Selenium(IV) sorption onto γ-Al2O3: a consistent description of the

Manuscript submitted to Environmental Science & Technology. 1. Selenium(IV) sorption onto γ-Al2O3: a consistent. 1 description of the surface speciat...
0 downloads 7 Views 505KB Size
Subscriber access provided by UNIV OF DURHAM

Article

Selenium(IV) sorption onto #-Al2O3: a consistent description of the surface speciation by spectroscopy and thermodynamic modeling Natalia Mayordomo, Harald Foerstendorf, Johannes Lutzenkirchen, Karsten Heim, Stephan Weiss, Ursula Alonso, Tiziana Missana, Katja Schmeide, and Norbert Jordan Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b04546 • Publication Date (Web): 12 Dec 2017 Downloaded from http://pubs.acs.org on December 12, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 32

Environmental Science & Technology

Manuscript submitted to Environmental Science & Technology

1

Selenium(IV) sorption onto γ-Al2O3: a consistent

2

description of the surface speciation by spectroscopy

3

and thermodynamic modeling

4

Natalia Mayordomo1, Harald Foerstendorf2,*, Johannes Lützenkirchen3, Karsten Heim2,

5

Stephan Weiss2, Ursula Alonso1, Tiziana Missana1, Katja Schmeide2, Norbert Jordan2,*

6 7

1

8

2

CIEMAT, Department of Environment, Avenida Complutense 40, CP 28040, Madrid (Spain) Helmholtz-Zentrum Dresden - Rossendorf (HZDR), Institute of Resource Ecology, Bautzner

9 10 11

Landstraße 400, 01328 Dresden (Germany) 3

Institute for Nuclear Waste Disposal, Karlsruhe Institute of Technology, Hermann-vonHelmholtz Platz 1, 76344 Eggenstein-Leopoldshafen (Germany)

12

* Corresponding authors. Phone +49 351 260 21 48 (N.J.), +49 351 260 36 64 (H.F.)

13

E-mail address: [email protected] (N.J.), [email protected] (H.F.)

14

1 ACS Paragon Plus Environment

Environmental Science & Technology

Page 2 of 32

Manuscript submitted to Environmental Science & Technology 15

TABLE OF CONTENT

16 17

2 ACS Paragon Plus Environment

Page 3 of 32

Environmental Science & Technology

Manuscript submitted to Environmental Science & Technology 18

ABSTRACT

19

The sorption processes of Se(IV) onto γ-Al2O3 were studied by in situ Infrared spectroscopy,

20

batch sorption studies, zeta potential measurements and Surface Complexation Modeling (SCM)

21

in the pH range from 5 to 10. In situ Attenuated Total Reflection Fourier-transform Infrared

22

(ATR FT-IR) spectroscopy revealed the predominant formation of a single inner-sphere surface

23

species at the alumina surface, supporting previously reported EXAFS results, irrespective of the

24

presence or absence of atmospherically derived carbonate. The adsorption of Se(IV) decreased

25

with increasing pH, and no impact of the ionic strength was observed in the range from 0.01 to

26

0.1 mol L−1 NaCl. Inner-sphere surface complexation was also suggested from the shift of the

27

isoelectric point of γ-Al2O3 observed during zeta potential measurements when Se(IV)

28

concentration was 10−4 mol L−1. Based on these qualitative findings, the acid-base surface

29

properties of γ-Al2O3 and the Se(IV) adsorption edges were successfully described using a 1-pK

30

CD-MUSIC model, considering one bidentate surface complex based on previous EXAFS

31

results. The results of competitive sorption experiments suggested that the surface affinity of

32

Se(IV) towards γ-Al2O3 is higher than that of dissolved inorganic carbon (DIC). Nevertheless,

33

from the in situ experiments, we suggest that the presence of DIC might transiently impact the

34

migration of Se(IV) by reducing the number of available sorption sites on mineral surfaces.

35

Consequently, this should be taken into account in predicting the environmental fate of Se(IV).

36

Keywords: Se(IV); Sorption; ATR FT-IR spectroscopy; CD-MUSIC; Alumina; Carbonate;

37

Competition

3 ACS Paragon Plus Environment

Environmental Science & Technology

Page 4 of 32

Manuscript submitted to Environmental Science & Technology 38

INTRODUCTION

39

Selenium (Se) is present in the environment due to natural or anthropogenic sources (e.g. the

40

use of fertilizers).1 Selenium removal is a matter of concern because it represents health hazards

41

for living organisms.2 Under near-neutral pH and slightly reducing conditions, the

42

hydrogenselenite (HSeO3−) and selenite (SeO32−) anions are the most stable chemical species in

43

aqueous solutions.3 Consequently, understanding the fate of selenium in the environment and the

44

processes governing its migration are of high importance.

45

Adsorption onto colloidal particles (e.g. iron oxyhydroxides or clays), which are ubiquitous in

46

the environment, is a process potentially facilitating the dissemination of selenium. Gamma

47

alumina (γ-Al2O3) has been widely used in different industrial and research applications (e.g. as

48

catalytic support) due to its surface characteristics,4 although it is metastable with respect to other

49

aluminum (hydro)oxide phases.5 Considering rising demand, it is likely that nano γ-alumina can

50

be released into the bio/geosphere. γ-Al2O3 is also a good candidate to sorb anions within a wide

51

range of pH conditions, because of its amphoteric behavior and high point of zero charge (pHPZC

52

≈ 8-9).6-8 Furthermore, γ-Al2O3 can be used as a model oxide for clay minerals, since alumina

53

expose surface functional groups comparable to those present at the edge of clay particles.9

54

Identifying the surface interactions at a molecular scale is of interest since the type of surface

55

complexation affects the reversibility of the sorption processes and, thus, the retention capability.

56

X-ray Absorption Spectroscopy (XAS) has been previously applied to get insights on the Se(IV)

57

surface speciation on γ-alumina.10 The formation of bidentate bridging complexes with AlO6

58

surface groups between pH 4 and 8 was observed. The possibility of additional outer-sphere

59

Se(IV) complexes at the γ-Al2O3 surface was not completely ruled out, however.10 Accurate

60

knowledge about the occurring surface processes is mandatory for a robust thermodynamic

4 ACS Paragon Plus Environment

Page 5 of 32

Environmental Science & Technology

Manuscript submitted to Environmental Science & Technology 61

description based on surface complexation modeling, to constrain the number of surface species

62

and their stoichiometry.

63

Under groundwater conditions, Se(IV) coexists with other anions (SO42−, Cl− and NO3−), which

64

in principle will be competing for sorption sites on (Fe,Al) oxides surfaces.11-14 On such surfaces,

65

the sorption of Se(IV) was found to be significantly impacted and decreased in the presence of

66

such competitors, depending on the competitor to selenium concentration ratio, and on the

67

prevailing pH. However, to the best of our knowledge, less attention has been paid to the

68

competing effect of dissolved inorganic carbon (DIC) on Se(IV) surface coordination and

69

reversibility.

70

In this work, in situ Attenuated Total Reflection Fourier-transform Infrared (ATR FT-IR)

71

spectroscopy15-17 was applied to identify Se(IV) surface complexes on the γ-Al2O3 surface, and to

72

thereby understand molecular interactions and sorption and desorption processes including the

73

impact of carbonate at different pH values and ionic strengths. At the macroscopic scale, batch

74

studies and zeta potential measurements were performed in order to investigate the impact of pH

75

and ionic strength on the sorption of Se(IV) onto γ-Al2O3. Potentiometric titrations and zeta

76

potential measurements were carried out and the acid-base surface properties of γ-Al2O3 were

77

modeled using a 1-pK CD-MUSIC approach. The derived properties (binding electrolyte

78

constants and capacitances) were then used to describe Se(IV) adsorption edges and the

79

isoelectric points, constraining surface speciation based on the IR investigations.

80 81 82 83 84 5 ACS Paragon Plus Environment

Environmental Science & Technology

Page 6 of 32

Manuscript submitted to Environmental Science & Technology 85

MATERIALS AND METHODS

86

γ-Al2O3 nanoparticles characterization. γ-Al2O3 nanoparticles (>99% purity, Aldrich) from

87

two batches with nominal diameter < 50 nm were used. The first batch was used for the

88

potentiometric titrations, zeta potential and Se(IV) adsorption edges. The specific surface area of

89

this batch (Multi-point Beckman Coulter surface analyzer SA 3100) was determined to be 127 m2

90

g−1 by applying the Brunauer–Emmett–Teller (BET) equation with nitrogen adsorption isotherms

91

at 77 K. Characterization details of the second batch exclusively used for the spectroscopic

92

investigations were presented elsewhere.14 The N2-BET specific surface area of this sample was

93

136 m2·g−1 and its isoelectric point pHIEP equal to 8.5. XRD measurements confirmed the purity

94

of γ-Al2O3 and the absence of other Al-phases such as bayerite, gibbsite, etc. in both batches

95

(Figure S1, Supporting Information (SI)).

96

Several studies highlighted the transformation of γ-Al2O3 at room temperature and in aqueous

97

suspensions after a few days into Al (oxy)hydroxides.18-22 The γ-Al2O3 sample (batch 1) was

98

equilibrated for three and seven days in 0.1 mol L−1 NaCl. After freeze-drying, samples were

99

analyzed by XRD and IR spectroscopy. No significant differences compared to the raw material

100

were observed (Supporting Information Figures S2 and S3). During the in situ ATR FT-IR

101

sorption studies, the same suspension was used for all experiments and no hints for surface

102

alterations with time (over two months) were noticed. Consequently, bulk and surface

103

transformations of γ-Al2O3 were not observed under our experimental conditions. This apparent

104

contradiction with former studies18-22 might arise from differences in the surface to bulk ratios.

105

For the modeling purposes, surface acid-base properties of γ-Al2O3 were comprehensively

106

characterized by potentiometric titrations and zeta potential studies. Potentiometric titrations (pH

107

range 5 to 9.5) were performed at different ionic strengths of NaCl (0.1, 0.05 and 0.01 mol L−1)

108

with a Metrohm 736 GP Titrino titrator. For each titration, a 20 g L−1 suspension of γ-Al2O3 (50 6 ACS Paragon Plus Environment

Page 7 of 32

Environmental Science & Technology

Manuscript submitted to Environmental Science & Technology 109

mL volume) was inserted in a borosilicate vessel and equilibrated over night at pH ~5. A

110

continuous argon flux (Argon N50 from Air Liquide) was streamed over the suspension to avoid

111

dissolution of atmospheric CO2. To ensure a homogeneous suspension, a Teflon propeller was

112

used. After overnight pre-equilibration, base titration was performed by addition of aliquots (20

113

µL) of 0.1 mol L−1 NaOH. The pH electrode (Schott BlueLine 11pH) was calibrated using a three

114

point calibration with buffer solutions (pH 4.01, 6.87 and 9.18). Zeta potential measurements

115

were also performed using a Laser-Doppler-Electrophoresis instrument (nano-ZS, Malvern

116

Instruments Ltd.), under CO2 exclusion, at a mass to volume ratio m/v = 0.25 g L−1 and at two

117

ionic strengths (0.005 and 0.01 mol L−1 NaCl).

118 119

Solutions. Detailed descriptions of the preparation of the solutions are given in the Supporting Information.

120

ATR FT-IR experiments. The in situ IR spectroscopic experiments allowed a monitoring of

121

the sorption and desorption processes on solid phase surfaces. For this, a mineral film was

122

prepared as a stationary phase directly onto the surface of the ATR crystal by repeatedly pipetting

123

aliquots of 1 µL from a suspension (2.5 g·L−1 γ-Al2O3 in deionized Milli-Q water) and

124

subsequently evaporating the solvent by gentle drying with N2(g). The amount of alumina on the

125

crystal surface was verified by the transmission of the IR light which was not allowed to fall

126

below a value of ~15% in the mid-IR frequency region. Subsequently, a flow cell was mounted

127

which was connected to a peristaltic pump providing a continuous flow rate of 100 µL·min−1

128

from the reservoirs containing the solutions of the background electrolyte or of the sorbate, that is

129

Se(IV) or DIC. A valve allows the selection of the respective solution to be flushed through the

130

setup.

131

Sorption and desorption experiments were monitored by continuously measuring IR single

132

beam spectra during three sequential steps of the experiment, that is (i) equilibration, (ii) sorption 7 ACS Paragon Plus Environment

Environmental Science & Technology

Page 8 of 32

Manuscript submitted to Environmental Science & Technology 133

and (iii) flushing/desorption as described earlier.15, 16 During the equilibration step, the stationary

134

phase was flushed with a blank solution only containing the background electrolyte for at least 60

135

min until spectral changes minimized. Subsequently, sorption was induced by switching to the

136

respective solution containing the sorbate until no further spectral changes occurred. Finally,

137

desorption was induced by switching back to the blank solution. This setup allowed to induce

138

adsorption and desorption processes of Se(IV) in the presence or absence of DIC. Moreover, both

139

anions could be optionally added consecutively or simultaneously providing insight into

140

competitive processes of both anions at the solid-liquid interface. With respect to the absorption

141

properties of the aqueous phase, the relatively low concentrations and the characteristic frequency

142

ranges of the vibrational modes of the Se(IV) and HCO3− anions, i.e. 1000–600 cm−1 and 1600–

143

1300 cm−1, the experiments had to be performed in D2O and H2O, respectively. In D2O, the pD

144

was calculated by the following equation23:  =    + 0.4

145

All IR experiments were performed with a Bruker Vertex 80/v spectrometer equipped with a

146

horizontal ATR diamond crystal (SamplIIR II, Smiths inc., 9 reflections, 45º angle of incidence)

147

and a Mercury Cadmium Telluride detector with a low frequency cut-off at 580 cm−1. Spectral

148

resolution was 4 cm−1 and the spectra were averaged over 256 scans. All manipulations of the

149

spectra were accomplished with the OPUS™ software.15, 16 Further details of the acquisition of

150

reference spectra are given in the Supporting Information.

151

Batch sorption and zeta potential studies. γ-Al2O3 (10 mg) was suspended in 40 mL of NaCl

152

solution in 50 mL polypropylene tubes resulting in m/v = 0.25 g L−1. Two background electrolyte

153

concentrations were used, i.e. 0.1 and 0.01 mol L−1. The initial Se(IV) concentrations were 10−5

154

and 10−4 mol L−1. The pH of the alumina suspensions was adjusted by addition of HCl or NaOH

155

and was recorded after being stable for more than 10 minutes. Experiments were performed in a 8 ACS Paragon Plus Environment

Page 9 of 32

Environmental Science & Technology

Manuscript submitted to Environmental Science & Technology 156

glove box in N2 atmosphere (O2 < 5 ppm). After 3 days of equilibration, samples were

157

centrifuged for 2 hours at 6,800 × g (Avanti J-20 XP Beckman Coulter) and the remaining

158

selenium concentration in the supernatants was determined by ICP-MS (PerkinElmer ELAN

159

9000). All the experiments were carried out in duplicate. Zeta potential measurements were

160

carried out on the same suspensions (prior to the centrifugation), the measuring cell being filled

161

in a glovebox under a N2 atmosphere. A more detailed description of the instrumental analytic

162

techniques applied can be found elsewhere.24

163

Surface Complexation Modeling. The most frequently used model to describe the surface of

164

γ-Al2O3 consists of five different kinds of OH groups, namely AlIVOH−0.25, AlVIOH−0.25, [AlIV-

165

OH-AlVI]+0.25, [AlVI-OH-AlVI]0 and [AlVI-OH-(AlVI)2]+0.5, derived from IR and NMR studies.4

166

The surface properties of γ-Al2O3 using one equivalent site ≡SO−0.5 representing the acidic

167

groups AlIVOH−0.25, AlIVAlVIO−0.75 and AlVI3O−0.5, as well as one site ≡AlOH−0.5 with singly

168

coordinated hydroxyl groups were described previously.25 25% of both types of sites were

169

assumed to be located in the subsurface. Satisfactory results were obtained by assuming charge

170

penetration of both Na and Cl electrolyte ions for the subsurface sites. A total site density of 6.6

171

sites nm−2 (2.5 for each normal site and 0.8 for each subsurface site) was used by Hiemstra et al.25

172

As shown later, a pragmatic approach considering one site ≡AlOH−0.5 (thought to represent

173

singly coordinated hydroxyl groups with fractional charges) is sufficient to provide an adequate

174

description of the surface properties of γ-Al2O3. A surface site density of 7 sites nm−2 was chosen

175

in this study, close to the total surface site density used by Hiemstra et al.25

176

Surface complexation modeling was performed using the CD-MUSIC model with a one-pK

177

approach.26 The titration and zeta potential data for the sorbent were modeled using one site as

178

discussed above with a pK equal to 9.0. Since a better description of the data in comparison to a

179

two plane model (Basic Stern) could be obtained, a three plane model with electrolyte binding in 9 ACS Paragon Plus Environment

Environmental Science & Technology

Page 10 of 32

Manuscript submitted to Environmental Science & Technology 180

the plane 2 (head-end of the diffuse layer) was used. The adjustable parameters were the two

181

electrolyte association constants (Na+ and Cl−) and the capacitance value C1 (C2 being fixed at 5 F

182

m−2).25 Furthermore, the slip-plane distance, s, was fitted to obtain a good fit to the zeta potential

183

data. Here, we followed previously published procedures.27 The fitting procedure was performed

184

using a modified version of FITEQL coupled to UCODE.28, 29

185 186

RESULTS AND DISCUSSION

187

Se(IV) surface species on γ-Al2O3. To identify oxyanion sorption onto mineral surfaces by

188

vibrational spectroscopy, the knowledge of the spectral fingerprints of the predominant aqueous

189

species is required. With respect to the experimental setup, contributions from aqueous species to

190

the spectra of the sorbed species cannot be ruled out. Thus, the respective spectra obtained for

191

aqueous Se(IV) solutions at selected pD values are provided in Figure 1A serving as references

192

for the spectra obtained during the sorption and desorption processes (Figures 1B–E). The Se(IV)

193

aqueous speciation (Figure S4) as well as a brief assignment of the observed bands to vibrational

194

modes of the Se(IV) species is given in the Supporting Information.

195

The transient spectra recorded for the sorbed species showed increasing amplitudes in the

196

spectral region of the ν(SeO) stretching modes reflecting the accumulation of surface species at

197

the alumina surface (Supporting Information Figure S5). In all experiments of this work, the

198

amplitudes generally reached maximum values about 40 min after sorption was initiated

199

indicating that a steady state was obtained. In addition, prolonged sorption up to 120 min

200

revealed no spectral changes ruling out the formation of surface precipitates. Representative

201

spectra recorded after 60 min of sorption are shown in Figure 1B.

202

The spectra following Se(IV) sorption (Figure 1B) were significantly different from the spectra

203

of the aqueous species (Figure 1A) evidencing the formation of surface species at the alumina 10 ACS Paragon Plus Environment

Page 11 of 32

Environmental Science & Technology

Manuscript submitted to Environmental Science & Technology 204

surface. Under mildly acidic conditions (pD 4.9), a predominant broad feature was observed

205

showing two maxima at 844 and 762 cm−1 representing the ν1,s and ν3,as(Se–O) modes (see

206

Supporting Information).30 Under more alkaline conditions, the intensity of the bands decreased

207

with increasing pH reflecting a decreased affinity of the Se(IV) anions for the alumina surface

208

most likely due to less favorable electrostatics.

209

The ν3,as(Se–O) mode was bathochromically shifted to 756 and 747 cm−1 with increasing pD

210

(Figure 1B). Such a shift potentially might indicate the formation of another species due to the

211

increased pD. However, the band showed no broadening with increasing pD suggesting that the

212

frequency of the respective vibrational mode is just lowered and no additional mode contributed

213

to this spectral feature. This interpretation was corroborated by spectral decomposition and fitting

214

of the spectra. The results obtained demonstrated that the spectra could be satisfactorily

215

reproduced with only two bands representing two different vibrational modes, one constantly

216

located around 840 cm−1 and another slightly shifting from ~760 to ~745 cm−1 with increasing pD

217

(Supporting Information Figure S6). The occurrence of additional species is expected to require

218

further components to obtain adequate fitting results. Thus, the pD depending shifting of the

219

ν3,as(Se–O) mode likely reflects modifications of the alumina surface charge due to the prevailing

220

pD value. Together with the results of the previous EXAFS study of Elzinga et al.10 which did not

221

reveal the presence of another species related to a second Se-Al distance, the formation of a

222

single predominant Se(IV) species at the alumina surface, independently of the pH, can be

223

assumed with confidence.

224

Subsequent to sorption, desorption was induced by flushing the alumina film with a 0.1 M

225

NaCl blank electrolyte solution for one hour. The respective spectra are shown as negative bands

226

representing species which were released from the mineral phase during the desorption step

227

(Figure 1C). The high congruence of the spectral shapes observed during the sorption and 11 ACS Paragon Plus Environment

Environmental Science & Technology

Page 12 of 32

Manuscript submitted to Environmental Science & Technology 228

desorption steps strongly suggests that the same Se(IV) species were involved in adsorption and

229

desorption processes. The amplitudes of these spectra were found to be nearly independent of pD

230

indicating an increasing relative reversibility with increasing pD. However, at pD 4.9 the

231

amplitude of the desorption spectra did not exceed ~55% of the sorption spectra. This rather low

232

reversibility of the sorption reaction indicates predominant formation of inner-sphere complexes

233

under those conditions. Prevalence of outer-sphere complexation was found to result in high

234

reversibility, which is in good agreement with the amplitudes of sorption and desorption spectra

235

in recent in situ ATR FT-IR spectroscopic experiments.16, 31

236

To verify the inner-sphere Se(IV)-alumina species suggested in our experiments, we have

237

performed in situ experiments at different ionic strengths (Supporting Information Figure S7).

238

Increasing the ionic strength by a factor up to 50 (from 0.01 to 0.5 M) revealed no significant

239

impact on the resulting spectra irrespective of the pH conditions. This insensitivity of the spectra

240

with ionic strength can be interpreted in terms of predominant formation of inner-sphere

241

complexes, which is in agreement with the observed decrease to more acidic pH value of the

242

isoelectric point of alumina upon Se(IV) sorption. 10, 14 While these macroscopic observations are

243

suggestive, the formation of inner-sphere complexes upon Se(IV) sorption onto γ-Al2O3 was

244

definitely shown from EXAFS measurements.10

245

The derivation of structural molecular information, such as molecule symmetry and binding

246

mode of surface complexes, from vibrational spectra is potentially accessible due to the

247

correlation between the molecule symmetry and occurring vibrational modes. However, an

248

unequivocal interpretation of the spectral features observed for the Se(IV) surface species is

249

rather difficult due to low molecule symmetry. Such low molecule symmetries generally reveal

250

numerous vibrational modes characterized by weak intensities, broad line widths and poor

251

energetic separation. From vibrational data of SeO32− metal complexes it is known that the 12 ACS Paragon Plus Environment

Page 13 of 32

Environmental Science & Technology

Manuscript submitted to Environmental Science & Technology 252

selenite anion in the complex shows Cs symmetry. This symmetry group is expected for both a

253

monodentate and a bidentate coordination of the selenite anion.32 Thus, the type of coordination

254

cannot be intrinsically distinguished based on the number of vibrational modes to be observed in

255

the spectra. Moreover, an assignment of mono- and bidentate coordinated SeO32− ions by the

256

fingerprint of vibrational modes in IR spectra has not been accurately described up to now. Only

257

the frequency ranges from 805 to 832 cm−1 and from 755 to 770 cm−1 were identified to show

258

vibrational modes of SeO32− ions in metal complexes,32 which is in agreement with the spectra of

259

the surface complexes of this work (Figure 1B–E). However, previous EXAFS studies reported

260

the formation of bidentate bridging complexes with AlO6 surface groups onto γ-alumina at

261

circumneutral pH.10 Thus, the formation of Se(IV) bidentate inner-sphere surface complexes at

262

the alumina surface was assumed.

263

To gain insight into the sorption processes of Se(IV) under ambient conditions, simultaneous

264

sorption of Se(IV) and DIC was studied. As the IR measurements of the Se(IV) sorption

265

processes had to be performed in D2O, the atmospherically equimolar amount of DIC was added

266

to the Se(IV) solution as solid NaDCO3, which avoided H2O contamination. The respective

267

Se(IV) spectra obtained after a prolonged equilibration of the alumina phase under inert gas

268

conditions and subsequent 60 min of incubation are shown in Figure 1D. The experiments were

269

performed at three different pD values. The spectra of the subsequent flushing steps are shown in

270

Figure 1E.

271

The overall shapes of the spectra are in very good agreement with those obtained in the absence

272

of DIC indicating that the same Se surface species are formed, regardless on DIC presence, which

273

is also supported by the same observed bathochromic shift of the band’s maximum with

274

increasing pD. However, the amplitudes of the spectra obtained at increased pD values were

275

reduced in the presence of DIC (Figure 1D). From these findings, it is assumed that prevailing 13 ACS Paragon Plus Environment

Environmental Science & Technology

Page 14 of 32

Manuscript submitted to Environmental Science & Technology 276

carbonate species hampered the accessibility of Se(IV) to the alumina surface, as the number of

277

available surface sites decreases.

278

In contrast, the coordination of carbonate on the alumina surface was almost completely

279

suppressed in the presence of 5 × 10−4 mol L−1 Se(IV). It was explicitly shown that carbonate

280

ions showed an affinity to the alumina surface in the absence of Se(IV) (Supporting Information

281

Figure S8). The high degree of reversibility found for these sorption processes and the spectral

282

characteristics of the carbonate modes suggested the formation of monodentate outer-sphere

283

species at the alumina surface (Supporting Information Figure S8). During simultaneous sorption

284

of Se(IV) and DIC, hardly any bands representing DIC were observed reflecting a negligible

285

uptake in the presence of Se(IV) on alumina which corroborates the higher affinity of Se(IV) for

286

the alumina surface (Supporting Information Figure S8).

287

This higher affinity is expected to cause removal of carbonate ions from the surface by Se(IV),

288

provided that both species bind to the same sites. In fact, this assumption was confirmed by an

289

additional experiment where Se(IV) was exposed to the alumina phase which had been allowed

290

to equilibrate with DIC beforehand. The spectral modifications related to Se(IV) sorption

291

processes performed at circumneutral pH are shown in Figure 2A and B for the frequency ranges

292

of the carbonate and Se(IV) modes, respectively.

293

The spectra clearly demonstrate that during the sorption processes of Se(IV), characterized by

294

the band at 757 cm−1 (Figure 2B), carbonate was released from the alumina surface as reflected

295

by negative bands observed at 1504 and 1392 cm−1 (Figure 2A). Thus, it is obvious that Se(IV)

296

replaced carbonate ions on the alumina surface again confirming the higher affinity of Se(IV).

297

Subsequent flushing with pure background electrolyte demonstrated a fractional release of Se(IV)

298

(Figure 2D), which is in accordance with previous results (Figures 1C and E), and with a very

299

small residual amount of remaining carbonate (Figure 2C).The high congruence of vibrational 14 ACS Paragon Plus Environment

Page 15 of 32

Environmental Science & Technology

Manuscript submitted to Environmental Science & Technology 300

modes of Se(IV) throughout the spectra of the sorption and desorption processes strongly

301

suggests the formation of a single predominant inner-sphere surface species on alumina

302

irrespective of the conditions (i.e. inert or DIC containing solutions). Although the derivation of

303

detailed molecule symmetries from the spectral fingerprinting is not feasible given the current

304

state of knowledge, a bidentate coordination of the anion to the surface is assumed as suggested

305

from earlier investigations.10 These qualitative findings constitute a prerequisite for the

306

comprehensive description of the sorption system Se(IV)/γ-Al2O3 studied by batch sorption

307

experiments and thermodynamic modeling.

308 309

Surface acid-base behavior, zeta potentials and modeling. To model the Se(IV) adsorption

310

data successfully, the determination of the acid-base parameters of the solid phase is required.

311

The experimental titration results (surface charge density vs. pH) at different ionic strengths are

312

shown in Figure 3A, while zeta potential measurements as a function of pH are given in Figure

313

3B. Solid lines show the modeling results in each figure. The reactions used and the

314

corresponding obtained parameters are summarized in Table 1. The capacitance C1 and the

315

parameter x were 1.39 F m−2 and 0.44, respectively. The fitted parameter x is related to the slip

316

plane distance s and the ionic strength dependent Debye length κ by x = s × κ.33

317

As it can be seen in Figure 3A, a satisfactory description of the titration data is obtained with a

318

one site and a three plane model. The pHIEP of γ-Al2O3, located at pH 9.0, was insensitive to the

319

background electrolyte concentration (Figure 3A), suggesting the identical or absence of stronger

320

adsorption of Na or Cl to γ-Al2O3. The potentiometric data derived in the present work were

321

compared with former literature studies (Supporting Information Figures S9 and S10). A

322

comparison of the results with literature shows that the pHIEP of the present study agrees very

323

well with literature data. Indeed, compilations of Kosmulski7, 34 show that the major part of the 15 ACS Paragon Plus Environment

Environmental Science & Technology

Page 16 of 32

Manuscript submitted to Environmental Science & Technology 324

PZCs/IEPs of alumina are in the range between pH 8 and 10. Some gamma-alumina data show

325

the very same value that we have obtained. The zeta potential curves and the pHIEP were rather

326

well described (Figure 3B). Charge penetration of the electrolyte ions in the plane 0 or 1 as

327

proposed earlier did not significantly improve the quality of the fit.25 Consequently, this model

328

was accepted and used further for the modeling of the Se(IV) adsorption data.

329

Batch sorption studies, zeta potentials and modeling. The amount of Se(IV) sorbed onto γ-

330

Al2O3 decreased with increasing pH (Figure 4A). This sorption trend with pH is observed

331

typically in anion retention on metal oxides, since when increasing pH, the surface of the mineral

332

becomes more negative and solid-anion interaction is hindered due to electrostatic repulsion.35

333

Raising the ionic strength from 0.01 M to 0.1 mol L−1 NaCl did not significantly impact the

334

Se(IV) sorption, which was already observed10 and supports the formation of inner-sphere

335

complexes.10, 24

336

No significant shift of the pHIEP towards more acidic pH values was observed after addition of

337

10−5 mol L−1 of Se(IV) (Figure 4B). However, a clear shift of the pHIEP from 9.0 to 8.1–8.2

338

occurred with 10−4 mol L−1 of Se(IV), at the two ionic strengths used (Figure 4B). The decrease

339

of alumina pHIEP was also observed at similar Se(IV) concentrations10, 14 and agrees with the

340

formation of Se(IV) inner-sphere surface complexes.10, 24, 36

341

The parameters from the acid-base model were kept constant for the Se(IV)-adsorption model.

342

The model involves the assumptions that (i) the Se adsorption does not affect the slip-plane

343

distance and (ii) both inner and outer layer capacitance are not changed by Se adsorption.

344

For several reasons, such as e.g. the lack of knowledge on the exact position of the slipping

345

plane, we did not intend to perfectly reproduce the zeta potential values upon Se(IV)

346

adsorption.36 We rather focused on reproducing the pHIEP and its shift due to interaction with

16 ACS Paragon Plus Environment

Page 17 of 32

Environmental Science & Technology

Manuscript submitted to Environmental Science & Technology 347

Se(IV). The aqueous Se(IV) protonation constants were taken from the NEA-TDB book for

348

selenium.3

349 350

SeO32− + H+ ⇌ HSeO3−, log K = 8.36

351

SeO32− + 2 H+ ⇌ H2SeO3, log K = 11.00

352 353

To design a Se-adsorption model, we started from a previously published Se(IV)-goethite

354

model, i.e. a three-plane electrostatic model.36, 37 The spectroscopic data suggested one bidentate

355

surface complex and we therefore used only one surface species (cf. Table 1), namely

356

[{(≡AlO)(≡AlOH)}SeO]0. The interaction between the Se(IV) and the surface of γ-Al2O3 was

357

assumed to involve the singly coordinated hydroxyl groups only. The location of the Se(IV)

358

charge at the interface was described via the charge distribution factor. This resulted for the

359

single surface complex in two adjustable parameters (the log K and its respective CD-factor).

360

Since no shift of the isoelectric point upon Se(IV) adsorption was experimentally observed

361

upon increasing the NaCl concentration, it was not necessary to include sodium binding to the

362

adsorbed Se(IV) in the bidentate surface complex involved in the model, as proposed for the

363

phosphate/goethite binary system.38 The model is summarized in the following reaction equation

364

and the model fit to the batch data is shown in Figure 4.

365 366

2(≡AlOH−0.5) + 3 H+ + SeO32− ⇌ [{(≡AlO)(≡AlOH)}SeO]0 + 2 H2O, log K = 30.0.

367 368

The model describes the adsorption data satisfactorily, except for a slight underestimation at

369

10−5 mol L−1 Se(IV) and I = 0.1 mol L−1. The predominance of a bidentate surface species is in

370

full agreement with previous studies.36, 37 17 ACS Paragon Plus Environment

Environmental Science & Technology

Page 18 of 32

Manuscript submitted to Environmental Science & Technology 371

Figure 4B shows the model description of the zeta potentials of the sorbent in absence and

372

presence of Se(IV). The model accurately describes the shift of the pHIEP, although not perfectly.

373

It slightly underestimates the shift of the pHIEP of γ-Al2O3 at 10−4 mol L−1 Se(IV) and I = 0.1 mol

374

L−1. However, we think that the model is adequate since it captures the experimental observations

375

and does not disagree with the spectroscopic data.

376

Other model options tested (deprotonated bidentate species [(≡AlO)2SeO]− with or without

377

coadsorption of sodium) did not improve the fit. The model used to calculate the lines in Figure 4

378

and considering the [{(≡AlO)(≡AlOH)}SeO]0 species was the one that yielded the best fit among

379

all the variations tested. The CD values are related to the position of the proton in the bidentate

380

surface complex. The modeling outcome showed that the obtained CD factor makes sense with

381

the proton close to the surface and not on the “free” oxygen that is oriented towards the solution

382

(an oxygen which would have little proton affinity). The surface complex might be interpreted as

383

involving a hydrogen bond with a neighbouring oxygen, as for example discussed by Loring et

384

al.39 This interpretation necessarily remains tentative and is based on best fit. On the sorbent a

385

sound structural assignment becomes difficult because the surface structure is not known. As

386

such the obtained parameters and stoichiometries can also be understood as the best-fit results,

387

which do not contradict the spectroscopic data. When the parameter x was also fitted, a much

388

better description of the experimental zeta potential values was obtained at 10−4 mol L−1 Se(IV),

389

while the zeta potentials at 10−5 mol L−1 Se(IV) were slightly overestimated. Using this value for

390

x to simulate the acid-base surface properties of alumina did not have major effects (Supporting

391

Information Figures S11 and S12), besides a slight overestimation of the zeta potentials of the

392

neat alumina surface at acidic pH.

393 394 18 ACS Paragon Plus Environment

Page 19 of 32

Environmental Science & Technology

Manuscript submitted to Environmental Science & Technology 395

ENVIRONMENTAL IMPLICATIONS

396

Although the most important exposure route to humans is food, exposure to low-level selenium

397

containing drinking water (< 10 µg L−1) has been reported to induce severe health effects.2 A

398

detailed knowledge of the mobility and bioavailability of selenium is therefore of great

399

importance for a safe disposal of radioactive waste as well as for the optimization of

400

decontamination processes of polluted sites.

401

A reliable description of the interaction at the solid/liquid interface can only be achieved based

402

on a multi-component approach including adsorption data, spectroscopic characterization and

403

thermodynamic modeling, as shown in this study. This work revealed the interaction of Se(IV)

404

with γ-Al2O3 nanoparticles, not occurring naturally, but serving as a model for clay minerals.

405

Since the interaction of Se(IV) with γ-alumina nanoparticles was evidenced to proceed via the

406

formation of strong covalent bonds, stable surface complexes are to be expected at the solid

407

surface. Alumina nanoparticles in ground- or waste-waters might therefore transport and enhance

408

selenium migration in the environment. In natural systems, competitive effects with ubiquitous

409

species such as carbonate ions have to be considered as well. Indeed, they can hinder the

410

adsorption by reducing the availability of surface sites. At higher pH levels, dissolved carbonate

411

ions impede the adsorption of Se(IV) at alumina phases, which can in turn increase the mobility

412

of Se in the environment. These results can be implemented in reactive transport models to

413

improve the monitoring of the environmental fate of selenium. An accurate description of the

414

carbonate interaction with alumina surfaces including batch studies and thermodynamic modeling

415

is also of high environmental relevance and will be the scope of future studies.

416 417

19 ACS Paragon Plus Environment

Environmental Science & Technology

Page 20 of 32

Manuscript submitted to Environmental Science & Technology 418

ASSOCIATED CONTENT

419

Supporting Information. The Supporting Information (SI) contains detailed descriptions

420

concerning the characterization of γ-Al2O3, the preparation of the solutions, the aqueous

421

speciation of Se(IV) and carbonate ions, the performance of the in situ ATR FT-IR measurements

422

(impact of pD, ionic strength and dissolved inorganic carbon), spectral fitting results of the

423

sorption spectra obtained, a comparison with former potentiometric titration as well as details

424

about surface complexation modeling. This material is available free of charge via the Internet at

425

http://pubs.acs.org.

426 427

AUTHOR INFORMATION

428

Corresponding Authors

429

*Phone: +49 351 260 2148, e-mail: [email protected] (NJ)

430

*Phone: +49 351 260 3664, e-mail: [email protected] (HF)

431 432 433

Author Contributions

434

The manuscript was written through contributions of all authors. All authors have given approval

435

to the final version of the manuscript. NJ and HF are the lead authors from Helmholtz-Zentrum

436

Dresden - Rossendorf.

437

Funding Sources

438

The authors declare no competing financial interest. 20 ACS Paragon Plus Environment

Page 21 of 32

Environmental Science & Technology

Manuscript submitted to Environmental Science & Technology 439

ACKNOWLEDGMENTS

440

This work has been partially supported by MIRAME (CTM-2014-60482-P) Spanish Ministry

441

of Economy and Competitiveness project. N.M. acknowledges FPI BES-2012-056603 and

442

EEBB-I-15-09446 grants, both from MINECO (Spain). Dr. Luis Gutiérrez Nebot from CIEMAT

443

and Dr. Atsushi Ikeda-Ohno are acknowledged for XRD measurements. Birke Pfützner, Heidrun

444

Neubert, Christa Müller and Salim Shams Aldin Azzam are acknowledged for support during

445

batch, zeta potential and ICP-MS measurements.

446

21 ACS Paragon Plus Environment

Environmental Science & Technology

Page 22 of 32

Manuscript submitted to Environmental Science & Technology 447

REFERENCES

448

1.Fernández-Martínez, A.; Charlet, L., Selenium environmental cycling and bioavailability: a

449

structural chemist point of view. Rev. Environ. Sci. Biotechnol. 2009, 8, (1), 81-110.

450

2.Selenium in Drinking-Water. Background Document for Development of WHO Guidelines for

451

Drinking-Water Quality; World Health Organization: Geneva, 2011.

452

3.Olin, A.; Noläng, B.; Osadchii, E. G.; Öhman, L.-O.; Rosén, E., Chemical thermodynamics of

453

selenium. Elsevier: Amsterdam, 2005.

454

4.Trueba, M.; Trasatti, S. P., γ−Alumina as a support for catalysts: A review of fundamental

455

aspects. European Journal of Inorganic Chemistry 2005, (17), 3393-3403.

456

5.Levin, I.; Brandon, D., Metastable alumina polymorphs: Crystal structures and transition

457

sequences. Journal of the American Ceramic Society 1998, 81, (8), 1995-2012.

458

6.Sposito, G., Chemical equilibrium and Kinetics in soils, First Edition. Oxford University Press,

459

New York, 1994.

460

7.Kosmulski, M., The pH-dependent surface charging and points of zero charge V. Update. J.

461

Colloid Interface Sci. 2011, 353, (1), 1-15.

462

8.Mayordomo, N.; Alonso, U.; Missana, T., Analysis of the improvement of selenite retention in

463

smectite by adding alumina nanoparticles. Science of the Total Environment 2016, 572, 1025-

464

1032.

465

9.Lagaly, G.; Dékány, I., Colloid Clay Science. In Handbook of Clay Science, Second ed.;

466

Elsevier: Amsterdam, The Nederlands, 2013.

467

10.Elzinga, E. J.; Tang, Y. Z.; McDonald, J.; DeSisto, S.; Reeder, R. J., Macroscopic and

468

spectroscopic characterization of selenate, selenite, and chromate adsorption at the solid-water

469

interface of γ-Al2O3. J. Colloid Interface Sci. 2009, 340, (2), 153-159.

22 ACS Paragon Plus Environment

Page 23 of 32

Environmental Science & Technology

Manuscript submitted to Environmental Science & Technology 470

11.Balistrieri, L. S.; Chao, T. T., Selenium adsorption by goethite. Soil Science Society of

471

America Journal 1987, 51, (5), 1145-1151.

472

12.Fujikawa, Y.; Fukui, M., Radionuclide sorption to rocks and minerals: Effects of pH and

473

inorganic anions .1. Sorption of cesium, cobalt, strontium and manganese. Radiochimica Acta

474

1997, 76, (3), 153-162.

475

13.Kim, S. S.; Min, J. H.; Lee, J. K.; Baik, M. H.; Choi, J. W.; Shin, H. S., Effects of pH and

476

anions on the sorption of selenium ions onto magnetite. Journal of Environmental Radioactivity

477

2012, 104, 1-6.

478

14.Missana, T.; Benedicto, A.; Mayordomo, N.; Alonso, U., Analysis of anion adsorption effects

479

on alumina nanoparticles stability. Applied Geochemistry 2014, 49, 68-76.

480

15.Foerstendorf, H.; Jordan, N.; Heim, K., Probing the surface speciation of uranium (VI) on iron

481

(hydr)oxides by in situ ATR FT-IR spectroscopy. J. Colloid Interface Sci. 2014, 416, 133-138.

482

16.Jordan, N.; Ritter, A.; Foerstendorf, H.; Scheinost, A. C.; Weiss, S.; Heim, K.; Grenzer, J.;

483

Mücklich, A.; Reuther, H., Adsorption mechanism of selenium(VI) onto maghemite. Geochim.

484

Cosmochim. Acta 2013, 103, 63-75.

485

17.Lefèvre, G., In situ Fourier-transform infrared spectroscopy studies of inorganic ions

486

adsorption on metal oxides and hydroxides. Advances in Colloid and Interface Science 2004,

487

107, (2-3), 109-123.

488

18.Carrier, X.; Marceau, E.; Lambert, J. F.; Che, M., Transformations of gamma-alumina in

489

aqueous suspensions 1. Alumina chemical weathering studied as a function of pH. J. Colloid

490

Interface Sci. 2007, 308, (2), 429-437.

491

19.Dyer, C.; Hendra, P. J.; Forsling, W.; Ranheimer, M., Surface hydration of aqueous γ-Al2O3

492

studied by Fourier transform Raman and infrared spectroscopy—I. Initial results. Spectrochimica

493

Acta Part a-Molecular and Biomolecular Spectroscopy 1993, 49, (5-6), 691-705. 23 ACS Paragon Plus Environment

Environmental Science & Technology

Page 24 of 32

Manuscript submitted to Environmental Science & Technology 494

20.Laiti, E.; Persson, P.; Ohman, L. O., Balance between surface complexation and surface phase

495

transformation at the alumina/water interface. Langmuir 1998, 14, (4), 825-831.

496

21.Lefèvre, G.; Duc, M.; Lepeut, P.; Caplain, R.; Fedoroff, M., Hydration of γ-alumina in water

497

and its effects on surface reactivity. Langmuir 2002, 18, (20), 7530-7537.

498

22.Wijnja, H.; Schulthess, C. P., ATR-FTIR and DRIFT spectroscopy of carbonate species at the

499

aged γ-Al2O3/water interface. Spectrochimica Acta Part a-Molecular and Biomolecular

500

Spectroscopy 1999, 55, (4), 861-872.

501

23.Glasoe, P. K.; Long, F. A., Use of glass electrodes to measure acidities in deuterium oxide. J.

502

Phys. Chem. 1960, 64, (1), 188-190.

503

24.Jordan, N.; Ritter, A.; Scheinost, A. C.; Weiss, S., Selenium(IV) Uptake by Maghemite (γ-

504

Fe2O3). Environmental Science & Technology 2014, 48, (3), 1665-1674.

505

25.Hiemstra, T.; Yong, H.; Van Riemsdijk, W. H., Interfacial charging phenomena of aluminum

506

(hydr)oxides. Langmuir 1999, 15, (18), 5942-5955.

507

26.Hiemstra, T.; Van Riemsdijk, W. H.; Bolt, G. H., Multisite proton adsorption modeling at the

508

solid/solution interface of (hydr)oxides: A new approach: I. Model description and evaluation of

509

intrinsic reaction constants. J. Colloid Interface Sci. 1989, 133, (1), 91-104.

510

27.Bouby, M.; Lützenkirchen, J.; Dardenne, K.; Preocanin, T.; Denecke, M. A.; Klenze, R.;

511

Geckeis, H., Sorption of Eu(III) onto titanium dioxide: Measurements and modeling. J. Colloid

512

Interface Sci. 2010, 350, (2), 551-561.

513

28.Westall, J. C. FITEQL: A Computer Program for Determination of Chemical Equilibrium

514

Constants from Experimental Data; Department of Chemistry, Oregon State University:

515

Corvallis, OR, U.S.A., 1982.

516

29.Poeter, E. P.; Hill, M. C. Documentation of UCODE: A Computer Code for Universal Inverse

517

Modeling; 1998. 24 ACS Paragon Plus Environment

Page 25 of 32

Environmental Science & Technology

Manuscript submitted to Environmental Science & Technology 518

30.Kretzschmar, J.; Jordan, N.; Brendler, E.; Tsushima, S.; Franzen, C.; Foerstendorf, H.;

519

Stockmann, M.; Heim, K.; Brendler, V., Spectroscopic evidence for selenium(IV) dimerization in

520

aqueous solution. Dalton Transactions 2015, 44, (22), 10508-10515.

521

31.Jordan, N.; Foerstendorf, H.; Weiss, S.; Heim, K.; Schild, D.; Brendler, V., Sorption of

522

selenium(VI) onto anatase: Macroscopic and microscopic characterization. Geochim.

523

Cosmochim. Acta 2011, 75, (6), 1519-1530.

524

32.Fowless, A. D.; Stranks, D. R., Selenitometal Complexes . 1 . Synthesis and Characterization

525

of Selenito Complexes of Cobalt (III) and Their Equilibrium Properties in Solution. Inorganic

526

Chemistry 1977, 16, (6), 1271-1276.

527

33.Lützenkirchen, J.; Preocanin, T.; Kallay, N., A macroscopic water structure based model for

528

describing charging phenomena at inert hydrophobic surfaces in aqueous electrolyte solutions.

529

Physical Chemistry Chemical Physics 2008, 10, (32), 4946-4955.

530

34.Kosmulski, M., Surface Charging and Points of Zero Charge. CRC Press 2009; Vol. 145.

531

35.Szczepaniak, W.; Koscielna, H., Specific adsorption of halogen anions on hydrous γ-Al2O3.

532

Analytica Chimica Acta 2002, 470, (2), 263-276.

533

36.Nie, Z.; Finck, N.; Heberling, F.; Pruessmann, T.; Liu, C. L.; Lützenkirchen, J., Adsorption of

534

Selenium and Strontium on Goethite: EXAFS Study and Surface Complexation Modeling of the

535

Ternary Systems. Environmental Science & Technology 2017, 51, (7), 3751-3758.

536

37.Hiemstra, T.; Rietra, R.; Van Riemsdijk, W. H., Surface complexation of selenite on goethite:

537

MO/DFT geometry and charge distribution. Croatica Chemica Acta 2007, 80, (3-4), 313-324.

538

38.Rahnemaie, R.; Hiemstra, T.; van Riemsdijk, W. H., Geometry, charge distribution, and

539

surface speciation of phosphate on goethite. Langmuir 2007, 23, (7), 3680-3689.

540

39.Loring, J. S.; Sandström, M. H.; Norén, K.; Persson, P., Rethinking Arsenate Coordination at

541

the Surface of Goethite. Chem.-Eur. J. 2009, 15, (20), 5063-5072. 25 ACS Paragon Plus Environment

Environmental Science & Technology

Page 26 of 32

Manuscript submitted to Environmental Science & Technology 542

Tables

543 544

Table 1. Parameters for the surface species in the Best-Fit Model (surface acid-base properties

545

and Se(IV) adsorption). Surface species

∆z0

∆z1

≡AlOH−0.5

0

0

0

0

≡AlOH2+0.5

1

0

0 ≡AlOH−0.5 + H+ ⇌ ≡AlOH2+0.5

9.0

≡AlOH−0.5…Na+

0

0

1 ≡AlOH−0.5 + Na+ ⇌ ≡AlOH−0.5…Na+

−0.096

≡AlOH+0.5…Cl−

0

0

−1 ≡AlOH2+0.5 + Cl− ⇌ ≡AlOH2+0.5…Cl−

−0.176

∆z2 reaction

log K

Surface acid-base properties

Se(IV) adsorption [{(≡AlO)(≡AlOH)}SeO]0

2(≡AlOH−0.5) + 3H+ + SeO32− ⇌ 2.21

−1.21

0

30.0 0

[{(≡AlO)(≡AlOH)}SeO] + 2H2O 546

26 ACS Paragon Plus Environment

Page 27 of 32

Environmental Science & Technology

Manuscript submitted to Environmental Science & Technology 547

Figure Captions

548 549

Figure 1. (A) IR spectra of aqueous Se(IV) recorded at pD 4.0 (0.1 mol L−1 Se, 0.1 mol L−1

550

NaCl), pD 8.9 (5 × 10−3 mol L−1 Se, 0.1 mol L−1 NaCl), and pH 13.4 (8.8 × 10−2 mol L−1 Se). (B,

551

D) In situ IR spectra of the Se(IV) surface species observed after 60 min of sorption on γ-Al2O3

552

and (C, E) after subsequent flushing with background electrolyte for 60 min (B, C) under inert

553

gas conditions (N2) and (D, E) in presence of carbonate (5 × 10−4 mol L−1 DCO3−) at pD 4.9, 6.9

554

and 8.9 ([Se(IV)]init. = 5 × 10−4 mol L−1, I = 0.1 M NaCl).

555

Figure 2. (A, B) In situ IR spectra of Se(IV) sorption process on γ-Al2O3 which was

556

preconditioned with atmospherically derived carbonate and (C, D) after subsequent flushing with

557

background electrolyte for 60 min at pD 6.9. (A, C) The spectral regions of the carbonate and (B,

558

D) Se(IV) are shown. Experimental conditions were as given in Figure 1.

559

Figure 3. (A) Surface charge of the neat surface of γ-Al2O3 (m/v = 20 g L−1, I = 0.01, 0.05 and

560

0.1 mol L−1 NaCl) (□, ○, ∆ experiment;

561

Al2O3 (m/v = 0.25 g L−1, I = 0.005 and 0.01 mol L−1 NaCl) (■ experiment; ______ fit).

562

Figure 4. (A) Se(IV) sorption edges onto γ-Al2O3 ([SeIV]initial = 1 × 10−5 and 1 × 10−4 mol L−1, I =

563

0.1 and 0.01 mol L−1 NaCl; ●, ▲ experiment;

564

Al2O3 ([SeIV]initial = 0, 1 × 10−5 and 1 × 10−4 mol L−1, 0.1 and 0.01 mol L−1 NaCl). (●, ▲, □

565

experiment;

566

chosen for all experiments.

______

______

fit). (B) Zeta potential of the neat surface of γ-

______

fit). (B) Zeta potential of the surface of γ-

fit). Mass/volume ratio: m/v = 0.25 g L−1, 3 days of shaking under N2 was

567

27 ACS Paragon Plus Environment

Environmental Science & Technology

Page 28 of 32

Manuscript submitted to Environmental Science & Technology 568

Figures

569

28 ACS Paragon Plus Environment

Page 29 of 32

Environmental Science & Technology

aq. Se(IV) 817 749 734 pD 4.0 pD 8.9 814 pH 13.4

A

851 807 pD 4.9 pD 6.9 pD 8.9

762 B 756 747

844

Absorp. / a.u.

Manuscript submitted to Environmental Science & Technology

Absorption / OD

↑ Sorption inert atm. ↓ Desorption C 747 762 756 757

pD 4.9 pD 6.9 pD 8.9

D

752 747

844

3 mOD

↑ Sorption + 0.5 mM DCO–3 ↓ Desorption E

747 1000 900 800 700 Wavenumber / cm–1

570

600

Figure 1.

571

29 ACS Paragon Plus Environment

Environmental Science & Technology

Page 30 of 32

Manuscript submitted to Environmental Science & Technology

B

757

Se(IV)-Sorption

C

1392

D

3 mOD

1 mOD

1504

Se(IV)-Desorption

1600 1400 1200 800 –1 Wavenumber / cm

572

Absorption / OD

Absorption / OD

A

600

Figure 2.

573

30 ACS Paragon Plus Environment

Page 31 of 32

Environmental Science & Technology

Manuscript submitted to Environmental Science & Technology

0.25

A

σ / C⋅⋅m–2

0.20

I = 0.1 M I = 0.05 M I = 0.01 M

0.15 0.10 0.05 0.00

Zeta potential / mV

-0.05 50 40 30 20 10 0 -10 -20 -30 -40 -50

B

I = 0.01 M I = 0.005 M Solid lines: modeling 5

574

6

7

8 pH

9

10

11

Figure 3.

575

31 ACS Paragon Plus Environment

Environmental Science & Technology

Page 32 of 32

Manuscript submitted to Environmental Science & Technology

I/M [Se(IV)] 0.01 0.1 10–5 M 10–4 M

Se(IV) sorbed / %

100

A

80 60 40 20

Zeta potential / mV

0 50 40 30 20 10 0 -10 -20 -30

B

0 M Se(IV); I = 0.01 M solid lines: modeling 5

576

6

7

8 pH

9

10

Figure 4.

32 ACS Paragon Plus Environment