Effects of mixing granular iron with sand on the efficiency of methylene blue 
discoloration 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
26 
Miyajima K.a, Noubactep C.a,b,* 
aAngewandte Geologie, Universität Göttingen, Goldschmidtstraße 3, D-37077, Göttingen, Germany. 
bKultur und Nachhaltige Entwicklung CDD e.V., Postfach 1502, D-37005 Göttingen, Germany 
* corresponding author: e-mail: cnoubac@gwdg.de; Tel. +49 551 39 3191, Fax: +49 551 399379. 
 
Abstract 
The influence of granular sand on the efficiency of metallic iron (Fe0) for the discoloration of 
a methylene blue (MB) solution was investigated in the current work. The initial MB 
concentration was 10 mg L-1 and mass loadings within the range of 0 to 90 g L-1 for sand and 
0 to 45 g L-1 for Fe0 were applied. The batch reaction vessel used was a graduated essay tube 
containing 22.0 mL of the MB solution. Shaking intensities of 0 and 75.0 rpm were applied 
for experimental durations of 7, 21 and 45 days. Results provide clear evidence that both Fe0 
and sand were independently effective for the discoloration of MB. However, the latter 
material was significantly less effective, recording 54.0 % compared to 82.0 % recorded for 
the Fe0 after 45 days in experiment with 45.0 g L-1 of each material. Similarly, mixing 90 g L-
1 sand with 45.0 g L-1 of Fe0 depicted a MB discoloration efficacy of 72.0 % demonstrating 
that the discoloration capability of the Fe0 was significantly ‘masked’ by the presence of sand. 
This observation provides clear evidence to question the common approach of using 
adsorbents for contaminant accumulation in the vicinity of Fe0 materials in order to facilitate 
chemical reduction by Fe0. Further research is required to determine the relative affinity of 
different materials that can be used in Fe0 mixtures for maximum contaminant removal 
efficacies. 
Keywords: Dye discoloration, Fe0/sand mixtures, Methylene blue, Water treatment, 
Zerovalent iron. 
1 Introduction 27 
28 
29 
30 
31 
32 
33 
34 
35 
36 
37 
38 
39 
40 
41 
42 
43 
44 
45 
46 
47 
48 
49 
50 
51 
52 
Metallic iron (Fe0) has been demonstrated in numerous studies to represent the best available 
material for subsurface permeable reactive barriers [1-5]. Fe0 has also been demonstrated as a 
highly efficient material for wastewater treatment and safe drinking water provision [5-10]. In 
all these applications Fe0 is routinely mixed with inert materials. The most used additive is 
sand [11-15]. Reported goals of mixing sand and Fe0 are: (i) meeting design requirements 
(goal 1), (ii) saving Fe0 costs (goal 2), and (iii) delaying particle clogging (goal 3). However, 
actually there is no conclusive experimental evidence to demonstrate that Fe0/sand mixtures 
are more or less effective than pure Fe0 systems [16]. 
The relevance of mixing iron and sand was recognized since the early phase of technology 
development [17]. However, the literature still contains limited information on Fe0/sand 
mixtures [15]. The need for systematic work aiming at establishing the practical use of 
Fe0/sand mixtures has been recently theoretically discussed as summarized in ref. [18]. 
Results concluded that, when designing a Fe0 treatment system, priority must be placed on the 
aforementioned goals number 1 and 3, stating that goal number 2 (low cost) is required, but 
not an instrinsic requirement of a Fe0 filtration system. In fact, mixing Fe0 and sand is 
regarded as reducing the proportion of Fe0, and thus ‘creating’ or ‘leaving’ room for sustained 
iron corrosion [19,20]. In other words, theoretical studies disprove the view of Ulsamer [16]. 
Moreover, Fe0/sand systems are more sustainable than pure Fe0 system as a rule [18]. On the 
other hand, the statement of Ulsamer [16] resulted from a critical literature review, showing 
that available data are not univocal [15]. These data resulted mostly from columns studies 
[15,21,22]. While a column experiment is the most effective method to investigate the 
mechanisms behind the efficacy of a Fe0 filtration system, a well-designed batch experiments 
could be useful in fine-tuning some relevant aspects at the laboratory scale [5]. For example 
goal 3 (avoiding or delaying particle clogging) can be properly investigated in batch systems 
using tubular vessels enabling juxtaposition of used materials at the bottom of the vessel. 
 2
Laboratory essay tubes are such vessels [23]. A survey of the literature revealed that the effect 
of mixing sand and Fe
53 
54 
55 
56 
57 
58 
59 
60 
61 
62 
63 
64 
65 
66 
67 
68 
69 
70 
71 
72 
73 
74 
75 
76 
77 
78 
0 has been addressed since the early phase of research on Fe0 
remediation technology [24-28]. However, sand has been constantly used as inert additive 
which contribution to contaminant removal is related to the presence of Fe0.  For example, 
Song et al. [14] discussed the effect of mixing Fe0 and sand on the extent of CrVI removal in 
batch studies, using experimental results from Kim et al. [12]. 
Kim et al. [12] investigated aqueous CrVI removal in batch studies, and reported that, the CrVI 
removal efficiency after 48 hours greatly increased from about 50 % in pure Fe0 systems to 85 
% in Fe0/sand systems. The authors reported that reduction products (CrIII species) adsorbed 
onto sand which shows higher sorption affinity than Fe0, thus “leaving a large portion of 
active sites of the Fe0 unblocked” [14]. This conclusion seems counter-intuitive as in Fe0/sand 
mixtures, reaction rates should diminish due to the decrease in reactive surface concentration 
(disadvantaged mass transfer of CrIII to Fe0 surface). On the other hand, the adsorption 
capacity of sand for Cr is limited. Therefore, coating sand with iron oxides is an established 
tool to sustain metal removal in sand filters [29,30]. The efficiency of iron-coated sand filters 
suggests that in the experiments of Kim et al. [12], in-situ generated iron oxides were 
adsorbed onto sand and subsequently removed CrVI or CrIII. Alternatively, CrVI adsorbed onto 
in-situ generated iron-coated sand can be further reduced to CrIII by FeII or H2 from 
continuous Fe0 oxidation [31,32]. 
The conclusions of Kim et al. [12] are consistent with the view, that sand delays clogging 
[17,22]. Clogging delay is achieved by decreasing the proportion of reactive Fe0 which 
progressively fill the connected pores with expansive corrosion products having a volume 2.1 
to 6.4 larger than Fe0 is the metal [33]. By mixing Fe0 with inert sand, a certain pH control 
can also be achieved as the proportion of Fe0 inducing pH elevation is limited [15]. While the 
optimal proportion of Fe0 relevant for the design of Fe0 filtration systems can not be achieved 
in batch experiments the qualitative compaction can be observed. 
 3
The objective of the present communication is to provide evidence for the effect of mixing 
Fe
79 
80 
81 
82 
83 
84 
85 
86 
87 
88 
89 
90 
91 
92 
93 
94 
95 
96 
97 
98 
99 
100 
101 
102 
103 
104 
0 and sand on the process of aqueous contaminant removal. A commercial Fe0 material is 
used. Methylene blue (MB - 10 mg L-1) is used as model contaminant. The extent of MB 
discoloration is characterized using three different systems: (i) Fe0 alone, (ii) sand alone, and 
(iii) “Fe0 + sand”. Non-disturbed experiments and experiments shaken at 75 rpm were 
performed. Results are comparatively discussed. 
2 Background of the experimental methodology 
Methylene blue (MB) is a cationic dye which is preferentially adsorbed onto negatively 
charged surfaces [34,35]. The application of MB adsorption in environmental science was 
demonstrated affordable, applicable and rapid [34,36,37]. MB adsorption is also widely used 
in the context of ‘Fe0 for environmental remediation’ (e.g. in Fe0/H2O systems) [37,38]. 
2.1 MB discoloration in Fe0/H2O systems 
In Fe0/H2O systems, MB discoloration results from the synergy between (i) adsorption onto 
in-situ generated iron oxides and (ii) co-precipitation with nascent iron hydroxides [8,10]. It is 
essential to note that discoloration by a redox process (E0 = 0.01 V at pH 7 vs. E0 = -0.44 V 
for FeII/Fe0) was proven insignificant as the large polar organic molecule can not 
quantitatively access the Fe0 surface [8,10]. The Fe0 surface is permanently covered with a 
multi-layered oxide scale. On the other hand, MB adsorption onto natural sand has been 
documented by Mitchell et al. [39]. These authors tested 65 sand samples from 45 localities 
from which only 6 samples (9.2 %) depicted ‘low’ adsorptive affinity for MB. The following 
observation of Mitchell et al. [39] is essential for Fe0/H2O systems: “sand artificially coated 
with Fe2O3 or Cr2O3 adsorbed very little methylene blue”. Recently, Varlikli et al. [40] tested 
the suitability of Sahara desert sand (SaDeS) for removing organic (cationic and anionic) dyes 
from aqueous solutions. They observed that MB demonstrated the strongest affinity for 
SaDeS with an adsorption capacity of 11.98 mg g-1 (initial MB concentration: 11.2 mg L-1) 
2.2 MB discoloration in Fe0/sand/H2O systems  
 4
The observation of Mitchell et al. [39] that coated metal oxides impairs the adsorption of MB 
by sand suggests that a sand sample is a better MB adsorbent than the resulted iron oxide 
coated one. A Fe
105 
106 
107 
108 
109 
110 
111 
112 
113 
114 
115 
116 
117 
118 
119 
120 
121 
122 
123 
124 
125 
126 
127 
128 
129 
130 
0/sand/H2O system is a dynamic system containing 100 % ‘clean’ sand only 
at the initial time (t = 0). Once iron corrosion starts, sand surface is progressively coated with 
in-situ generated iron oxides. In other words, the adsorption capacity of sand for MB in a 
Fe0/sand/H2O system is maximal at t = 0. This adsorption capacity decreases with time as a 
function of three key operational parameters: (i) the kinetics of Fe0 oxidation (Fe0 intrinsic 
reactivity), (ii) the available amount of sand, and (iii) the available amount of Fe0.  
Since Fe0 alone (e.g. Fe0/H2O system) is efficient in decolourising aqueous MB [40,41], the 
effects of sand on MB discoloration in the presence of Fe0 may be summarized in two 
hypothesis. First, MB discoloration by Fe0 is not affected by the presence of sand 
(Assumption 1). Second, MB discoloration by Fe0 is significantly retarded by the presence of 
sand (Assumption 2). Based on the results of Mitchell et al. [39], the obvious case that sand 
addition may increase MB discoloration is not considered. 
The used methodology for the investigation of the impact of sand on MB discoloration by Fe0 
comprises testing the validity of Assumption 1 and Assumption 2 by following the MB 
discoloration in the presence (absence) of sand. For this purpose, the addition of sand is tested 
as a tool to delay the availability of “free” corrosion products. In-situ generated iron corrosion 
products are adsorbed onto the sand surface, worsening its capacity of MB adsorption [39]. 
Adsorbed corrosion products are not available to enmeshed MB in the vicinity of Fe0. 
3 Materials and methods 
3.1 Solutions 
The used MB was of analytical grade. The working solution was 10.0 mg L-1. The solutions 
were prepared by diluting a 1000 mg L-1 stock solution. 
3.2 Solid materials 
3.2.1 Metallic iron (Fe0) 
 5
The used Fe0 material was purchased from iPutech (Rheinfelden, Germany). The material is 
available as fillings with a particle size between 0.3 and 2.0 mm. Its average elemental 
composition as specified by iPutech was: C: 3.52 %; Si: 2.12 %; Mn: 0.93 %; Cr: 0.66 %. The 
material was used without any pre-treatment. Fe
131 
132 
133 
134 
135 
136 
137 
138 
139 
140 
141 
142 
143 
144 
145 
146 
147 
148 
149 
150 
151 
152 
153 
154 
155 
0 was proven a powerful discoloration agent 
for MB. In particular, discoloration agents are progressively generated in-situ [41,42]. 
Therefore, the discoloration capacity of used Fe0 can not be exhausted within the 
experimental duration (≤ 42 days). 
3.2.2 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. The particle size was between 2.0 and 4.0 mm. Sand was used as an 
adsorbent because of its worldwide availability and its common use as admixing agent in 
Fe0/H2O systems. The adsorption capacity of various sand samples for MB has been largely 
documented in the literature [39,40,43,44]. Unlike Fe0, the adsorption capacity of sand for 
MB can be exhausted within the experimental duration. 
3.3 MB discoloration 
3.3.1 Non-disturbed batch experiments 
Batch experiments without shaking were conducted in essay tubes for an experimental 
duration of 21 and 42 days. The batches consisted of 0.0 to 2.0 g of sand, 0.0 to 1.0 g of Fe0 
and 22 mL of a 10 mg L-1 MB solution. A reaction time of 21 d was selected to allow a MB 
discoloration efficiency larger than 50 % in the system with Fe0 alone. The extent of MB 
discoloration in investigated systems was characterized at laboratory temperature (about 22° 
C). Initial pH was ~8.2. After equilibration, up to 3 mL of the supernatant solutions was 
carefully retrieved (no filtration) for MB measurement (no dilution). 
3.3.2 Shaken batch experiments 
 6
156 
157 
158 
159 
160 
161 
162 
163 
164 
165 
166 
167 
168 
169 
170 
171 
172 
173 
174 
175 
176 
177 
178 
179 
180 
181 
The essay tubes were amended with materials and MB solution as described above and 
allowed to equilibrate on a rotary shaker at 75 rpm for 7 days. 
Each experiment was performed in triplicate and averaged results are presented. 
3.4 Analytical methods 
MB aqueous concentrations were determined by a Cary 50 UV-Vis spectrophotometer 
(Varian) at a wavelength of 664.5 nm. Cuvettes with 1 cm light path were used. The 
spectrophotometer was calibrated for MB concentrations ≤ 15 mg L-1. The pH value was 
measured by combined glass electrodes (WTW Co., Germany). 
3.5 Expression of experimental results 
After the determination of the residual MB concentration (C) in batch studies, the 
corresponding percent MB discoloration was calculated according to the following equation 
(Eq. 1): 
P = [1 - (C/C0)] * 100%      (1) 
where C0 is the initial aqueous MB concentration (10.0 mg L-1), while C gives the MB 
concentration after the experiment. The operational initial concentration (C0) for each case 
was acquired from a triplicate control experiment without additive material (so-called blank). 
This procedure was to account for experimental errors during dilution of the stock solution, 
MB adsorption onto the walls of the reaction vessels and all other possible side reactions 
during the experiments. 
4 Results and Discussion 
The initial pH value of the MB solution was 8.2. The final pH value varies between 7.9 and 
8.8 for all investigated systems (results not shown). With a variation of only one pH unit, the 
process of MB discoloration depends primarily on the availability (and amount) of Fe0 and 
sand. MB discoloration as used in this work is not interchangeable with MB removal as the 
DOC values were not characterized. However, from a pure thermodynamic perspective MB is 
not redox sensitive in Fe0/H2O systems [41,42] (see section 2). 
 7
4.1 Non-disturbed batch experiments 182 
183 
184 
185 
186 
187 
188 
189 
190 
191 
192 
193 
194 
195 
196 
197 
198 
199 
200 
201 
202 
203 
204 
205 
206 
207 
Figure 1 summarizes the results of non-disturbed batch experiments for 21 and 45 days. Fig. 
1a compares the extent of MB discoloration by sand (≤ 90 g L-1) and Fe0 (≤ 45 g L-1) after 21 
days. It is seen that, based on the used masses, Fe0 is a more efficient MB discoloration agent 
than sand. MB discoloration is levelled at about 60 % in the pure Fe0 system and at about 45 
% in the pure sand system. In both systems, a pseudo-equilibrium is observed due to the 
slowness of MB diffusion to the surface of the adsorbing material at the bottom of the essay 
tubes. In the pure sand system, the adsorption capacity might be achieved for a long enough 
experimental duration. In Fe0-containing systems on the contrary, the adsorption capacity can 
not be exhausted at the laboratory scale, because completed depletion of Fe0 should be 
achieved. 
Fig. 1b shows that mixing 90 g L-1 sand to Fe0 significantly increased the efficiency of the 
system only for low Fe0 loading (< 6 g L-1). For these Fe0 loadings the surfaces of Fe0 and 
sand compete for MB discoloration and the amount of in-situ produced iron oxides has not 
significantly impaired the affinity of the sand surface for MB molecules [39]. For higher Fe0 
mass loadings, sand addition even impaired the process of MB discoloration. Thus, 
Assumption 2 (MB discoloration by Fe0 is significantly retarded by the presence of sand)  is 
validated. This observation is rationalized at first glance by the fact that sand is a good 
adsorbent for MB [39,40,43,44]. Adsorption on sand occurs with a more rapid kinetics than 
Fe0 oxidation which produced MB scavengers. This seemingly surprising observation was 
further investigated. Results from Fig. 1a clearly disproves Assumption 1 (“MB discoloration 
by Fe0 is not affected by the presence of sand”). 
Fig. 1b also shows that lengthening the reaction time from 21 to 45 days yield to increased 
MB discoloration. MB discoloration is now levelled to about 80 % for both systems. It is 
essential to recall that the systems were not shaken and that the reactive mixture was located 
at the bottom of the essay tubes. In this manner, gravitational diffusion was the sole driving 
 8
force for MB transport from the bulk solution to the adsorptive particles. Gravitational 
diffusion is induced by concentration gradient due to MB adsorption and/or co-precipitation at 
the bottom of the essay tubes. In the pure Fe
208 
209 
210 
211 
212 
213 
214 
215 
216 
217 
218 
219 
220 
221 
222 
223 
224 
225 
226 
227 
228 
229 
230 
231 
232 
233 
0 systems particle cementation was observed at 
the end of the experiments both after 21 and 45 days, with the cake after 45 days being more 
harder than that after 21 days. In Fe0/sand system not such cementation was observed. These 
observation suggested that the pure Fe0 system was already at a pseudo-steady state while the 
Fe0/sand system was still progressing to such a stage. Accordingly, it can be postulated that 
for longer experimental duration, Fe0/sand systems will be more efficient than pure sand 
systems. However, this will not result from MB adsorption onto sand but to continuous Fe0 
corrosion and scavengers generation. Accordingly, mixing Fe0 and sand is an efficient tool to 
sustain Fe0 corrosion [10,15]. 
The last important feature from Fig. 1 is that while using essay tubes in non-disturbed batch 
experiments, no pseudo-equilibrium was achieved even after 45 days. Accordingly, no effort 
was made to model the results using available equilibrium models (e.g. Freundlich, Langmuir, 
Temkin). Efforts to model Fe0 removal results by these models are faulty because Fe0 is 
supposed to react until complete depletion. Therefore, any equilibrium model should take the 
intrinsic reactivity of used Fe0 and its reaction kinetics into account. No such a study could be 
found in the literature. In contrary, the literature contain many studies characterizing Fe0 
materials by their adsorption capacity (e.g. in mg contaminant per g Fe0) based on result of 
short term batch experiments [1,5]. Usually these experiments are performed under mixing 
conditions (mixing type, mixing intensity) which are no relevant for experimental situations 
[45]. Recent works have shown that while using the experimental design of the present study, 
the shaking intensity should not exceed 50 rpm [23,41,42] to be relevant for field situations. 
The next section presents results of experiments obtained when the systems presented above 
were shaken at 75 rpm for 7 days (1 week). This shaking intensity (> 50 rpm) was selected to 
increase the probability to observe difference with non-disturbed conditions within 7 days. 
 9
4.2 Shaken batch experiments 234 
235 
236 
237 
238 
239 
240 
241 
242 
243 
244 
245 
246 
247 
248 
249 
250 
251 
252 
253 
254 
255 
256 
257 
258 
259 
Figure 2 summarizes the results of MB discoloration in the three investigated systems, shaken 
at 75 rpm for 7 days. Figure 2 clearly shows that the pure Fe0 system was the most efficient, 
followed by the “Fe0i + sand” system. The observation that MB is continuously adsorbed and 
co-precipitated by iron corrosion products is confirmed by the course of the line “Fe0i”. The 
least efficient system was “Fe0 + sandi”. This corresponds to the system with the smallest Fe0 
loading (5 g L-1) and the largest sand loading (up to 90 g L-1). It is important to notice that 
parallel “Fe0i” and “sand + Fe0i” systems contained exactly the same amount of Fe0 but “sand 
+ Fe0i” additionally contains 90 g L-1 sand. For Fe0 mass loadings larger that 11 g L-1, “Fe0” 
was more efficient than “sand + Fe0i”. Thus the validity of Assumption 2 is confirmed by 
shaken experiments. 
These results from Fig. 2 clearly show that the presence of sand impairs the accessibility of 
MB to the vicinity of Fe0 particle where absorption and co-precipitation occurs. It should be 
kept in mind that in-situ iron oxide coating at the surface of sand impair MB discoloration in 
two ways: (i) decreasing the affinity of sand surface for MB [39], and (ii) consuming “free” 
corrosion products with would have co-precipitated MB. At the end of the experiment minor 
compaction was observed in some reaction vessels for the Fe0/sand system. As a rule, when 
compaction become significant the access of sand and Fe0 surface is limited and MB 
discoloration is impaired. Accordingly, while mixing Fe0 and sand a balance most be found 
between (i) “decreased reactive surface” because of sand addition and (ii) “sustained Fe0 
reactivity” by virtue of sand addition.  
4.3 Removal mechanisms 
The use of adsorbing agents to accumulate reducible species in the vicinity of Fe0 media has 
been extensively reported in the literature [5]. However, the driving force for the transfer of 
adsorbed species to the Fe0 surface has not been identified/reported [46]. The discussion 
above has recalled that in-situ coating of the Fe0 surface can even impair adsorptive 
 10
accumulation. However, when accumulation is favourable, adsorbed species are more or less 
attached to the adsorbents (including sand and iron oxides). Adsorbed reducible species will 
not readily migrate to the Fe
260 
261 
262 
263 
264 
265 
266 
267 
268 
269 
270 
271 
272 
273 
274 
275 
276 
277 
278 
279 
280 
281 
282 
283 
284 
0 surface, but could be easily chemically transformed in the 
adsorbed state by FeII and H2 from continue iron corrosion. Reduction by FeII and H2 is a 
chemical reaction while reduction by Fe0 is an electrochemical reaction [47]. In other words, 
although iron corrosion is an electrochemical process, contaminant reduction in Fe0/H2O 
system is not induced by electrons from the metal body (Fe0). This has been demonstrated for 
example by Odziemkowski et al. [48], Lavine et al. [49], Jiao et al. [50] and Ghauch et al. 
[51]. Therefore, the reported efficiency of Fe0/sand system for contaminant reduction 
corroborates the adsorption co-precipitation concept [46,52-54], arguing that direct reduction 
(electrons from Fe0) does not play any significant role in the process of contaminant 
reduction. 
5 Concluding remarks 
This study has investigated the effect of mixing granular iron with sand for methylene blue 
(MB) discoloration. Three ranges of sand mass loading were tested: (i) absence of sand (pure 
Fe0), (ii) limited amount of sand (11 < [sand] (g L-1) < 45, (iii) abundance of sand (> 50 g L-1). 
Results showed that the kinetics of mass transfer, the relative amounts of Fe0 and sand all 
affect the efficiency of Fe0/sand mixtures for MB discoloration. The results achieved herein 
are purely qualitative and suggest that more research it needed with other contaminants in 
batch and column experiments before accumulated data can be converted into useful 
knowledge for the further development of the Fe0 technology. For example, based of the 
findings that (i) anionic (e.g. Acid Orange-7, Congo Red, Eriochrome Black T, Rose Bengal) 
and cationic (e.g. Janus Green, Rhodamine 6G, Toluidine Blue) dyes exhibit very different 
adsorption affinity for iron oxides [40,55] and (ii) metal oxide coated sands are worse 
adsorbents for MB than ‘virgin’ sands [39], working in parallel with organic anionic and 
 11
cationic dyes will enable a rapid characterization of sand on contaminant removal in Fe0/H2O 
systems. 
285 
286 
287 
288 
289 
290 
291 
292 
293 
294 
295 
296 
297 
298 
299 
300 
301 
302 
303 
304 
305 
306 
307 
308 
309 
310 
This study clearly shows that in delaying particle compaction, Fe0 “dilution” with sand is 
certainly beneficial for the long-term efficiency of Fe0 remediation systems. Accordingly, for 
the long term, mass transfer limitations are compensated by sustained reactivity. Practitioners 
of Fe0 technology have already taken advantage of this phenomenon [1,2,11,13]. However, 
available data were gained in a very pragmatic approach. With the recent establishment of the 
equation of a Fe0 bed [20] and the proper consideration of the volumetric expensive nature of 
iron corrosion [19,56,57], more research is needed to determine more precisely the conditions 
for a non site specific Fe0/sand system design. The present study has not attempted to 
introduce any mathematical model of the investigated systems which could account for 
example for the mass of material. Rather, the study is regarded as preliminary guidelines for 
the optimisation of filtration Fe0/H2O systems. 
Acknowledgements 
Thoughtful comments provided by Richard A. Crane (University of Bristol/UK) on the 
revised manuscript are gratefully acknowledged. 
References 
[1] G. Bartzas, K. Komnitsas, Solid phase studies and geochemical modelling of low-cost 
permeable reactive barriers, J. Hazard. Mater. 183 (2010) 301–308. 
[2] L. Li, C.H. Benson, Evaluation of five strategies to limit the impact of fouling in 
permeable reactive barriers, J. Hazard. Mater. 181 (2010) 170–180. 
[3] 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. 
[4] M. Gheju, Hexavalent chromium reduction with zero-valent iron (ZVI) in aquatic systems, 
Water Air Soil Pollut. 222 (2011) 103–148. 
 12
[5] ITRC (Interstate Technology & Regulatory Council) Permeable Reactive Barrier: 
Technology Update. PRB-5. Washington, D.C.: Interstate Technology & Regulatory 
Council, PRB: Technology Update Team. www.itrcweb.org (2011) (access: 
29.04.2012). 
311 
312 
313 
314 
315 
316 
317 
318 
319 
320 
321 
322 
323 
324 
[6] A. Hussam, Contending with a development disaster: SONO filters remove arsenic from 
well water in Bangladesh, Innovations 4 (2009) 89–102. 
[7] M.I. Litter, M.E. Morgada, J. Bundschuh, Possible treatments for arsenic removal in Latin 
American waters for human consumption, Environ. Pollut. 158 (2010) 1105–1118. 
[8] C. Noubactep, Metallic iron for safe drinking water worldwide, Chem. Eng. J. 165 (2010) 
740–749. 
[9] D.E. Giles, M. Mohapatra, T.B. Issa, S. Anand, P. Singh, Iron and aluminium based 
adsorption strategies for removing arsenic from water, J. Environ. Manage. 92 (2011) 
3011–3022. 
[10] C. Noubactep, Metallic iron for safe drinking water production, Freiberg Online Geol. 27 
(2011) 38 pp, ISSN 1434-7512. (www.geo.tu-freiberg.de/fog). 325 
326 
327 
328 
329 
330 
331 
332 
333 
334 
335 
[11] S.F. O´Hannesin, R.W. Gillham, Long-term performance of an in situ "iron wall" for 
remediation of VOCs, Ground Water 36 (1998) 164–170. 
[12] Y.H. Kim, S.-O. Ko, H.C. Yoo, Simultaneous removal of tetrachlorocarbon and 
chromium(VI) using zero valent iron, J. Korean Soc. Environ. Eng. 24 (2002) 1949–
1956. 
[13] J.F. Devlin, J. Patchen, The effect of diluting granular iron with a non-reactive porous 
medium on contaminant transformation rates. 5th Joint Conference of the IAH-CNC 
and the Canadian Geotechnical Society (CGS), October 24–27, Quebec City (2004). 
[14] D.-I. Song, Y.H. Kim, W.S. Shin, A simple mathematical analysis on the effect of sand 
in Cr(VI) reduction using zero valent iron, Korean J. Chem. Eng. 22  (2005) 67–69. 
 13
336 
337 
338 
339 
340 
341 
342 
343 
344 
345 
346 
347 
348 
349 
350 
351 
352 
353 
354 
355 
356 
357 
358 
359 
360 
361 
[15] E. Bi, J.F. Devlin, B. Huang, Effects of mixing granular iron with sand on the kinetics of 
trichloroethylene reduction, Ground Water Monit. Remed. 29  (2009) 56–62. 
[16] S. Ulsamer, A model to characterize the kinetics of dechlorination of tetrachloroethylene 
and trichloroethylene by a zero valent iron permeable reactive barrier, Master thesis, 
Worcester Polytechnic Institute (2011) 73 pp. 
[17] P.D. Mackenzie, D.P. Horney, T.M. Sivavec, Mineral precipitation and porosity losses in 
granular iron columns, J. Hazard. Mater. 68  (1999) 1–17. 
[18] C. Noubactep, S. Caré, B.D. Btatkeu K., C.P. Nanseu-Njiki, Enhancing the sustainability 
of household Fe0/sand filters by using bimetallics and MnO2, Clean – Soil, Air, Water 
40 (2012) 100–109. 
[19] C. Noubactep, S. Caré, Enhancing sustainability of household water filters by mixing 
metallic iron with porous materials, Chem. Eng. J. 162 (2010) 635–642. 
[20] C. Noubactep, S. Caré, Designing laboratory metallic iron columns for better result 
comparability, J. Hazard. Mater. 189  (2011) 809–813. 
[21] P. Westerhoff, J. James, Nitrate removal in zero-valent iron packed columns, Wat. Res. 
37  (2003) 1818–1830. 
[22] D.I. Kaplan, T.J. Gilmore, Zero-valent iron removal rates of aqueous Cr(VI) Measured 
under flow conditions, Water Air Soil Pollut. 155 (2004) 21-33. 
[23] A.-M. Kurth, Discoloration of Methylene Blue by Elemental Iron - Influence of the 
Shaking Intensity, Bachelor thesis, Universität Göttingen  (2008) 45 pp. 
[24] M.R. Powell, W.R. Puls, K.S Hightower, A.D. Sebatini, coupled iron corrosion and 
chromate reduction: Mechanisms for subsurface remediation, Environ. Sci. Technol. 29  
(1995) 1913–1922.  
[25] D.W. Blowes, C.J. Ptacek, J.L. Jambor, In-situ remediation of Cr(VI)-contaminated 
groundwater using permeable reactive walls: laboratory studies, Environ. Sci. Technol. 
31 (1997) 3348–3357. 
 14
362 
363 
364 
365 
366 
367 
368 
369 
370 
371 
372 
373 
374 
375 
376 
377 
378 
379 
380 
381 
382 
383 
384 
385 
[26] R.M. Powell, R.W. Puls, Proton generation by dissolution of intrinsic or augmented 
aluminosilicate minerals for in situ contaminant remediation by zero-valence-state iron, 
Environ. Sci. Technol. 31 (1997) 2244–2251. 
[27] Y.J. Oh, H. Song, W.S. Shin, S.J. Choi, Y.H. Kim, Effect of amorphous silica sand on 
removal of chromium(VI) by zero-valent iron, Chemosphere 66 (2007) 858–865. 
[28] J. Guo, D.J. Jiang, Y. Wu, P. Zhou, Y.Q. Lan, Degradation of methyl orange by Zn(0) 
assisted with silica gel, J. Hazard. Mater. 194 (2011) 290–296. 
[29] M. Edwards, M. Benjamin, Adsorptive filtration using coated sand: A new approach for 
treatment of metal-bearing wastes, J. Water Pollut. Control Fed. 61  (1989) 1523–1533. 
[30] D. Dermatas, X. Meng, Removal of As, Cr and Cd by adsorptive filtration, Global Nest: 
the Int. J. 6 (2004) 73–80. 
[31] J.P. Gould, The kinetics of hexavalent chromium reduction by metallic iron, Water Res. 
16 (1982) 871–877. 
[32] 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. 196 (2011) 131–138. 
[33] 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 Concrete Res. 38 
(2008) 1079–1091. 
[34] P.T. Hang, G.W. Brindley, Methylene blue absorption by clay minerals. Determination 
of surface areas and cation exchange capacities (clay–organic studies XVIII), Clay Clay 
Min. 18 (1970) 203–212. 
[35] S. Fityus, D.W. Smith, A.M. Jennar, Surface area using methylene blue adsorption as a 
measure of soil expansivity. Geo2000 Conference (on CD), Australia (2000). 
 15
[36] A.A. Attia, B.S. Girgis, N.A. Fathy, Removal of methylene blue by carbons derived from 
peach stones by H
386 
387 
388 
389 
390 
391 
392 
393 
394 
395 
396 
397 
398 
399 
400 
401 
402 
403 
404 
405 
406 
407 
408 
409 
410 
411 
3PO4 activation: Batch and column studies, Dyes and Pigments 76 
(2008) 282–289. 
[37] R.L. Frost, Y. Xi, H. He, Synthesis, characterization of palygorskite supported zero-
valent iron and its application for methylene blue adsorption, J. Colloid Interface Sci. 
341 (2010) 153–161. 
[38] P.G. Tratnyek, T.E. Reilkoff, A. Lemon, M.M. Scherer, B.A. Balko, L.M. Feik, B. 
Henegar, Visualizing redox chemistry: Probing environmental oxidation-reduction 
reactions with indicator dyes, Chem. Ed. 6 (2001) 172–179. 
[39] G. Mitchell, P. Poole, H.D. Segrove, Adsorption of methylene blue by high-silica sands. 
Nature 176 (1955) 1025–1026. 
[40] C. Varlikli, V. Bekiari, M. Kus, N. Boduroglu, I. Oner, P. Lianos, G. Lyberatos, S. Icli, 
Adsorption of dyes on Sahara desert sand, J. Hazard. Mater. 170 (2009) 27–34.  
[41] C. Noubactep, Characterizing the discoloration of methylene blue in Fe0/H2O systems, J. 
Hazard. Mater. 166 (2009) 79–87. 
[42] C. Noubactep, A.-M.F. Kurth, M. Sauter, Evaluation of the effects of shaking intensity 
on the process of methylene blue discoloration by metallic iron, J. Hazard. Mater. 169  
(2009) 1005–1011. 
[43] S. Chakrabarti, B.K. Dutta, On the adsorption and diffusion of methylene blue in glass 
fibers, J. Colloid Interf. Sci. 286  (2005) 807–811. 
[44] S.B. Bukallah, M.A. Rauf, S.S. AlAli, Removal of methylene blue from aqueous solution 
by adsorption on sand, Dyes and Pigments 74 (2007) 85–87. 
[45] C. Noubactep, Investigating the processes of contaminant removal in Fe0/H2O systems, 
Korean J. Chem. Eng. (2012) doi: 10.1007/s11814-011-0298-8. 
[46] C. Noubactep, A critical review on the mechanism of contaminant removal in Fe0–H2O 
systems, Environ. Technol. 29  (2008) 909–920. 
 16
412 
413 
414 
415 
416 
417 
418 
419 
420 
421 
422 
423 
424 
425 
426 
427 
428 
429 
430 
431 
432 
433 
434 
435 
[47] F. Togue-Kamga, K.B.D. Btatkeu, C. Noubactep, P. Woafo, Metallic iron for 
environmental remediation: Back to textbooks, Fresenius Environ. Bull. 21 (2012), 
(Accepted 09.05.2012). 
[48] M.S. Odziemkowski, L. Gui, R.W. Gillham, Reduction of n-nitrosodimethylamine with 
granular iron and nickel-enhanced iron. 2. mechanistic studies, Environ. Sci. Technol. 
34 (2000) 3495–3500. 
[49] B.K. Lavine, G. Auslander, J. Ritter, Polarographic studies of zero valent iron as a 
reductant for remediation of nitroaromatics in the environment, Microchem. J. 70 
(2001) 69–83.
[50] Y. Jiao, C. Qiu, L. Huang, K. Wu, H. Ma, S. Chen, L. Ma, L. Wu, Reductive 
dechlorination of carbon tetrachloride by zero-valent iron and related iron corrosion, 
Appl. Catal. B: Environ. 91  (2009) 434–440. 
[51] 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. 
172 (2011) 1033–1044. 
[52] C. Noubactep, The fundamental mechanism of aqueous contaminant removal by metallic 
iron, Water SA 36  (2010) 663–670. 
[53] C. Noubactep, Aqueous contaminant removal by metallic iron: Is the paradigm shifting? 
Water SA 37 (2011) 419–426. 
[54] C. Noubactep, Metallic iron for water treatment: A knowledge system challenges 
mainstream science, Fresenius Environ. Bull. 20 (2011) 2632–2637. 
[55] B. Saha, S. Das, J. Saikia, G. Das, Preferential and enhanced adsorption of different dyes 
on iron oxide nanoparticles: A comparative study, J. Phys. Chem. C 115 (2011) 8024–
8033. 
 17
436 
437 
438 
439 
440 
441 
442 
443 
[56] C. Noubactep, E. Temgoua, M.A. Rahman, Designing iron-amended biosand filters for 
decentralized safe drinking water provision, Clean: Soil, Air, Water (2012) doi: 
10.1002/clen.201100620. 
[57] S. Caré, R. Crane, P.S. Calabro, A. Ghauch, E. Temgoua, C. Noubactep, Modelling the 
permeability loss of metallic iron water filtration systems. Clean – Soil, Air, Water 
(2012) doi: 10.1002/clen.201200167. 
 
 18
Figure 1 443 
444  
0 20 40 60 80 100
0
20
40
60
80
100
(a)
 Fe0, 21 days
 sand, 21 days
M
B
 d
is
co
lo
ra
tio
n 
/ [
%
]
material / [g L-1]
 445 
446  
0 10 20 30 40 50
0
20
40
60
80
100
(b)
 Fe0i, 45 days
 sand + Fe0i, 45 days
 Fe0i, 21 days
 sand + Fe0i,  21 days
P 
/ [
%
]
Fe0 / [g L-1]
 447 
448 
 19
Figure 2 448 
449  
0 20 40 60 80 100
0
20
40
60
80
 Fe0i
 sand + Fe0i
 Fe0 + sandi
P 
/ [
%
]
material / [g L-1]
 450 
451 
 20
Figure Captions 451 
452 
453 
454 
455 
456 
457 
458 
459 
460 
461 
462 
463 
464 
465 
466 
 
Figure 1: Evolution of the methylene blue (MB) discoloration (P in %) as a function of 
material loading in non-disturbed batch experiments for 21 and 45 days: (a) pure Fe0 and sand 
systems, (b) Fe0 and Fe0/sand systems. The Fe0/sand systems (Fig. 1b) content 90 g L-1 of 
sand and variable amounts of Fe0. The experiments were conducted with a 10 mg L-1 MB 
solution in graduated essay tubes containing of 22 mL of the MB solution. The lines are not 
fitting functions, they simply connect points to facilitate visualization. 
 
Figure 2:  Evolution of the methylene blue (MB) discoloration (P in %) as a function of 
material loading in shaken batch experiments (75 rpm) for 7 days. The experiments were 
conducted with a 10 mg L-1 MB solution in graduated essay tubes containing of 22 mL of the 
MB solution. The subscript “i” designates the material which mass loading varied. The other 
material is present in constant amount (5 g L-1 for Fe0 and 90 g L-1 for sand). The lines are not 
fitting functions, they simply connect points to facilitate visualization. 
 
 21