0 + sand
 C r
 Cu
 Zn
 M o
 Sb
 U
el
em
en
t /
 [μ
M
]
elapsed time / [days]
 450 
0 5 10 15 20 25 30 35
0
10
20
30
40
50
60
70
80
90
(c)
pH = 6.0
Fe0 + MnO2
 Cr
 Cu
 Zn
 Mo
 Sb
 U
el
em
en
t /
 [m
M
]
elapsed time / [dyas]
 451 
452 
 25 
Figure Captions 452 
453 
454 
455 
456 
457 
458 
459 
460 
461 
462 
463 
464 
465 
466 
467 
468 
469 
470 
471 
472 
473 
474 
475 
476 
Figure 1: Variation of the pH value in the investigated systems. (a) time-dependant evolution 
for the first 33 days, and (b) variation with additive loading after 60 days. The used Fe0 and 
additive (MnO2 or sand) loadings are 5.0 and 2.5 g/L respectively. The lines simply connect 
points to facilitate visualization. 
 
Figure 2: Time-dependant evolution of the iron concentration for the first 33 days in the 
experiments at pH 6.0 with the three investigated systems. The used Fe0 and additive (MnO2 
or sand) loadings are 5.0 and 2.5 g/L respectively. The lines simply connect points to facilitate 
visualization. 
 
Figure 3: Time-dependant evolution of the manganese concentration for the first 33 days in 
the experiments at pH 6.0 with the three investigated systems. The used Fe0 and additive 
(MnO2 or sand) loadings are 5.0 and 2.5 g/L respectively. The lines simply connect points to 
facilitate visualization. The lines simply connect points to facilitate visualization. 
 
Figure 4: Evolution of the concentration of dissolved metals as function of the additive mass 
loading for the experiment with “Fe0 + MnO2” and pH 4.0 as initial value: (a) aqueous metal 
removal, and (b) metal removal from Fe0 and MnO2. The used Fe0 mass loading is 5.0 g/L. 
The lines simply connect points to facilitate visualization. 
 
Figure 5: Time-dependant evolution of the concentration of dissolved metals for the 
experiment with pH 6.0 as initial value: (a) system “MnO2 alone”, (b) “Fe0 + sand” and (c) 
“Fe0 + MnO2”. The used Fe0 and additive (MnO2 or sand) loadings are 5.0 and 2.5 g/L 
respectively. The lines simply connect points to facilitate visualization. 
 26