Characterizing the reactivity of metallic iron in Fe0/UVI/H2O systems by long-term column experiments 1 2 3 4 C. Noubactep Angewandte Geologie, Universität Göttingen, Goldschmidtstraße 3, D - 37077 Göttingen, Germany. 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Tel.: +49 551 39 3191, Fax.: +49 551 399379, e-mail: cnoubac@gwdg.de Abstract Aqueous iron corrosion has been identified as a powerful process for environmental remediation and safe drinking water provision. The intrinsic reactivity of conventional micro- scale metallic iron (Fe0) has been mostly enhanced by the association of other metals (e.g. Cu, Ni, Pd) or by using nano-scale Fe0. However, the mechanisms responsible for increased reactivity are yet to be univocally elucidated. Therefore, it appears that basic knowledge on Fe0 intrinsic reactivity decreases as Fe0 technology advances. This work was performed to address the intrinsic reactivity of four selected conventional Fe0 materials in long-term column studies. Five systems were investigated. Each system was made up of two columns in series. The first column contains 5.0 g of a U-bearing rock mixed with sand. The second sand column contained 0.0 or 4.0 g of each Fe0. The extent of U fixation in the systems was used to characterize the reactivity of the Fe0 materials. Results showed a clearer differentiation among the three Fe0 materials which exhibited very closed extents of iron dissolution in 2 mM EDTA under dynamic conditions. The crucial importance of long-term column experiments to test the suitability of Fe0 for field applications is delineated. A new research trend is presented. Keywords: Column study, Material reactivity, Natural rock, Water treatment, Zerovalent iron. 1 1 Introduction 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 Science and technology have developed various efficient methods for safe drinking water provision, wastewater recycling, and groundwater remediation. These methods can be classified on the basis of biological, chemical, physical, and thermal principles [1]. The most current methods include adsorption, centrifugation, coagulation, crystallization, distillation, evaporation, flotation, ion exchange, oxidation, precipitation, sedimentation (and gravity separation), and reverse osmosis [1-7]. Among these methods, adsorption, ion exchange and reverse osmosis are excellent technologies, adsorption being the most affordable one [6,7]. The use of metallic iron (Fe0) for safe drinking water production has been discussed in the literature during the last 10 to 15 years [8-16]. Actually, detailed concepts for the application of Fe0 for safe drinking water production at several scales have been presented. In particular, Fe0 seems to be an ideal safe water production medium in emergency situations, for remote and in low-income communities [12,15]. For the realization of these concepts, a profound knowledge on Fe0 intrinsic reactivity under field conditions is needed. A purposeful use or modification of Fe0 for each application can only be based on such a profound knowledge. The present work is a continuation of an effort which objective is to establish a reliable tool to characterize Fe0 intrinsic reactivity under natural near conditions. Previous works have characterized the reactivity of 18 materials using batch EDTA-test [17,18]. The batch EDTA- test was significantly disturbed by the abundance of atmospheric corrosion products and was not suitable for powdered materials [18]. On the other hand, the long-term EDTA leaching column test could not really differentiate the reactivity of seven granular materials [19]. The shortcoming of the EDTA-test is that Fe0 materials are characterized by the extent of Fe dissolution which is not necessarily linearly coupled to contaminant removal. Another short term batch test accounting for this shortcoming has been developed [20]. This test is based on the extend of discoloration of a methylene blue solution by Fe0 in the presence of a well- 2 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 characterized MnO2 (MB-test). The MB-test suffers from the lack of a referential MnO2 material but could be satisfactorily used for internal purposes. Filtration on Fe0 beds is a deep-bed filtration. This means that contaminants are not only removed in the entrance zone or at a reaction front but in the whole bed [15,21]. Accordingly, characterizing the reactivity of Fe0 in column studies should not be limited to determine the time to contaminant breakthrough, but rather the reactivity of Fe0 in the beds. Based on this evidence, this work intentionally used small amounts of Fe0 and characterized the systems by the extent of U fixation in the bed (also see ref. [22]). The other peculiarity is that, not a U solution is piped into the columns, but U was leached in-situ from a natural rock by a natural spring water. The objective of this study was to further differentiate the reactivity of Fe0 materials from recent works by using near natural systems. For this purpose, U was leached from a natural rock by a spring water and the intrinsic reactivity of Fe0 materials was characterized as the extent of fixation of leached U. Preliminary works enable the optimisation of experimental conditions. Results delineate the importance of natural near systems for the characterization of Fe0 materials for practical use. A new experimental trend for laboratory experiments regarding the characterization of the suitability of Fe0 under natural conditions is introduced. 2 Materials and methods 2.1 Solid materials 2.1.1 U-bearing rock The used U-bearing rock was crushed and sieved at the Helmholtz-Zentrum Dresden- Rossendorf. The fraction 0.200 to 0.500 mm was used without any further pre-treatment. The rock contains around 0.6 % U and is further composed of: 74.16 % SiO2, 0.19% TiO2; 7.42 % Al2O3, 1.64 % Fe2O3, 0.03% MnO; 0.86 % MgO, 12.68 % CaO, 1.53 % Na2O, 1.45 % K2O, 0.04 % P2O5 and 0.18% SO3. The composition was determined by X-ray fluorescence (XRF) analysis. The EDX analysis (results not shown) revealed that the used rock is a multi- 3 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 mineralic one containing, among others, uraninite (UO2), arsenopyrite (FeSAs), and galena (PbS). Associations of U with arsenopyrite was also encountered. 2.1.2 Fe0 materials One scrap iron (ZVI1), and three commercially available Fe0 materials have been tested. The main characteristics of these materials are summarized elsewhere [19]. Before being used ZVI4 was crushed and sieved; the size fraction 1.0-2.0 mm was used without any further pre- treatment. The specific surface area of the materials varies between 0.043 and 0.63 m2 g-1. These data were compiled from the literature (Tab. 1). The materials were compared solely on the basis of the extent of U removal by the same initial mass of Fe0 (1.2 or 4.0 g) under similar experimental conditions. Apart from ZVI4, all other materials were used as obtained. Crushing and sieving ZVI4 aimed at working with materials of particle size relevant for field applications. The materials differ regarding their characteristics such as iron content, nature and proportion of alloying elements, and shape. The four used materials were selected from nine materials which were recently characterized by leaching with 0.2 M EDTA in column study [19]. In turn, the tested nine materials were selected from eighteen materials after a screening in batch experiments using the EDTA-test [17,18]. Both tests could not really differentiate the reactivity of ZVI1, ZVI3 and ZVI4. The reactivity of these three materials toward U removal from a natural rock was investigated in this study. For comparison the least reactive commercial Fe0 (ZVI2) was incorporated in this study. 2.1.3 Other materials Dolomite: The used dolomite mineral was crushed, sieved and the fraction 0.63 to 1.0 mm was used. The mineralogical composition (XRF) is: SiO2: 1.2%, TiO2: 0.03%; Al2O3: 0.4%, Fe2O3 0.6%, MgO: 20.24%, CaO: 30.94%, Na2O: 0.04%. Dolomite is a carbonate mineral; its dissolution increases the kinetics of U leaching because of the formation of stable complex 4 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 with carbonates [23,24]. The dolomite layer was placed after of the rock layer and thus its capacity in retarding U transport is investigated. Apatite: The used hydroxyapatite was a natural sample from a private collection (origin unknown). The material was crushed, sieved and the fraction 0.5 to 1.0 mm was used. Glass beads: The used glass beads were from Roth (Karlsruhe, Germany). Their particle size was 0.25 < d (mm) < 0.50. They were used as filling material. Their adsorption capacity was also tested in using three different volumes of glass beads in columns containing sand. 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. 2.2 Leaching solution The leaching solution was spring water from the Lausebrunnen in Krebeck (administrative district of Göttingen). Spring water was used as proxy for natural groundwater and was supplied weekly. 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 an initial pH 7.8. 2.3 Uranium leaching and immobilization Laboratory scale glass columns were operated in up-flow mode (Fig. 1). The influent solution was spring water (Lausebrunnen/Krebeck) and was pumped upwards from PE bottles using a peristaltic pump (Ismatec, ICP 24). Tygon tubes were used to connect inlet reservoir, pump, column and outlet. Ten glass columns (40 cm long, 2.6 cm inner diameter) were used in two series of experiments. The columns were mostly packed with sand (Fig. 1). The effective length, the bulk density and the porosity of the packed columns were not characterized as they were not necessary for the discussion of the results. The extent of U dissolution by water and the extent of its removal by selected materials were the sole targets. The experiments were performed at room temperature (22 ± 3 °C) in the ISOLAB at the Department of Physical 5 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 Chemistry. In individual experiment series, a stable flow rate was maintained throughout the experiment. Samples for analysis were collected at periodic intervals. 2.3.1 One-column experiments U rock and other materials (including Fe0) were packed into the same column (Fig. 1 - left). The rock was placed at the inlet, at 5.0 cm from the bottom on a layer of sand. The other material was placed at 5.0 cm from the outlet. 10 parallel experiments were performed with 3.0 g of the rock, 1.2 g of Fe0, 6.0 g of dolomite, 6.0 g of apatite and three different depths of glass beads (5.0, 10.0 and 20.0 cm – mass not measured) . The columns were fed with spring water at a constant flow rate of 19.0 mL/h. The experiments were stopped after 37 days when each column was leached by about 17.0 L of water. 2.3.2 Two-column experiments Based on the results of the one-column experiments, the two-column experiments were designed to better characterize the reactivity of tested Fe0 materials (Fig. 1 - right). Five parallel experiments were performed. The same mass of the rock (5.0 g) was placed in a first column and 4.0 g of each tested Fe0 was placed in the second column. In the reference system, the second column contained only sand (no Fe0). The experiments were stopped after 84 days when each column was leached by 24.0 L of water. The water flow rate was constant at 12.0 mL/h. 2.4 Analytical methods Analysis for U was performed by inductively coupled plasma mass spectrometry (ICP-MS) at the Department of Geochemistry (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). The electrodes were calibrated with five standards following a multi-point calibration protocol [25] and in agreement with the new IUPAC recommendation [26]. 6 3 Results and discussion 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 3.1 Expression of experimental results The mass (m) of leached U (mg) at any time (t) is calculated from the concentration of the effluent using Eq. (1): m = P.V (1) Where P is the U concentration (in mg/L) and V the volume (L). At the end of the experiment the total amount of leached U can be calculated by addition and the extent of U leaching by spring water deduced. Knowing the percentage of U in the natural rock (0.6 %), the maximal leachable mass (m0) of uranium can be calculated. The percentage (P) uranium leaching at each time is given by Eq. (2): P = 100 * m/m0 (2) At each time the amount of U leached in the reference system (Pref) can be set to 100 and the relative leaching percent (Prel) for all other systems deduced by Eq. (3): P* = Prel = 100 * P/Pref (3) Finally, the relative percent of U removal (Pfix) by each material is given by Eq. (4): Pfix = 100 - Prel (4) 3.2 Uranium removal in one-column experiments Table 2 and Figure 2 summarize the results of U leaching (reference system) and removal by glass beads, apatite, dolomite and the four Fe0 materials. It can be seen from Tab. 2 that the maximal U leaching corresponds to 10.43 mg of U. This value corresponds to 29 % of the total amount of U contained in the 3.0 g U rock. It is important to note that in all systems, the effluent U concentration after 37 days was larger than 350 μg/L (Fig. 2). That is more than 16 times the maximum contaminant level (MCL) of the World Health Organization (WHO MCL = 20 μg/L) [27]. This experimental fact suggests that instead of synthetic solutions, natural rocks and minerals can be leached by natural or relevant synthetic solutions and the leachates used as initial solution for decontamination experiments. The double advantage of this 7 approach is: (i) the leaching process of contaminant from natural phases can be accurately investigated/characterized, and (ii) contaminant removal is investigated with concentration nearer to the real world as with all possible synthetic solutions. Note that the used rock was a multi-mineralic one, making the leachates multi-contaminant systems. Presently, however, multi-contaminant systems have received less attention in the Fe 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 0 remediation community [28]. Fig. 2a shows clearly that the process of contaminant removal under natural near conditions is complex. There is no uniform trend for the evolution of U concentration. The elevated U release at the beginning of the experiment (up to day 12) could be attributed to the presence of fines and very reactive site at the surface. Based on relative steady state after day 20 it was decided at day 30 to stop the reaction. However, prolonged sojourn in the laboratory (“restricted access area”) was not possible from day 27 to 37 due to technical problems. Accordingly, the experiments were stopped at day 27 and the peak at day 37 is due to a 10- days-stagnation which resulted in increased U leaching. Table 2 shows no U removal in the systems with 5 cm and 10 cm glass beads. This both systems are very comparable to the reference system showing that the results were reproducible. The 6 % U removal in the column with 20 cm glass beads can be attributed to U precipitation in the glass beads-layer at the beginning of the experiment as the solution are necessarily over-saturated [24]. In fact, given the small size of glass beads, precipitation of U as U(VI) phases (e.g. as schoepite UO3.2H2O) is very likely. Table 2 also shows that used apatite and dolomite are not suitable for U removal. The suitability of apatite for U removal is well documented [29-31]. Fig. 2b shows the cumulative mass of uranium in selected systems representative for all investigated systems. It can be seen that the curves have the same allure and are very closed to each other. Remember that experiments aimed at differentiating materials ZVI1, ZVI3 and ZVI4 for their intrinsic reactivity using less reactive ZVI2 as negative reference. P* values 8 from Tab. 2 confirmed that ZVI2 is less reactive than the three other materials while a slight differentiation between (ZVI1, ZVI3) and ZVI4 was possible. These results are not satisfactorily. Accordingly, the system was modified to exploit the huge potential of this experimental tool. The water flow velocity was reduced from 19 to 12 mL/h, the Fe 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 0 mass was increased from 1.2 to 4.0 g and the mass of U rock was increased from 3.0 to 5.0 g. Additionally, the Fe0 and rock were packed in different columns (Fig. 1) and the experimental duration was lengthened. 3.3 Uranium removal in two-column experiments Table 3 and Figure 3 summarize the results of U leaching (reference system) and removal by four Fe0 materials in the two-column experiment mode. It is clear from Fig. 3 a and b, that a differentiation of ZVI1 and ZVI3 is not possible. ZVI4 is clearly the most reactive material. This observation is confirmed by Pfix values from Tab. 3 (83.4 % U removal). Accordingly the order of increasing reactivity for the Fe0 materials is: ZVI2 < ZVI1 ≅ ZVI3 < ZVI4. It should be noted that in this experimental procedure, the U concentration was still higher than 20 μg/L (WHO MCL value) after 70 days. On the contrary, in Fe0 systems the U concentrations were lower than 15 μg/L after 55 days (see Fig. 4). This observation corroborates the suitability of long term column experiment with natural rocks and minerals for more reliable data. In such experiments, the natural material could be initially washed with the used water to eliminate fines at the surface and avoid the artificial peak at the beginning of the experiment. Alternatively, the first pore volumes (e.g. 10 or 20) can be discarded and the Fe0 characterization starts when the initial peak is over. As a rule of thumb it can be considered that any natural rock or mineral from which relevant contaminants could be leached for a long time (e.g. 2 months) at concentrations larger to the MCL value is suitable for this type of experiment. To optimise the experimental conditions, the material particle size and the used amount should be carefully tested. Testing a large number of materials under such conditions could enable the establishment of a data base 9 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 which could be used to bridge the huge gap between short term laboratory experiments and long-term processes on the field. 3.3 Discussion Since the discovery of Fe0 materials as water treatment medium [32] several groups of Fe- based materials have been tested for their suitability for field application [17,33-42]. However, apart from Reardon [33,39] and Noubactep et al. [17,18] all Fe0 characterization attempts were based on the extent of the removal of selected contaminants by the materials. Given that large variability of experimental conditions [18,22,35], available results are not really comparable. Furthermore, there is still no single standard test (or proposition therefore) for the characterization of the intrinsic reactivity of used materials. This situation is not acceptable, efforts should be directed at developing such tests. The objective being to generate the chain of knowledge on which the promising technology of using Fe0 for water treatment will be further developed. The next point will be to characterize materials and group them for individual purposes. For example, Fe0 for drinking water, Fe0 for wastewater, Fe0 for carbonate-rich groundwater, or Fe0 for MnO2-containing soils. The objective should not be to select the most reactive material but the material that is appropriate for a given objective (site-specific). For example a material suitable for carbonate-rich groundwater could be too reactive for a MnO2-containing soil because the reactivity is sustained by MnO2 [43-45]. Similarly, plating a Fe0 could be superfluous or even disadvantageous at certain sites. In other words, plating Fe0 to generate bimetallic systems or even composites should be justified by field requirements. For this purpose a better material characterization is needed. After the development of a two simple tests for the characterization of intrinsic Fe0 reactivity using a diluted EDTA solution [17-19], the present study has used a natural near system to further differentiate the reactivity of Fe0. It is shown that the proper characterization of the reactivity of Fe0 materials is necessarily time consuming. In fact the 30 day-experiment was 10 not really conclusive. A clear conclusion could only be given after the 84 day-experiments. It should be recalled that the 62 day-column-experiment with EDTA [19] was not sufficient to differentiate the reactivity of Fe 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 0 tested here. It appears that an attitude change is necessary to overcome the current impasse characterized by the intensive production of non comparable data. It has already been reiterated that all current data for the characterization of Fe0 materials (surface area, elemental composition, surface charge, surface roughness, surface state…) are not sufficient to evaluate/predict the intrinsic reactivity of Fe0 materials [18]. Only a purposeful characterization of available materials will enable the manufacture of new efficient materials. On this path, efforts should be directed at obtaining particularly corrosive materials which are necessarily composite containing for instance over-proportional proportions of corrosive alloying element like S or Mn. It is likely, that such material with electrochemical cells in the whole body will be more reactive in the long term than bimetallic or multi- metallic coatings. The fundamental suitability of Fe0 composites for water treatment and their superiority over conventional Fe0 materials has been demonstrated by Hussam et al. [9,46]. These authors replaced conventional iron shavings/fillings by a porous composite iron material and could transform a highly efficient but not sustainable system (3-Kolshi filter) into a highly functional and sustainable system (SONO filter) [46]. 4 Concluding remarks The future of Fe0 as water treatment material, and thus the fate for the Fe0 remediation technology as a whole, depends on the ability of researchers to traceably define/characterize used materials. Since used materials considerably affect the success of the systems, there is a dire need to optimize material characterization. This basic operation has the advantage of minimizing system failures at the origin. This work has presented a possible way for the characterization of Fe0 intrinsic reactivity in long-term column studies using natural near conditions. A beneficial side-effect of the 11 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 presented concept is the investigation of the process of contaminant release from natural rocks and minerals and the investigation of contaminant in multi-contaminant systems at concentrations close to those encountered in nature. Using such systems will ease the bridging on the huge gap between short term laboratory experiments and field observations [47]. Finally, a full commitment of Fe0 remediation operators with independent researchers is needed to enable applied research to develop relevant scientific models and management schemes based on sound laboratory results from long-term experiments. Acknowledgments The experiments with uranium were performed in the Department of Physical Chemistry (ISOLAB) of the University of Göttingen. Dr. Ralf. Köber (Institute Earth Science of the University of Kiel) kindly purchased the commercial ZVI samples. The scrap iron (Sorte 69) was kindly purchased by the branch of the MAZ (Metallaufbereitung Zwickau, Co) in Freiberg. Klaus Simon (Department of Geochemistry - Universität Göttingen) is acknowledged for the ICP uranium analyses. Mohammad A. Rahman (Angewandte Geologie - Universität Göttingen) is acknowledged for technical support. The work was supported by the Deutsche Forschungsgemeinschaft (DFG-No 626/2-2). Cited references [1] V.L. Snoeyink, D. Jenkins, Water Chemistry, John Wiley & Sons (1980) 480 pp. [2] T. Wigmans, Industrial aspects of production and use of activated carbons, Carbon 27 (1989) 13–22. [3] R.D. Vidic, M.T. Suidan, U.K. Traegner, G.F. Nakhla, Adsorption isotherms: illusive capacity and role of oxygen, War. Res. 24 (1990) 1187–1195. [4] S.J.T. Pollard, G.D. Fowler, C.J. Sollars, R. Perry, Low-cost adsorbents for waste and wastewater treatment: A review, Sci. Total Environ. 116 (1992) 31–52. 12 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 [5] M. Streat, J.W. Patrick, M.J. Camporro-Perez, Sorption of phenol and p-chlorophenol from water using conventional and noval activated carbons, Water Res. 29 (1995) 467– 472. [6] M.K. Doula, Simultaneous removal of Cu, Mn and Zn from drinking water with the use of clinoptilolite and its Fe-modified form, Water Res. 43 (2009) 3659–3672. [7] I. Ali, The quest for active carbon adsorbent substitutes: inexpensive adsorbents for toxic metal ions removal from wastewater. Sep. Purif. Rev. 39 (2010) 95–171. [8] 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 A35 (2000) 1021–1041. [9] 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. [10] 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 A42 (2007) 1879–1888. [11] C. Noubactep, A. Schöner, P. Woafo, Metallic iron filters for universal access to safe drinking water. Clean Soil Air Water 37 (2009) 930–937. [12] D.D.J. Antia, Sustainable zero-valent metal (ZVM) water treatment associated with diffusion, infiltration, abstraction and recirculation. Sustainability 2 (2010) 2988–3073. [13] A.M. Gottinger, Chemical-free arsenic removal from potable water with a ZVI-amended biofilter, Master thesis, University of Regina (Saskatchewan, Canada) (2010) 90 pp. [14] 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. [15] C. Noubactep, Metallic iron for safe drinking water worldwide. Chem. Eng. J. 165 (2010) 740–749. 13 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 [16] C. Noubactep, A. Schöner, Metallic iron: dawn of a new era of drinking water treatment research? Fresen. Environ. Bull. 19 (2010) 1661–1668. [17] C. Noubactep, G. Meinrath, P. Dietrich, M. Sauter, B. Merkel, Testing the suitability of zerovalent iron materials for reactive walls, Environ. Chem. 2 (2005) 71–76. [18] C. Noubactep, T. Licha, T.B. Scott, M. Fall, M. Sauter, Exploring the influence of operational parameters on the reactivity of elemental iron materials. J. Hazard. Mater. 172 (2009) 943–951. [19] C. Noubactep, Characterizing the reactivity of metallic iron in Fe0/EDTA/H2O systems with column experiments, Chem. Eng. J. 162 (2010) 656–661. [20] C. Noubactep, Characterizing the reactivity of metallic iron upon methylene blue discoloration in Fe0/MnO2/H2O systems. J. Hazard. Mater. 168 (2009) 1613–1616. [21] C. Noubactep, The fundamental mechanism of aqueous contaminant removal by metallic iron. Water SA 36 (2010) 663–670. [22] C. Noubactep, S. Caré, Designing laboratory metallic iron columns for better result comparability. J. Hazard. Mater. (2011) doi: 10.1016/j.jhazmat.2011.03.016. [23] D. Langmuir, Uranium solution-mineral equilibria at low temperatures with application to sedimentary ore deposits, Geochim. Cosmochim. Acta 42 (1978) 547–569. [24] M. Kalin, W.N. Wheeler, G. Meinrath, The removal of uranium from mining waste water using algal/microbial biomass. J. Environ. Radioact. 78 (2005) 151–177. [25] G. Meinrath, P. Spitzer, Uncertainties in determination of pH, Mikrochem. Acta 135 (2000) 155–168. [26] 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. 14 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 [27] WHO, Uranium in drinking-water. Background document for preparation of WHO Guidelines for drinking-water quality. Geneva, World Health Organization (WHO/SDE/WSH/03.04/118) (2003). [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] R. Gauglitz, M. Holterdorf, W. Franke, G. Marx, Immobilization of heavy metals by hydroxylapatite, Radiochim. Acta 58–59 (1992) 253–257. [30] F.G. Simon, V. Biermann, C. Segebade, M. Hedrich, Behaviour of uranium in hydroxyapatite-bearing permeable reactive barriers: investigation using 237U as a radioindicator. Sci. Tot. Environ. 326 (2004) 249–256. [31] F.G. Simon, V. Biermann, B. Peplinski, Uranium removal from groundwater using hydroxyapatite. Appl. Geochem. 23 (2008) 2137–2145. [32] G.W. Reynolds, J.T. Hoff, R.W. Gillham, Sampling bias caused by materials used to monitor halocarbons in groundwater, Environ. Sci. Technol. 24 (1990) 135–142. [33] E.J. Reardon, Anaerobic corrosion of granular iron: Measurement and interpretation of hydrogen evolution rates, Environ. Sci. Technol. 29 (1995) 2936–2945. [34] R.L. Landis, R.W. Gillham, E.J. Reardon, R. Fagan, R.M. Focht, J.L. Vogan, An examination of zero-valent iron sources used in permeable reactive barriers, 3rd International Containment Technology Conference (10-13 June 2001), Florida State University, Tallahassee. Orlando, FL. (2001) 5 pages. [35] R. Miehr, G.P. Tratnyek, Z.J. Bandstra, M.M. Scherer, J.M. Alowitz, J.E. Bylaska, Diversity of contaminant reduction reactions by zerovalent iron: Role of the reductate, Environ. Sci. Technol. 38 (2004) 139–147. [36] A.H. El-Shazly, A.A. Mubarak, A.H. Konsowa, Hexavalent chromium reduction using a fixed bed of scrap bearing iron spheres. Desalination 185 (2005) 307–316. 15 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 [37] O.X. Leupin, S.J. Hug, Oxidation and removal of arsenic (III) from aerated groundwater by filtration through sand and zero-valent iron, Wat. Res. 39 (2005) 1729–740. [38] O.X. Leupin, S.J. Hug, A.B.M. Badruzzaman, Arsenic removal from Bangladesh tube well water with filter columns containing zerovalent iron filings and sand, Environ. Sci. Technol. 39 (2005) 8032–8037. [39] E.J. Reardon, Zerovalent irons: Styles of corrosion and inorganic control on hydrogen pressure buildup, Environ. Sci. Tchnol. 39 (2005) 7311–7317. [40] M. Gheju, A. Iovi, Kinetics of hexavalent chromium reduction by scrap iron, J. Hazard. Mater. 135 (2006) 66–73. [41] M. Gheju, A. Iovi, I. Balcu, Hexavalent chromium reduction with scrap iron in continuous-flow system: Part 1: Effect of feed solution pH, J. Hazard. Mater. 153 (2008) 655–662. [42] M. Gheju, I. Balcu, Hexavalent chromium reduction with scrap iron in continuous-flow system. Part 2: Effect of scrap iron shape and size, J. Hazard. Mater. 182 (2010) 484– 493. [43] 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. [44] 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. [45] A. Ghauch, H. Abou Assi, S. Bdeir, Aqueous removal of diclofenac by plated elemental iron: Bimetallic systems, J. Hazard. Mater. 182 (2010) 64–74. [46] A. Hussam, Contending with a development disaster: SONO filters remove arsenic from well water in Bangladesh, Innovations 4 (2009) 89–102. [47] T.-H. Wang, M.-H. Li, S.-P. Teng, Bridging the gap between batch and column experiments: A case study of Cs adsorption on granite, J. Hazard. Mater. 161 (2009) 409–415. 16 408 409 17 Table 1. Elemental composition and specific surface area (SSA) of iron materials used in this study. n.d. = not determined. Modified from ref. [19]. The materials are range in increasing order of the carbon content to ease comparison with previous work. 409 410 411 412 ZVI Element (%) SSA C Si Mn Cr Mo Ni Fe (m2/g) ZVI4 1.96 0.12 0.09 0.003 n.d. <0.001 86.3 0.63 ZVI2 3.39 0.41 1.10 0.34 n.d. 0.088 91.5 0.043 ZVI1 3.52 2.12 0.93 0.66 n.d. n.d. 99.8 0.29 ZVI3 3.13 2.17 0.36 0.077 n.d. 0.056 96.7 0.50 413 414 18 Table 2: Extent of uranium fixation by apatite, dolomite, and Fe0 materials in one-column experiments. It can be seen that the column containing 20 cm of glass beads (Glass beads 3) could significantly removed U. The extent of U removal by Fe 414 415 416 417 418 419 420 0 varies from 10.0 to 16.0 %. 3 g of U-mineral was used and each column was flooded by a volume VT of spring water. mf is the leached mass of U, P is the percent leaching relative to the total amount of U in the mineral and P* is the relative percent of U leaching. The sand column is the reference system. System Material VT mf P P* (g) (L) (mg) (%) (%) Ref. 0.0 17.32 10.43 29.0 100.0 Glass beads 1 5 cm 17.46 11.32 31.4 108.6 Glass beads 2 10 cm 17.48 10.56 29.3 101.3 Glass beads 3 20 cm 18.25 9.82 27.3 94.2 Apatite 6.0 17.67 9.77 27.1 93.7 Dolomite 6.0 17.16 9.63 26.7 92.3 ZVI1 1.2 16.96 9.05 25.1 86.8 ZVI2 1.2 17.40 9.39 26.1 90.0 ZVI4 1.2 17.96 8.77 24.4 84.1 ZVI3 1.2 18.10 9.01 25.0 86.4 421 422 19 Table 3: Extent of uranium fixation by Fe0 materials in two-column experiments. The extent of U removal by Fe 422 423 424 425 426 427 0 varies from 37.3 to 83.4 %. 5 g of U-mineral was used and each column was flooded by a volume VT of natural water. mf is the leached mass of U, P is the percent leaching relative to the total amount of U in the mineral and P* is the relative percent of U leaching. The sand column is the reference system. 4.0 g of Fe0 was used in each system. System VT mf PU P* Pfix (L) (mg) (%) (%) (%) Ref. 23.9 18.40 30.7 100.0 0.0 ZVI1 23.9 7.13 11.9 38.8 61.2 ZVI2 24.3 11.53 19.2 62.7 37.3 ZVI3 23.9 7.08 11.8 38.5 61.5 ZVI4 24.1 3.06 5.1 16.6 83.4 428 429 20 Figure 1: 429 430 431 Figure 1° (left) 432 433 21 Figure 1b (right) 433 434 435 436 437 22 Figure 2 437 0 5 10 15 20 25 30 35 40 0 400 800 1200 1600 (a) reference glass bead dolomite apatite metallic iron ur an iu m / [μg ] elapsed time / [days] 438 439 0 10 20 30 40 0 2000 4000 6000 8000 10000 12000 (b) reference glass bead dolomite apatite metallic iron ur an iu m / [μg ] elapsed time / [days] 440 441 442 23 Figure 3: 442 0 20 40 60 80 0 1000 2000 3000 4000 5000 6000 7000 (a) Reference ZVI1 ZVI2 ZVI3 ZVI4 ur an iu m / [μg ] elapsed time / [days] 443 444 0 20 40 60 80 0 2500 5000 7500 10000 12500 15000 17500 20000 (b) Reference ZVI1 ZVI2 ZVI3 ZVI4 ur an iu m / [μg ] elapsed time / [days] 445 446 24 Figure 4 0 15 30 45 60 75 90 0 10 20 30 40 50 60 WHO MCL Reference ZVI1 ZVI2 ZVI4 ZVI3 ur an iu m / [μg /L ] elapsed time / [days] 446 447 448 449 25 Figure Captions 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 Figure 1: Schematic diagram of the experimental design: one-column (left) and two-column experiments (right). For the one-column experiments used materials were Fe0 (1.2 g), dolomite (6.0 g), apatite (6.0 g) and glass beads (5.0, 10.0 and 20.0 cm). Figure 2: Representative trend of uranium release from the one-column systems for 37 days by tested materials: (a) variation of the extent of U release with time, and (b) cumulative uranium release (μg). The lines are not fitting functions, they simply connect points to facilitate visualization. Figure 3: Extent of uranium release from the two-column-systems for 84 days by tested Fe0 materials: (a) variation of the extent of U release with time, and (b) cumulative uranium release (μg). The lines are not fitting functions, they simply connect points to facilitate visualization. Figure 4: Comparison of the evolution of the residual U concentration in the five systems of the two- column experiment with the WHO MCL (20 μg/L). It is shown that up to day 75 the U concentration of the effluent from the reference system was too high. Up to this date, lower U concentration in other system is attributed to the reactivity of Fe0 materials. The lines are not fitting functions, they simply connect points to facilitate visualization. 26