fe(ii)-initiated reduction of hexavalent chromium in...
TRANSCRIPT
1
Fe(II)-initiated Reduction of Hexavalent Chromium in 2
Heterogeneous Iron Oxide Suspension 3
4
Jeongyun Choi 1 Yoojin Jung
2, and Woojin Lee
1* 5
6
1 Department of Civil and Environmental Engineering, Korea Advanced Institute of Science and 7
Technology, 373-1 Guseong-Dong, Yuseong-Gu, Daejeon 305-701, Korea 8
2 Department of Civil and Environmental Engineering, University of Delaware, Newark, DE 19716, 9
USA 10
11
*Corresponding author; phone: 82-42-869-3624; fax; 82-42-869-3610; e-mail: 12
14
15
1
Abstract 16
The characteristics of Fe(II)-initiated reduction of Cr(VI) in iron oxide suspensions were 17
investigated by conducting a series of kinetic experiments. A modified Langmuir-Hinshelwood 18
kinetic model was used to provide the better description of Cr(VI) reduction kinetics which were 19
believed to be occurred on the limited reactive site of reductant. The concentration of magnetite 20
concentration as well as Cr(VI) concentration, significantly affected the reaction kinetics of Cr(VI). 21
The reduction kinetics were improved with increasing magnetite and Cr(VI) concentration. Almost 22
95% of Cr(VI) reduction was achieved within 10 min at the condition of 8 g/L of magnetite and 80 23
mg/L of initial Cr(VI), respectively. The solution pH also affected the reaction rate in the range of 24
5.5 and 8.0 where a lower pH produced a faster reaction rate. The addition of Fe(II) on soil and 25
magnetite showed the capability of improving Cr(VI) reduction kinetics and their reduction kinetics 26
was also well described using a Langmuir-Hinshelwood kinetic model. The experimental results 27
obtained in this research clearly show the advantage of additional reductant for reducing Cr(VI), and 28
they can provide basic knowledge for the development of remediation technology for the treatment 29
of groundwater and soil contaminated with Cr(VI). 30
31
Keywords: Chromium(VI), Modified Langmuir-Hinshelwood kinetics, Magnetite, Reduction, Fe(II) 32
addition 33
34
35
36
2
1. Introduction 37
Chromium generally exists in four oxidation states (Cr(0), Cr(II), Cr(III), and Cr(IV)) in natural 38
and engineered environments, but the trivalent and hexavalent forms are predominant. Most studies 39
have focused on Cr(VI) because of its toxic, carcinogenic, and mutagenic characteristics [1]. In 40
contrast to the harmful characteristics of Cr(VI), Cr(III) that is immobile and insoluble in subsurface 41
environments has been known to be less toxic and even an essential nutrient for plants and animals 42
[2]. Therefore, the redox chemistry reducing Cr(VI) to Cr(III) has been applied to the development of 43
most remedial technologies to treat groundwater and soil contaminated with Cr(VI). Natural 44
attenuation and in-situ redox manipulation are examples of the application of redox chemistry to 45
remediate contaminated sites. Surface-bound natural organic matters (NOMs) and iron-bearing soil 46
minerals such as magnetite, green rust, and zero-valent iron significantly affect the reduction of 47
Cr(VI) [3,4,5]. 48
As one of recent remedial alternatives to expedite natural attenuation, in-situ redox 49
manipulation by injecting reducing agents such as Fe(II) and sulfide into monitoring wells has been 50
studied. The significant reduction of organic compounds has been observed in in-situ redox 51
manipulation system as shown in soil mineral suspensions with reductants [6]. It has been reported 52
that heterogeneous surface reactions cause reductive degradations of organic compounds. Kinetic 53
rate constants for reductive degradation were influenced by the density of reactive Fe(II) bound on 54
the mineral surfaces. Williams and Scherer [7] showed that Fe(II) bound on the mineral surfaces 55
results in enhanced reductive degradation of organic compounds. In contrast, Cr(VI) can be 56
stoichiometrically reduced in the aqueous Fe(II) solution without any solid surfaces (except under 57
high pH conditions) forming reactive precipitates e.g., (CrxFe1-x)(OH)3(s)) [8]. Based on previous 58
experimental results, the Cr(VI) reduction seems to be affected by both aqueous Fe(II) and reactive 59
solids in soil mineral suspensions. 60
Research has been conducted to investigate the reaction kinetics for the reduction of Cr(VI) in 61
soil mineral suspensions with and without Fe(II) addition. However, few research has been carried 62
3
out to fully characterize the reduction of Cr(VI) at the reactive surfaces of soil minerals. This 63
research has investigated the reaction mechanism for the reduction of Cr(VI) at reactive solid 64
surfaces and identified the effect of important physicochemical parameters on the reaction kinetics of 65
Cr(VI). The contribution of aqueous Fe(II) and Fe(II) adsorbed on iron-bearing soil minerals to the 66
reduction of Cr(VI) has been also estimated. Synthesized magnetite and Fe(II) have been used as 67
representative iron-bearing soil mineral and reducing agent, respectively. Nontronite and soil have 68
been used for comparison. 69
70
2. Materials and Methods 71
2.1. Chemicals, Soil Minerals, and Soil 72
Reagents or higher grade chemicals were used as received without further treatment. Cr(VI) 73
(K2Cr2O7, 99.5%, Kanto) and FeSO4·7 H2O (99%, Sigma-Aldrich) were used as a target contaminant 74
and a reductant, respectively. FeCl2·4H2O (99~102%, Wako) and Fe(NO3)3·9H2O (99%, Yakuri) 75
were used to synthesize magnetite. Biological buffers, N-(2-hydroxyethyl)-piperazine-N-2-76
ethanosulfunic acid (HEPES, 99.5%, Sigma), 2-(N-Morpholino)ethanesulfonic acid (MES, 99.5%, 77
Sigma), and a mixture of Tris(hydroxymethyl)aminomethane (99.8%, Sigma-Aldrich) and 78
Tris(hydroxymethyl)aminomethane hydrochloride (99%, Sigma) (TRIS buffer), were used to keep 79
the constant pH of aqueous solutions and suspensions. The following chemicals were used for 80
colorimetric analyses: 1,5-Diphenylcarbonohydrazide (Kanto), 3-(2-Pyridyl)-5,6-diphenyl-1,2,4-81
triazine-p,p΄-disulfonic acid monosodium salt hydrate (97%, Aldrich), (NH4)2Fe(SO4)2 (99.9%, 82
Sigma-Aldrich), H2SO4 (95%, Junsei), HNO3 (60%, Junsei), NaOH (96%, Junsei), Na2SO4 (99%, 83
Junsei), ammonium acetate (97%, Yakuri), and hydroxylammonium chloride (97%, Junsei). All 84
chemical reagents were prepared with deaerated deionized water (ddw) by deoxygenating 18 MΩ·cm 85
deionized water with 99.9% nitrogen for two hours. 86
Magnetite was synthesized by modifying the method developed by Taylor at al. [9] and 87
identified using the XRD analysis. The result of XRD was shown in Fig. 1 with the obtained d-space 88
4
vales of the synthesized magnetite which were comparable to the results of magnetite in powder 89
diffraction file. The magnetite was used for a day after the synthesis to preclude aging effect. The 90
preparation and pretreatment of nontronite (Nontrone, France) and soil samples (Seoul, Korea) were 91
conducted by following Lee and Batchelor’s methods [10]. 92
93
2.2. Experimental Procedures 94
Batch kinetic experiments were conducted in 20 mL amber glass vials with open-top cap and 95
silicon septum lined with PTFE film. Magnetite (0.13 g), nontronite (2.5 g), and soil (2 g) were 96
transferred to vials, and 2 mL of 1M HEPES solution was added to adjust suspension pH to 7.2. The 97
total volume of the suspension was 20 mL resulting in concentrations of 6.5, 125, and 100 g/L for 98
magnetite, nontronite, and soil, respectively. An aliquot amount of Cr(VI) stock solution (0.4 mL, 99
1000 mg/L) was spiked to the vials resulting in an initial Cr(VI) concentration of 20 mg/L. 100
Based on the results of the batch kinetic experiments, parametric studies were conducted to 101
identify the effect of important geochemical parameters on the Cr(VI) reduction kinetics. Magnetite 102
was chosen as a representative soil mineral. The effects of magnetite concentration, target compound 103
concentration, and pH were separately investigated. Each parametric study had the following 104
experimental conditions: (1) The concentration of magnetite was varied from 5 to 8 g/L under the 105
initial Cr(VI) concentration of 50 mg L-1
, (2) the concentration of Cr(VI) differed from 20 to 80 106
mg/L, and (3) the pH of magnetite suspensions was adjusted from 5.5 to 8.0 using biological buffers 107
(MES, HEPES, and TRIS, respectively). Based on the study by Alowitz and Scherer, the Cr(VI) 108
reduction by iron metal was dependent on the type of biological buffer [11]. But we think that a 109
biological buffer is better than an inorganic buffer such as phosphate buffer to maintain the pH in 110
this study. Cr(VI) reduction might be less interfered by a biological buffer compared to an inorganic 111
buffer since the chemicals in an inorganic buffer could react with iron and chromium. Sample 112
preparation procedures generally followed that for the batch kinetic experiments described above. 113
5
To identify the effect of Fe(II) addition on the kinetics of Cr(VI) reduction, Fe(II) stock 114
solution was added to magnetite suspensions before spiking Cr(VI) resulting in the desired ratios of 115
ferrous iron to chromate in the range of 0.6 – 3.0. The concentrations of magnetite and Cr(VI) were 116
6.5 g/L and 80 mg/L, respectively. The pH of the mixed suspension was kept constant at 7.2 with a 117
TRIS buffer. Batch kinetic experiments were also conducted to compare the effect of pH in 118
magnetite suspensions with and without Fe(II) in the pH range of 5.5 – 8.0 under the same 119
concentrations of magnetite and Cr(VI). 120
All vials were quickly and tightly capped, mounted on a tumbler, and then completely mixed at 121
7 rpm under room temperature (20 ± 0.5 ℃). After each reaction time, a desorbing agent (0.284g of 122
Na2SO4) was added into the vials to result in 0.1 M sulfate. The suspensions were vigorously mixed 123
on an orbital shaker for 1 day to promote the desorption of Cr(VI) from soil mineral surfaces [12] 124
and centrifuged at 2000 g for five minutes. The supernatant from each vial was used for the 125
measurement of Cr(VI) and iron concentrations in aqueous solution. All samples were prepared in 126
duplicate. Control samples were prepared to check losses due to sorption on walls of vials and 127
septum and volatilization. The control samples were prepared in the same way described above 128
excluding the addition of soil minerals and soil. 129
130
2.3. Analytical Procedures 131
The concentrations of Cr(VI) and iron (Fe(II) and total iron) in the aqueous phase were 132
measured by colorimetric analysis using UV/VIS spectrophotometer (Spectronic® GENESYS™ 2, 133
Thermo Electron Corporation). The Cr(VI) concentration was determined using the 134
diphenylcarbazide method at a wavelength of 540 nm [13]. The concentrations of Fe(II) was directly 135
measured using the ferrozine method at a wavelength of 562 nm [14], and the concentration of total 136
iron was measured as a form of Fe(II) at the same wavelength after reducing Fe(III) to Fe(II) by 137
adding a 10% hydroxylamine solution. The concentration of Fe(III) was calculated by subtracting the 138
Fe(II) concentration from the total iron concentration. 139
6
X-ray diffraction analysis performed using a RigaKu automated diffractometer using Cu Kα 140
radiation. Solids were separated by centrifugation at 2960 g for 5 min and carefully dried in an 141
anaerobic chamber. The samples were scanned between 2.1° to 70° 2θ with a scan speed of 1° 2θ 142
min-1
. 143
144
2.4. Data Treatment for the Reduction Kinetics of Cr(VI) by magnetite 145
Reductive degradation of Cr(VI) by iron-bearing soil minerals and soil were monitored for 146
one hour. Cr(VI) in the soil mineral and soil suspensions showed rapid disappearance initially 147
followed by slow disappearance or constant concentrations depending on the reductive capacity of 148
soil minerals and soil. This indicates that the Cr(VI) reduction kinetics cannot be properly explained 149
by a pseudo-first-order or zero-order rate law. A modified Langmuir-Hinshelwood kinetic model (Eq. 150
1) was used to properly describe Cr(VI) reduction kinetics. This model could give a proper 151
description of kinetic data since the reduction of Cr(VI) might occur on the finite reactive site of 152
magnetite surface [10]. The kinetic model shown below can be obtained by combining a material 153
balance equation in a batch reactor and a rate equation. 154
155
)(
)()(
0
)()(
0
)()(
1
)(
VICr
VICrVICrVICrVICrRCVICrVICr
CK
CCCpCpk
dt
dC
(1) 156
157
where CCr(VI) is the aqueous concentration of Cr(VI) at time t; k is the rate constant for the decay of 158
Cr(VI) at the reactive sites in time-1
; pCr(VI) is the partitioning factor to explain the effect of 159
partitioning of Cr(VI) among aqueous, gas, and solid (soil mineral + vial wall + septum liner) phases 160
assuming instantaneous equilibrium among the three phases; C0
RC is the initial reductive capacity of 161
magnetite for the Cr(VI) in concentration; C0
Cr(VI) is the aqueous concentration of Cr(VI) at time 162
zero; and K is the sorption coefficient of Cr(VI) with units of inverse concentration. 163
164
7
3. Results and Discussion 165
3.1. Reaction Kinetics for the Reduction of Cr(VI) by Iron-bearing Soil Minerals and Soil. 166
Figure 2 shows the reaction kinetics for the reduction of Cr(VI) by magnetite, nontronite, and 167
soil. Cr(VI) was completely degraded in 10 minutes in the magnetite suspension, while more than 168
90% of the initial Cr(VI) was recovered in the soil and nontronite suspensions at the last sampling 169
time. Magnetite having the smallest surface concentration (284.7 m2/L) shows the greatest kinetic 170
rate constant (2.06 min-1
). Although nontronite and soil have approximately 46 and 8 times greater 171
surface concentrations than magnetite, they have smaller rate constants (1.42 min-1
and 1.28 min-1
, 172
respectively). This indicates the surface concentration is not a unique geochemical factor 173
significantly affecting the reduction of Cr(VI) in soil and soil mineral suspensions. On the other hand, 174
the total iron content of magnetite including both Fe(II) (201 mg/g) and Fe(III) (512 mg/g) is 175
approximately 140 and 7 times greater than those of soil and nontronite. Fe(II) content of soil was 176
measured to be less than 1 mg/g, and most iron measured for nontronite was identified as Fe(III). 177
Therefore, the iron content (especially, the surface Fe(II) concentration) could be the more significant 178
factor explaining the great differences in the reduction kinetics of Cr(VI) in the solid suspensions. 179
Peterson et al. showed similar experimental results. Cr(VI) in groundwater was fully reduced to 180
Cr(III) by the soil containing a high concentration of magnetite, but Cr(VI) was rarely reduced by the 181
soil containing natural iron-bearing phyllosilicates [15]. The results obtained in this research and by 182
Peterson et al. [15] confirm the assumption that the reaction kinetics for the reduction of Cr(VI) is 183
significantly influenced by the content of Fe(II) on the surface of soil minerals and soil. The curves 184
fitted by the Langmuir-Hinshelwood kinetic model at Fig. 2 seem to properly explain the reduction 185
kinetics of Cr(VI) in soil mineral and soil suspensions. 186
187
3.2. Effect of Physicochemical Parameters on the Reduction Kinetics of Cr(VI) 188
Figure 3 shows the effect of magnetite concentration on the reduction of Cr(VI) in the 189
magnetite suspension. The removal of Cr(VI) increased 5 times as the content of magnetite increased 190
8
by 1.6 times. This is due to the limited active site and different activity of sites. Almost 95% of 191
reduced Cr(VI) was removed in 0.5 hour from the suspension of different magnetite concentrations, 192
and part of the remaining Cr(VI) was slowly degraded. The result is consistent with previous results 193
[4]. The early rapid removal of Cr(VI) followed by its gradual decrease has been reported to be 194
mainly due to the accumulation of Cr(III) monolayer and subsequent exhaustion of available Fe(II) at 195
the soil mineral surfaces. This assumption was also supported in this study by that the content of 196
Fe(II) in the magnetite concentration of 6.5 g/L was 14 times higher than the stoichiometric amount 197
required for the full Cr(VI) reduction, but 30% of Cr(VI) was still recovered in the aqueous phase. 198
This means that part of Fe(II) measured in the magnetite exists as structural Fe(II), which cannot be 199
easily used for fast kinetic reactions, or that the Cr(III) layer formed hampers additional reactions. 200
These processes can cause a significant deviation from the well-fitted pseudo first-order kinetics for 201
the reduction of Cr(VI) in the soil mineral suspensions. 202
The effect of Cr(VI) concentration on the reaction kinetics is shown in Fig. 4. Cr(VI) at C0
Cr(VI) 203
= 20 mg/L was completely reduced in 10 minutes and well fitted by the first-order rate law with 204
respect to Cr(VI) concentration (R2 = 0.94). As the initial concentration of Cr(VI) increased, 205
different reduction patterns were observed. Although a similar rapid reduction of Cr(VI) at early 206
sampling times was found at the initial Cr(VI) concentration of 50 and 80 mg/L, the reduction 207
kinetics of Cr(VI) under these higher concentrations was not well fitted by the first-order kinetics but 208
properly fitted by the modified Langmuir-Hinshelwood kinetic model. This implies that the reaction 209
kinetics for the reduction of Cr(VI) is not simply dependent on the Cr(VI) concentration but 210
dependent on both concentrations of Cr(VI) and reactive surfaces on magnetite. 211
The pH of magnetite suspensions significantly affected the reduction kinetics of Cr(VI) as 212
shown in Fig. 5. The reduction rates of Cr(VI) increased with the decrease of pH (from 8 to 5.5), 213
which is similar to the experimental result observed in the reduction of aqueous transition metal 214
species on the surfaces of Fe(II)-bearing oxides [3]. The Cr(VI) injected was fully removed under pH 215
6.5, but 30 and 52% of Cr(VI) were recovered at pH 7.2 and 8.0, respectively. The initial rapid drop 216
9
of Cr(VI) concentration was mainly caused by fast Cr(VI) adsorption to magnetite, followed by the 217
slow decrease of Cr(VI) mainly due to Cr(VI) reduction [16]. The rapid adsorption is mainly related 218
with the zero point charge (zpc) of magnetite surface. The zpc of magnetite has been reported in the 219
pH range of 6.4 to 6.85 [17]. If the pH of suspensions is lower or higher than the zpc of magnetite, 220
the magnetite surfaces can be charged positively or negatively. Cr(VI) (Cr2O72-
) may experience 221
different electrostatic forces from the magnetite surfaces depending on the pH, i.e. attraction under 222
the zpc and repulsion above the zpc. Cr(VI) adsorption, the first step for the reduction of Cr(VI) at 223
the surfaces, at pH 7.2 and 8.0 could be interrupted by electrostatic repulsion. Thus, pH is an 224
important factor that controls the reaction kinetics of Cr(VI). In addition, the reduction of Cr(VI) can 225
consume the H+ from solution based on the following equation [11]. Therefore, the reduction of 226
Cr(VI) could be favorable at the lower pH condition and was comparable with the results of this 227
study . 228
O2H3Fe2Cr(OH)10H3Fe2CrO 22
30
4 (2) 229
230
3.3. Reduction of Cr(VI) in the Suspensions of Soil minerals and Soil with Fe(II) 231
Figure 6 shows the reduction kinetics of Cr(VI) in nontronite, soil, and magnetite with Fe(II) 232
(Fe(II)/Cr(VI) = 1.0). The reduction kinetics of Cr(VI) in the aqueous Fe(II) solution and in the 233
magnetite suspension are also shown as controls for comparison. The relative concentrations of 234
Cr(VI) dropped at first sampling time and slowly approached 0.7 during the reaction time. The 235
reduction patterns of Cr(VI) in nontronite and soil with Fe(II) are very similar, however more Cr(VI) 236
was removed by soil than by nontronite. This may be due to the high content of organic matter 237
(0.76%) in soil sample whose influence on the reduction kinetics has been reported [18]. In contrast 238
to the reductive dechlorination of chlorinated organic compounds, more Cr(VI) was removed in the 239
aqueous Fe(II) solution than in the soil and nontronite suspensions with Fe(II) addition. Chlorinated 240
organic compounds have been known to be reductively degraded at the reactive surface sites, but no 241
significant degradation has been reported in homogeneous reductant solutions such as Fe(II) solution. 242
10
This result indicates that the reactive Fe(II) adsorbed on the soil and nontronite surfaces in 243
heterogeneous systems does not effectively reduce Cr(VI) as much as aqueous Fe(II) does in the 244
homogeneous system. 245
The reactivity of magnetite suspension was enhanced by the addition of Fe(II). More Cr(VI) 246
was reduced in the magnetite suspension with Fe(II) (C/C0 at 60 min = 0.2) than in homogeneous 247
Fe(II) solution (0.6) and in the magnetite suspension without Fe(II) (0.4), respectively. This indicates 248
that the reductive capacity of magnetite was enhanced by forming the reactive surface sites (i.e., 249
surface complexes/precipitates) and/or by regenerating the oxidized surface sites by the addition of 250
Fe(II). However, the amount of Cr(VI) removed in magnetite suspension with Fe(II) is far less than 251
that estimated under an assumption that the removal of Cr(VI) in the suspension with Fe(II) is the 252
summation of each removal from aqueous Fe(II) solution and magnetite suspension under the same 253
experimental conditions. Figure 6 shows the estimated Cr(VI) removal completed in 30 min. The 254
part of adsorbed Fe(II) may be oxidized to Fe(III) at the surfaces forming non-reactive complexes 255
and/or precipitates. No measurement has been conducted to identify these non-reactive chemical 256
species on the surfaces in this study. 257
Figure 7 shows the effect of pH on the reduction of Cr(VI) in the magnetite suspension with 258
Fe(II) in the range of suspension pH 5.5 to 8.0. A similar effect shown in magnetite suspension (Fig. 259
5) was also observed in the magnetite suspensions with Fe(II). The reduction rates of Cr(VI) 260
increased with the decrease of suspension pH in both suspensions with and without Fe(II). The 261
magnetite suspension with Fe(II) at pH 5.5 has the greatest reduction rate constant showing the full 262
reduction of Cr(VI) in one hour, followed by those at pH 6.5 > pH 7.2 > pH 8.0 > pH 7.2 without 263
Fe(II). Because of the zpc of magnetite, the surfaces of magnetite at pH 5.5, pH 6.5, and > pH 7.2 are 264
charged with negative (-O-), neutral (-OH), and positive (-OH2
+) surface species, respectively. This 265
significantly affects the behavior of Cr(VI) and Fe(II) in the magnetite suspension. Gregory et al. 266
have reported that Fe(II) adsorbed on magnetite rapidly increases as the suspension pH increases 267
from 6 to 8 [6]. While 30% of initial Fe(II) was adsorbed on the surfaces of magnetite at pH 6, 90% 268
11
of the initial Fe(II) was adsorbed at pH 6.5 in this study. Fe(II) was not detected in the aqueous 269
solution at pH 7.2 and 8. The higher reactivity of magnetite suspension with Fe(II) observed at lower 270
pH (pH 5.5 and 6.5) is due to the significant reduction of Cr(VI) by Fe(II) in aqueous solution and 271
Fe(II) adsorbed on magnetite surfaces, suggesting that the Cr(VI) reduction by aqueous Fe(II) and 272
adsorbed Fe(II) is more reactive than that by Fe(II) adsorbed on magnetite surfaces alone. 273
274
4. Conclusions 275
The reduction of Cr(VI) in soil mineral suspensions was investigated. The concentration of 276
reactive Fe(II) on the magnetite surfaces was a very important factor that significantly affected the 277
reduction kinetics of Cr(VI). Unless the Cr(VI) concentration is much lower than the surface-bound 278
reactive Fe(II), the slow reduction of Cr(VI) followed the rapid initial sorption of Cr(VI) onto the 279
magnetite surfaces. Because of these characteristics of surface reduction, the reaction kinetics of 280
Cr(VI) cannot be properly described by the zero-order or pseudo-first-order rate law, but well 281
described by the modified Langmuir-Hinshelwood kinetic model. 282
The addition of Fe(II) to the magnetite suspensions enhanced the Cr(VI) reduction, which 283
indicates that the reduction can be controlled and accelerated by the simple treatment. Most Fe(II) 284
added into the magnetite suspension was adsorbed on the soil mineral surface and the Cr(VI) 285
reduction rate increased at constant pH (7.2). However the extent of Cr(VI) reduction in the 286
magnetite suspension with Fe(II) did not reach the estimated value, which was a simple summation 287
of the reduced Cr(VI) in the magnetite suspension and that in the Fe(II) solution. An identical result 288
was observed in the soil and the nontronite suspensions. These results showed that the soluble Fe(II) 289
added cannot stoichiometrically convert the surface Fe(III) to Fe(II) and that electrons may transfer to 290
the bulk suspension. The Fe(II)-initiated reduction of Cr(VI) was also significantly affected by 291
suspension pH. The reduction rate of Cr(VI) decreased as the pH of magnetite suspension increased 292
(above the zpc of magnetite). The results can be applied to the development of remediation 293
12
technology and used for the optimal operation of in-situ redox manipulation and natural attenuation 294
to treat groundwater and soil contaminated with Cr(VI). 295
296
Acknowledgements 297
This research was supported by the grants from the Korean Ministry of Environment (Eco-Technopia 298
21). The contents do not necessarily reflect the views and policies of the sponsor nor does the 299
mention of trade names or commercial products constitute endorsement or recommendation for use. 300
301
References 302
References 303
1. Barnhart, J. (1997). Occurrences, uses, and properties of chromium. Regul. Toxicol. Pharm. 304
26, S3-S7. 305
2. Eary, L.E., and Rai, D. (1987). Kinetics of chromium(III) oxidation to chromium(VI) by 306
reaction with manganese dioxide. Environ. Sci. Technol. 21, 1187-1193. 307
3. White, A.F., and Peterson, M.L. (1996). Reduction of aqueous transition metal species on the 308
surfaces of Fe(II)-containing oxides. Geochim. Cosmochim. Acta 60, 3799-3814. 309
4. Williams, A.G.B., and Scherer, M.M. (2001). Kinetics of Cr(VI) reduction by carbonate 310
green rust. Environ. Sci. Technol 35, 3488-3494. 311
5. Song, D., Kim, Y., and Shin, W.S. (2005). A simple mathematical analysis on the effect of 312
sand in Cr(VI) reduction using zero valent iron, Korean J. Chem. Eng., 22, 67-69. 313
6. Gergory, K.B., Larese-Casanova, P., Parkin, G.F., and Scherer, M.M. (2004). Abiotic 314
transformation of hexahydro-1,3,5-trinitro-1,3,5-triazine by Fe(II) bound to magnetite. 315
Environ. Sci. Technol 38, 1408-1414. 316
7. Williams, A.G.B., and Scherer, M.M. (2004). Spectroscopic evidence for Fe(II)-Fe(III) 317
electron transfer at the iron oxide-water interface. Environ. Sci. Technol 38, 4782-4790. 318
13
8. Eary, L.E., and Rai, D. (1988). Chromate removal from aqueous wastes by reduction with 319
ferrous ion. Environ. Sci. Technol 22, 972-997. 320
9. Taylor, R.M., Maher, B.A., and Self, P.G. (1987). Magnetite in soils: I. the synthesis of 321
singledomain and superparamagnetic magnetite. Clay Miner. 22, 411-422. 322
10. Lee, W., and Batchelor, B. (2002). Abiotic reductive dechlorination of chlorinated ethylenes 323
by iron-bearing soil minerals. 1. pyrite and magnetite. Environ. Sci. Technol. 36, 5147-5154. 324
11. Alowitz, M.J., and Scherer M.M. (2002). Kinetics of nitrate, nitrite, and Cr(VI) reduction by 325
iron metal. Environ. Sci. Technol. 36, 299-306. 326
12. Lee, W., Batchelor, B., and Schlautman, M.A. (2000). Reductive capaicty of soils for 327
chromium. Environ. Technol. 21, 953-963. 328
13. Clescerl, L.S., Greenberg, A.E., and Eaton, A.D. (1992). Standard Methods for the 329
Examination of Water and Wastewater (Boston: American Public Health Association). 330
14. Stookey, L.I. (1970). Ferrozine-A new spectrophotometric regent for iron. Anal. Chem. 42, 331
779-781. 332
15. Peterson, M.L., Gordon, E., Brown, J., Parks, G.A., and Stein, C.L. (1997). Differential redox 333
and sorption of Cr(III/VI) on natural silicate and oxide minerals; EXAFS and XANES results. 334
Geochim. Cosmochim. Acta 61, 3399-3412. 335
16 Peterson, M.L., White, A.F., Brown, G.E., and Parks, G.A. (1997). Surface passivation of 336
magnetite by reaction with aqueous Cr(VI): XAFS and TEM Results. Environ. Sci. Technol. 337
31, 1573-1576. 338
17. Regazzoni, A.E., Blesa, M.A., and Maroto, A.J.G. (1982). Interfacial properties of zirconium 339
dioxide and magnetite in water. J. Colloid Interf. Sci. 91, 560-570. 340
18. Jardine, P.M., Fendorf, S.E., Mayes, M.A., Larsen, I.L., Brooks, S.C., and Bailey, W.B. 341
(1999). Fate and transport of hexavalent chromium in undisturbed heterogeneous soil. 342
Environ. Sci. Technol. 33, 2939-2944. 343
344
14
Figure Captions. 345
Fig 1. X-ray diffraction of synthesized magnetite. The values represent the d-space values (Å) 346
Fig 2. Reduction of Cr(VI) by magnetite, nontronite, and soil at pH 7.2. 347
Fig 3. Effect of magnetite dosage on the reaction kinetics of Cr(VI) at pH 7.2. 348
Fig 4. Effect of initial Cr(VI) concentration on the reaction kinetics of Cr(VI) at pH 7.2. 349
Fig 5. Effect of pH on the reaction kinetics of Cr(VI). 350
Fig 6. Reduction of Cr(VI) in solid suspensions with and without Fe(II) and aqueous Fe(II) solution 351
(pH 7.2, Fe(II)/Cr(VI) = 1.0). 352
Fig 7. Effect of pH on the reaction kinetics of Cr(VI) by magnetite with Fe(II). 353
354
355
15
356
Fig 1. 357
358
359
16
360
Time (min)
0 10 20 30 40 50 60
Re
lative
co
nce
ntr
aito
n o
f C
r(V
I) (
C/C
o)
0.0
0.2
0.4
0.6
0.8
1.0
magnetite
nontronite
soil
control
361
362
363
364
365
Fig 2. 366
367
17
368
369
370
371
Time (hrs)
0 2 4 6 8
Re
lative
co
nce
ntr
atio
n o
f C
r(V
I) (
C/C
o)
0.0
0.2
0.4
0.6
0.8
1.05 g/L
6.5 g/L
8 g/L
372
373
374
375
Fig 3. 376
377
18
378
379
380
381
Time (min)
0 10 20 30 40 50 60
Re
lative
co
nce
ntr
aito
n o
f C
r(V
I) (
C/C
o)
0.0
0.2
0.4
0.6
0.8
1.0
20mg/L
50mg/L
80mg/L
382
383
384
Fig 4. 385
386
19
387
388
389
Time (min)
0 10 20 30 40 50 60
Re
lative
co
nce
ntr
atio
n o
f C
r(V
I) (
C/C
o)
0.0
0.2
0.4
0.6
0.8
1.0
pH 8.0
pH 7.2
pH 6.5
pH 5.5
390
391
392
Fig 5. 393
394
20
395
396
397
Time (min)
0 10 20 30 40 50 60
Rela
tive c
oncentr
ation o
f C
r(V
I) (
C/C
0)
0.0
0.2
0.4
0.6
0.8
1.0
Fe(II)+Nontronite
Fe(II)+Soil
Fe(II)+Magnetite
Only Fe(II) addition
Only Magnetite
Expected value for Fe(II)+Magnetite
398
399
400
Fig 6. 401
402
403
404
405
406
21
407
408
409
Time (min)
0 10 20 30 40 50 60
Re
lative
co
nce
ntr
atio
n o
f C
r(V
I) (
C/C
o)
0.0
0.2
0.4
0.6
0.8
1.0
magnetite only at pH 7.2
pH 8.0
pH 7.2
pH 6.5
pH 5.5
410
411
412
Fig 7. 413
414
415