Arsenic Incorporation in Synthetic Struvite - American Chemical Society

Subscriber access provided by BOSTON UNIV

Article

Arsenic incorporation in synthetic struvite (NH4MgPO4•6H2O): A synchrotron XAS and single-crystal EPR study Jinru Lin, Ning Chen, and Yuanming Pan Environ. Sci. Technol., Just Accepted Manuscript • Publication Date (Web): 23 Oct 2013 Downloaded from http://pubs.acs.org on November 1, 2013

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 28

Environmental Science & Technology

1

1 2 3 4 5 6 7 8 9

Arsenic incorporation in synthetic struvite (NH4MgPO4·6H2O): A synchrotron XAS and single-crystal EPR study Jinru Lin1, Ning Chen1,2, Yuanming Pan1,* 1 Department of Geological Sciences, University of Saskatchewan, Saskatoon, SK S7N 5E2, Canada 2 Canadian Light Source, University of Saskatchewan, Canada

10

KEYWORDS: struvite, wastewaters, fertilizer, arsenic, ICPMS analysis, µ-SXRF mapping,

11

synchrotron XAFS, single-crystal EPR

12 13 14

ABSTRACT

15

Struvite, a common biomineral and increasingly important fertilizer recovered from wastewater

16

treatment plants, is capable of sequestering a wide range of heavy metals and metalloids,

17

including arsenic. Inductively coupled plasma mass spectrometric (ICPMS) analyses and

18

microbeam synchrotron X-ray flurorescence (µ-SXRF) mapping show that struvite formed under

19

ambient conditions contains up to 547±15 ppm As and that the uptake of As is controlled by pH.

20

Synchrotron As K-edge XANES spectra measured at 20 K show that As5+ is the predominant

21

oxidation state in struvite, irrespective of Na2HAsO4·7H2O or NaAsO2 as the source for As.

22

Modeling of As K-edge EXAFS data suggest that local structural distortion associated with the

23

substitution of As5+ for P5+ in struvite reaches up to 3.75 Å. Single-crystal electron paramagnetic

24

resonance (EPR) spectra of gamma-ray-irradiated struvite disclose five [AsO3]2- radicals and one

25

[AsO4]2- radical. These arsenic-centered oxyradicals are all readily attributed to form from

26

diamagnetic [AsO4]3- precursors during irradiation, providing further support for exclusive

27

incorporation and local structural expansion beyond the first shell of As5+ at the P site in struvite.

ACS Paragon Plus Environment


Page 2 of 28

2 28

29

INTRODUCTION Wastewaters (e.g., domestic, commercial and industrial sewages, agricultural and surface

30 31

runoffs, and mine tailings) often contain high levels of phosphate ions. They, if discharged

32

without treatment, result in eutrophication of receiving rivers and lakes, threatening to the

33

environment and human health.1, 2 Another common nutrient in wastewaters is nitrogen that

34

exists in high abundances in human and livestock wastes and poultry manure.3, 4 Therefore,

35

recovery of phosphate and nitrogen from wastewaters is not only necessary for the sake of

36

environmental protection but also can be used as new fertilizers with significant economic

37

merits, making the wastewater treatment industry more sustainable. One common product from

38

spontaneous precipitation of phosphate- and nitrogen-rich wastewaters is struvite

39

(NH4MgPO4·6H2O), which is an important biomineral present as a major constituent of kidney

40

and urinary stones in human and animals and is often found as a major constituent in guano

41

deposits, soils and organic-rich sediments. 5, 6 Struvite containing both nitrogen and phosphorus

42

has obvious advantages over other fertilizers and has been the subject of numerous recent

43

studies, especially with respect to the sustainable treatment of domestic or animal wastewaters.1-

44

4, 7-13

45

Struvite crystallizes in the space group Pmn21 and consists of isolated PO4 tetrahedra

46

(site symmetry m), Mg(H2O)6 octehedra (m) and (NH4)O10 irregular polyhedra (m), held together

47

by a complex network of hydrogen bonds (Fig. 1).6, 14 One major concern of struvite as fertilizer

48

is that it has been suggested to sequester a wide variety of toxic heavy metals and metalloids

49

such as As, Cd, Co, Cr, Cs, Cu and Ni.14-26 For example, accommodation of As in struvite is

50

suggested by the presence of its arsenate analogs (NH4MgAsO4·6H2O and CsMgAsO4·6H2O).14,


Page 3 of 28


3 51

19

52

removal of As and other heavy metals and metalloids from wastewaters.27, 28 On the other hand,

53

direct use of struvite recovered from wastewaters as fertilizer, without removal of As and other

54

heavy metals and metalloids, represents a potential source of contamination.

55

Not surprisingly, spontaneous precipitation of struvite has been proposed to be useful for the

The presence of arsenate analogs is not proof for arsenic incorporation in struvite

56

precipitated spontaneously from wastewaters under ambient conditions. For example, apatites

57

Ca5(PO4)3(F,OH,Cl) have arsenate analogs Ca5(AsO4)3(F,OH,Cl), which are also known to form

58

extensive solid solutions at elevated temperatures.29-31 However, Bothe and Brown (1999) noted

59

that there is no solid solution between Ca5(PO4)3(OH) and Ca5(AsO4)3(OH) precipitated under

60

ambient conditions.32 In this contribution we report on the synthesis of struvite with

61

Na2HAsO4·7H2O or NaAsO2 in the starting materials over a range of pH values under ambient

62

conditions. Our struvite crystals reaching 5 mm in the maximum dimension allow detailed

63

analyses by use of multiple, high-sensitivity chemical (inductively coupled plasma mass

64

spectrometry, ICPMS), imaging (microbeam synchrotron X-ray fluorescence mapping, µ-

65

SXRF), and structural techniques (synchrotron X-ray absorption spectroscopy, XAS; and

66

electron paramagnetic resonance spectroscopy, EPR). These data not only confirm the uptake of

67

lattice-bound As in struvite formed under ambient conditions but also provide new insights into

68

its speciation and local structural environment, with direct relevance to struvite recovered from

69

wastewater treatment plants both for the removal of As and as a new fertilizer.

70

71

SYNTHESIS EXPERIMENTS AND CHARACTERIZATION TECHNIQUES



Page 4 of 28

4 72

Two series of synthesis experiments (I and II) by using the single diffusion gel

73

technique33-35 have been made to produce crystals of arsenic doped struvite. Five experiments of

74

Series I used mixtures of 20 mL sodium metasilicate solution (Na2SiO3) and 5 mL each of 1 M

75

ammonium dihydrogen phosphate (NH4H2PO4) and 0.01 M sodium hydrogen arsenate

76

heptahydrate (Na2HAsO4·7H2O), which were first adjusted to pH = 4, 6, 7, 8 and 9 with acetic

77

acid and then poured into test tubes to form gels. After the gel settled down in each tube, 10 mL

78

solution of 0.5 M magnesium sulphate (MgSO4) was poured into the tubes that were then

79

covered with aluminum foil and kept at room temperature for one month. Series II of only two

80

runs at pH = 6 and 8 utilized the same procedure described above and the same starting materials

81

except that 0.04 M NaAsO2 standard solution was added as the source for arsenic. Individual

82

runs at various pH values have been designated as I-4, 6, 7, 8 and 9 and II-6 and 8 hereafter.

83

Hand-picked crystals from each synthesis experiment were first washed with deionized

84

water and then pulverized for powder X-ray diffraction (PXRD) analysis on a PANalytical

85

X’pert PRO X-ray diffractometer using Co Kα1 radiation (λ =1.7809Å) at 40 kV and 45 mA,

86

and for ICPMS analyses of As with the HF-HNO3 dissolution method. Selected crystals were

87

also carbon-coated for examination on a JEOL 840A scanning electron microscope (SEM).

88

89

X-band EPR measurements

90

Single-crystal X-band EPR measurements at room temperature were made on a Bruker

91

EMX instrument at the Saskatchewan Structure Sciences Centre, University of Saskatchewan,

92

Selected crystals from both series of experiments were irradiated at room temperature in a 60Co

93

cell (dose rate = ~460 Gy/h) for 2 days. In particular, one crystal from I-8 was selected for


Page 5 of 28


5 94

detailed single-crystal EPR measurements at room temperature in four rotation planes

95

approximately parallel to the (100), (010), (001) and (101) crystal faces. Experimental

96

conditions included microwave frequencies from ~9.360 to ~9.386 GHz, a microwave power of

97

~20 mW, a modulation frequency of 100 kHz, and a modulation amplitude of 0.1 mT, a constant

98

angle interval of 5º, and a spectral resolution of ~0.137 mT (i.e., 2048 field data points over the

99

scan range from 190 to 470 mT).

100 101

Synchrotron µ-SXRF mapping and As K-edge XAS experiments

102

The µ-SXRF mapping and As K-edge XAS analyses were made at the VESPERS

103

beamline36,37 and the HXMA beamline38, respectively, at the Canadian Light Source, University

104

of Saskatchewan. µ-SXRF mapping for a crystal from I-8 (Fig. 2), which was mounted on a

105

three-way NanoMotion stage tilting at 45° azimuthally and orthogonally to the incoming beam,

106

was made with a beam of ~2 µm in diameter, a step size of 3 µm, and a constant counting time of

107

1s for each data point.

108

Arsenic K-edge (11, 867 eV) XAS data for powders obtained from selected crystals (I-8

109

and II-8) and seven model compounds (arsenopyrite, realgar, orpiment, arsenolite, synthetic

110

scorodite, and As2O3 and As2O5 from Alfa Aesar) were collected at in the fluorescence mode and

111

the transmission mode, respectively. A Si(111) monochromator crystal and Rh mirrors

112

(collimating and focusing mirrors) with a 32 element Ge detector and wiggler field of 1.9 T were

113

used during data collection. All samples were diluted with boron nitride to give a satisfactory

114

edge jump and absorption. The sizes of the scan steps for the pre-edge, XANES, and EXAFS

115

regions for all experiments were 10 eV/step, 0.25 eV/step, and 0.05 Å-1/step, respectively.

116



Page 6 of 28

6 117

RESULTS AND DISCUSSION

118

Characterization and composition of synthetic struvite

119

All synthesis experiments, except for I-4, yielded several colorless crystals of two

120

contrasting morphologies (i.e., prismatic and tabular) with the maximum dimension up to 5 mm.

121

PXRD analyses (Fig. S1) show that the prismatic crystals (Fig. S2) are struvite, whereas the

122

tabular ones are boussingaultite (NH4)2Mg(SO4)2•6H2O.

123

The µ-SXRF mapping shows that the distribution of As in struvite from I-8 is reasonably

124

homogeneous (i.e., intensity with a standard deviation of only 11%) without apparent zonation or

125

any visible As-rich inclusions (Fig. 2b). Triplicate ICPMS analyses of the struvite samples from

126

Series I experiments show an apparent trend of enhanced uptake of As with increasing pH, from

127

227±15 ppm in I-6 to 78±5, 352±33 and 547±15 ppm in I-7, I-8 and I-9, respectively (Fig. S3).

128

The standard deviation of ±9.4% obtained from ICPMS analyses of different crystals in I-8 (Fig.

129

S3) is similar to the variation in intensity present in one crystal as revealed by µ-SXRF (Fig. 2b).

130

Struvite crystals from Series II experiments contain significantly lower As contents (24 and 23

131

ppm) than their counterparts from Series I.

132 133 134

Synchrotron As K-edge XANES and EXAFS data Our initial XAS measurements at room temperature showed that the As K-edge XANES

135

spectra of struvite have notable variations between scans, suggesting significant radiation-

136

induced reduction of As5+. Subsequently, XAS measurements were made for 5 scans at 20 K by

137

using an Oxford Instruments liquid-helium cryostat. These spectra measured at 20 K, unlike their

138

counterparts collected at room temperature, do not exhibit any visible change in the line shape or

139

intensity between scans. The XANES spectra of struvite I-8 and II-8 are similar to each other and


Page 7 of 28


7 140

are characterized by a main peak centered at ~11,874 eV, which is in excellent agreement with

141

those of scorodite and As2O5 (Fig. 3a) and suggests the dominant oxidation state of As5+. This

142

suggestion is also evident in the first-derivative spectra of I-8 and II-8 (Fig. 3b).

143

The As K-edge EXAFS k3•χ(k) data of struvite I-8 and II-8 are also similar, except that

144

the former with a higher As content has better signal-to-noise ratios than the latter. These

145

EXAFS data have been analyzed for R space curve fittings based on the FEFF7 modeling39 by

146

using the XAFS data processing software WinXas 3.2. 40 Fourier transform was performed by

147

using a Gaussian window function over the k range of 2.78-14.63 Å-1 with the window parameter

148

of 30%, and the R-space data were fitted in a window of 2.79-3.73 Å (Fig. 4). The fitted

149

interatomic distance for the nearest neighbor As-O shell is 1.69 Å (Table 1), matching closely to

150

those of typical As5+.41-43 This interatomic distance is notably larger than that of the average P-O

151

bond distance in struvite (1.54 Å),6 requiring a significant expansion of the P site.

152

Beyond the nearest-neighbor As-O shell, two groups of scattering paths are included in

153

the fitting (regions I and II in Fig. 4). The first group is 3 types of multiple scatterings occurring

154

in the AsO4 tetrahedron. Following the approach of Kim et al (2013),44 the Debye-Waller factors

155

(σ2) for these paths were constrained, namely the σ2As-Oa-Ob and σ2As-Oa-As-Ob were correlated to

156

those of the nearest-neighbor As-O scattering by factors of 2 and 4, respectively. Meanwhile, the

157

σ2As-Oa-As-Ob values were changed by factors of 2 to 4 to that of the As-O path. The FEFF

158

modeling of these multiple scattering paths was based on the morphology of the AsO4

159

tetrahedron in the scorodite structure.41 Inclusion of these multiple scattering paths in the data

160

analysis is necessary for struvite, because the intensity contribution from these multiple

161

scattering is not negligible (Fig. 4). Noted that the estimated σ2 value for the nearest-neighbor

162

As-O path of struvite is slightly larger than that of scorodite,45 suggesting a larger distortion of



Page 8 of 28

8 163

the arsenate tetrahedron in the former. This local structural distortion is expected to partially

164

remove the degeneration of the multiple scattering paths As-Oa-Ob andAs-Oa-As-Ob. Therefore,

165

the path As-Oa-Ob was treated as two sets of separated paths from the scattering distances

166

perspective in fitting (Table 1), same was the path As-Oa-As-Ob. Nevertheless the FEFF

167

multiple scattering modeling on the basis of the scorodite structure is shown to be suitable.

168

The second group of scattering paths (II in Figure 4) include outer As-O paths from two

169

groups of oxygen atoms of the neighboring Mg(H2O)6 groups (Fig. 1). The calculated

170

interatomic distances for the outer As-O paths from the first group are larger than those in the

171

ideal struvite structure, suggesting that local structural distortion associated with As5+

172

incorporation reaches up to 3.75 Å. Although the absence of any strong back scattering elements

173

in this R region makes accurate analysis of the experimental data difficult, the fitted two sets of

174

the single scattering As-O paths are consistent with the extended PO4 site symmetry, supporting

175

the substitution of As5+ for P5+ in struvite.

176

177 178

Single-crystal X-band EPR spectra of struvite Single-crystal EPR measurements show that synthetic struvite does not contain any

179

paramagnetic defects before gamma-ray irradiation but is characterized by six sets of four-line

180

resonance signals all centered at the effective g value of ~2.00 (Centers I-VI in Fig. 5a) after

181

irradiation. Centers I to V are characterized by large but unequal separations of ~50-81 mT,

182

~55-82 mT, ~44-85 mT, ~47-91 mT and ~50-98 mT, respectively, and have approximately

183

similar linewidths of ~0.5 mT. The relative intensities of Centers I to V are 100:13:11:7:3. These

184

five centers are all attributable to simple spins (i.e., a single unpaired electron S=1/2) interacting


Page 9 of 28


9 185

with a 75As nucleus (i.e., nuclear spin number I = 3/2 and natural isotope abundance = 100%).

186

The individual hyperfine lines of Center II are resolved to a maximum of 4 when the external

187

magnetic field B is rotated away from symmetry axes (Fig. S4), indicative of a triclinic site

188

symmetry in orthorhombic struvite. Centers I, III, IV and V, however, are resolved to at most

189

two magnetically nonequivalent sites and therefore have monoclinic site symmetries (Fig. S5-

190

S8). At some orientations individual lines of Center I are resolved into triplets with an intensity

191

ratio of approximately 1:2:1 (Fig. 5b), characteristic of a superhyperfine structure arising from

192

interactions with two equivalent or nearly equivalent 1H nuclei.

193 194 195

The spin Hamiltonian for Centers I to V, ignoring the incompletely resolved 1H superhyperfine structure of Center I, can be written as follow: H = βeS•g•B + I•A•S + I•P•I - βnI•gn•B

(1),

196

where βe and βn are the electronic (Bohr) and nuclear magnetons, respectively; S and I are the

197

electron spin and nuclear spin operators, respectively; g is the Zeeman electron term; and A and

198

P are the nuclear hyperfine and quadrupole terms for 75As with an isotropic gn value of 0.959647.

199

All analyses have been made with the experimental axes x, y and z along the crystallographic

200

axes a, c and b, respectively. The total number of line-position data points used for the fittings of

201

Centers I to V were 1736, 1047, 1074, 838 and 948, respectively. The final values of the root-

202

mean-squares of weighed differences (RMSD) between the calculated and observed line

203

positions for the five centers are 0.132 mT, 0.167 mT, 0.158 mT, 0.167 mT and 0.193 mT,

204

respectively (Table 2), which are all less than half of the average linewidths. Spectral simulations

205

have been used to determine the two matrices A(1H) for the superhyperfine structure of Center I

206

(Table S1), which include an isotropic component a = 0.410 mT and an anisotropic component b

207

= 0. 278 mT. The directions of the unique principal A(1H) axes at (θ=30°, φ=270°) and (30°,



Page 10 of 28

10

208

90°) are close to the orientations from P to two nearest H atoms in struvite (Table S2),6 where θ

209

and φ are tilting angles from crystallographic axes b and a, respectively.

210

The best-fit matrices g and A(75As) of Centers I to V (Table S1) are approximately axial

211

in symmetry and are closely comparable to those of the well-established [AsO3]2− radical (Table

212

2).46-57 This radical is readily linked to the diamagnetic [AsO4]3− precursor and involves the sp3

213

hybridization for the unpaired electron orbital directed towards the missing O atom (Table 2).56

214

Specifically, the orientations of the unique g3 and A1 axes of Centers I, II and III (Table S1) are

215

approximately along the P-O1 bond direction (86.2°, 90°), the P-O3 bond direction (61.2°,

216

159.7°) and the P-O2 bond direction (23.1°, 270°), respectively. These [AsO3]2− radicals in

217

struvite are readily attributed to electron trapping on the central As5+ ion of the substitutional

218

[AsO4]3- group at the [PO4]3- position after removal of one of the four nearest O atoms, via the

219

following reaction:

220

[AsO4]3-

[AsO3]2- + O2-

(2).

221

Similarly, Centers IV and V are characterized by the orientations of their unique g3 and A1 axes

222

approximately along the P-O2 bond direction and may represent variants of Center I: i.e.,

223

[AsO3]2− radicals from electron trapping on substitutional [AsO4]3- group after the removal of the

224

O2 atom but involving slightly different distortions. The fact that the two O3 atoms coordinated

225

to P are symmetrically equivalent explains the presence of these [AsO3]2− radicals with three

226

distinct orientations only.

227

Center VI is characterized by 75As hyperfine coupling constants from ~1.5 to ~2.2 mT

228

(Fig. 5a) and possesses multiple and strong “forbidden” nuclear transitions arising from

229

significant nuclear quadrupole effects, typical of the well-established [AsO4]2− radical with the


Page 11 of 28


11 230

unpaired spin localized in the 2pz orbital of an O atom (Table 2).56 Unfortunately, quantitative

231

determination of the spin Hamiltonian parameters for this [AsO4]2− radical in struvite is not

232

possible owing to difficulties in tracking the angular dependence of individual lines at most

233

orientations, which are affected by not only “forbidden” nuclear transitions but also splittings

234

related to a triclinic site symmetry as well as overlapping lines from unknown centers (Fig. 5a).

235 236 237

Single-crystal EPR spectra of boussingaultite Single-crystal EPR measurements show that synthetic boussingaultite does not contain

238

any paramagnetic defect before gamma-ray irradiation but is characterized by three well-

239

resolved arsenic-centered oxyradicals after irradiation ([AsO4]2−, [AsO3]2− and [AsO2]2−; Fig.

240

5c). These three arsenic-centered oxyradicals are readily identified on the basis of their

241

diagnostic 75As hyperfine coupling constants (Table 2).46-57 Interestingly, the angular dependence

242

of the experimental line positions (i.e., roadmaps) and the best-fit spin Hamiltonian parameters

243

of the [AsO3]2− radical in boussingaultite are almost identical to those of an [AsO3]2− radical in

244

arsenate-doped struvite reported by Xu (1992). Also, the same [AsO2]2− radical (Fig. 5c) is

245

visible in the B//c spectrum reported by Xu (1992).57 Moreover, a broad signal at the central

246

magnetic field in the B//c spectrum reported by Xu (1992)57 corresponds to the well-resolved

247

[AsO4]2− radical in Figure 5c. These similarities lead us to suggest that “arsenate-doped struvite”

248

investigated by Xu (1992)57 was most likely boussingaultite. Indeed, PXRD analyses

249

demonstrate that crystals from our synthesis experiment using the same slow evaporation

250

method57 (i.e., equal volumes of the equimolar solutions of NH4H2PO4 and MgSO4 with the

251

addition of one molar percent (NH4)3AsO4) are boussingaultite.

252



Page 12 of 28

12 253

Insights into arsenate uptake in struvite

254

ICPMS analyses confirm µ-SXRF mapping that struvite formed under ambient conditions

255

contains up to 547±15 ppm As. Moreover, modeling of synchrotron As K-edge EXAFS data and

256

single-crystal EPR analyses of arsenic-centered oxyradicals demonstrate lattice-bound As in

257

struvite. In particular, the angular dependences of the experimental EPR spectra of all five

258

[AsO3]2− radicals, including agreements between their experimental lines with both the site

259

symmetries and the Laue crystal symmetry of struvite, rule out origins from As-bearing

260

inclusions or surface absorbents. Moreover, the close matches in directions between the unique g

261

and A(75As) axes with specific P-O bonds of the [PO4]3− group are unambiguous evidence for

262

incorporation of As5+ at the P site in the struvite lattice. In addition, the observed 1H

263

superhyperfine structure of Center I (Fig. 5b and Table S2) provides further support for the

264

location of As5+ at the P5+ site in the struvite structure.

265

Our results for the incorporation of up to 547±15 ppm As in struvite differ from those of

266

Bothe and Brown (1999), 32 whose did not detect any solid solution between Ca5(AsO4)3(OH)

267

and Ca5(PO4)3(OH) under ambient conditions. Elevated As contents in apatites from marine

268

phosphorite deposits are well known.58 The failure to detect any solid solution between

269

Ca5(AsO4)3(OH) and Ca5(PO4)3(OH) in Bothe and Brown (1999) 32 is most likely attributable to

270

the PXRD technique of limited sensitivity. Our observation of an enhanced As uptake in struvite

271

with increasing pH in Series I experiments is consistent with the increased stability of [AsO4]3−

272

relative to [HAsO4]2− and [H2AsO4]− at high pH in aqueous environments 58, 59 but must be

273

viewed with caution, because thermodynamic equilibrium was not demonstrated in our synthesis

274

experiments.


Page 13 of 28


13 275

The most salient feature from this combined synchrotron XAFS and EPR study is the

276

exclusive incorporation of As5+ in struvite, irrespective of Na2HAsO4·7H2O or NaAsO2 as the

277

source. In particular, the detection of only As5+ in struvite from Series II experiments

278

demonstrates that As3+ is highly unfavorable in this mineral. Here As5+ in Series II experiments

279

probably resulted from oxidation reaction(s) related to bacterial activity in the unsealed tubes.46

280

Other possible causes for the presence of As5+ in Series II experiments include 1) impurity in the

281

starting material and 2) abiotic oxidation related to atmospheric O2. The incorporation of only

282

As5+ in struvite differs from the accommodation of mixed As3+ and As5+ in gypsum formed

283

under similar experimental conditions.46

284

Although our design for the single diffusion gel technique does not allow any control or

285

monitoring of the oxidation state, the presence of mixed As3+ and As5+ species during our

286

experiments is confirmed by the single-crystal EPR spectra of co-precipitated boussingaultite

287

(Fig. 5c). The [AsO4]2− and [AsO3]2− radicals in boussingaultite, similar to their counterparts in

288

struvite, are readily linked to the diamagnetic [AsO4]3− group (i.e., forming from hole trapping

289

on one of the four oxygen atoms for the former and electron trapping on the central As atom after

290

removal of an oxygen atom for the latter), indicating the occurrence of As5+ .56 The [AsO2]2−

291

radical is known to form from the [AsO3]3− group by electron trapping on the As 4pz orbital

292

(Table 2) after removal of an O atom, confirming the presence of As3+ .56 The absence of any

293

[AsO2]2− radical in the EPR spectra of gamma-ray-irradiated struvite, considering the sensitivity

294

of this technique,55,56 suggests that As3+ in struvite, if present, is unlikely to exceed 1 ppm.

295

Local structural expansion beyond the first shell of substituting As5+ in struvite suggested

296

by modeling of EXAFS data is also corroborated by EPR results. For example, the anisotropic

297

component b from the 1H superhyperfine structure of Center I, using the point-dipole model [b =



Page 14 of 28

14 298

−(2µ0/4π)(gβegnβn/r3), where gn = 5.5856912 for 1H], predicts As-H distances of 2.76 Å, which

299

are significantly longer than those from P to the two nearest H atoms determined by neutron

300

diffraction studies (2.67 Å).6

301

Another interesting question is whether As5+ in struvite occupies the Mg octahedral site

302

or not, because octahedral arsenate groups, albeit rare, have been reported to occur in synthetic

303

compounds 60 and minerals61. Our EXAFS modeling suggests that as little as 5% As5+ at the

304

octahedral site (As-O = 1.83 Å)60 would be visible in experimental spectra. The fact that none of

305

the arsenic-centered oxyradicals is consistent with the orientation of the Mg(H2O)6 group (Tables

306

S1 and S2) also refutes any accommodation of As5+ at the octahedral site in struvite, even at the

307

level of the exceptionally sensitive EPR technique.

308

Finally the present study of arsenic-doped struvite formed under ambient conditions not

309

only provides new insights into the mechanism of As5+ incorporation but also raises serious

310

questions about the current use of this material as fertilizer. In particular, struvite precipitated

311

from sewage and wastewater treatment plants, without removal of As (and other heavy metals

312

and metalloids), is a potential source of contamination. Our recognition of As3+ exclusion from

313

struvite suggests that precipitation under reduced conditions is a possible route to minimize As.

314 315 316 317 318


Page 15 of 28


15 319 320

FIGURES

321 322 323

324 325 326 327 328

Figure 1. Crystal structure of struvite projection along [010] showing the local structural environment of the P site.



Page 16 of 28

16 329

330

331 332


Page 17 of 28


17 333 334 335

Figure 2. (a) SEM image of a prismatic crystal of synthetic struvite (I-8) showing three areas (1, 2 and 3) of notable beam damage from µ-SXRF mapping; (b) µ-SXRF map showing the distribution of As in an 150 µm x 150 µm area (outlined in a).



Page 18 of 28

18 336

337 338 339 340 341

Figure 3. (a) As K-edge XANES spectra of struvite samples (I-8 and II-8) and model compounds; and (b) corresponding first-derivative XANES spectra.


Page 19 of 28


19 342

343 344 345 346 347 348 349

Figure 4. R-space curve fittings for As K-edge EXAFS spectra of Struvite. The Fourier transforms of k3•χ(k) spectra were performed over the k ranges of 2.78-14.63 Å-1 and Rspace fitting windows of 2.79-3.73 Å. Also included in inserts are the k3•χ(k) spectra in comparison with those from R-space fittings.



Page 20 of 28

20 350

351 352


Page 21 of 28


21 353 354 355

356 357 358 359 360 361 362 363 364 365

Figure 5. Representative single-crystal EPR spectra of gamma-ray-irradiated struvite measured at: a) B//c showing six arsenic-centered oxyradicals with characteristic 75As hyperfine structures (labeled I to VI); and unknown peaks are marked by arrows; and b) B∧b = ~30° in the high magnetic field showing that Center I contains a well-resolved superhyperfine structure arising from interactions with two equivalent 1H nuclei. Also shown for comparison is c) a single-crystal EPR spectrum of gamma-ray-irradiated boussingaultite measured at B//c showing three well-resolved arsenic-centered oxyradicals ([AsO4]2−, [AsO3]2− and [AsO2]2−) .



Page 22 of 28

22 366 367 368 369

370 371

Table 1. Local environment of As in struvite from R space fittings of As K-edge EXAFS data Paths

CN

R (Å)

σ2 (Å2)

E0 (eV)

As-O

4.0§

1.69

0.0018

2.5

As-Oa-Ob

6.0

§

2.90

0.0036

2.5

As-Oa-Ob

6.0§

3.19

0.0036

2.5

As-Oa-As-Oa

4.0§

3.58

0.0072

2.5

As-Oa-As-Ob

6.0§

3.25

0.0040

2.5

As-Oa-As-Ob

6.0§

3.47

0.0050

2.5

As-O

3.3

3.75

0.0100

2.5

As-O

6.0

4.24

0.0100

2.5

Oa and Ob mark nonequivalent oxygen atoms; § denotes fixed values.

372


Page 23 of 28


23 373 374 375 376

Table 2. Spin Hamiltonian parameters of [AsO3]2−, [AsO4]2− and [AsO2]2− radicals in struvite, boussingaultite and other crystalline hosts

Host

struvite (I) struvite (II) struvite (III) struvite (IV) struvite (V) boussingaultite calcite (77K) Na2HAsO4·7H2O KH2AsO4 (K) KH2AsO4 (L) KH2AsO4 (M) betaine arsenate haidingerite(I) haidingerite(II) gypsum

scheelite (77K) (NH4)2HPO4 hemimorphite boussingaultite

g1

2.00431(4) 2.00467(6) 2.00419(7) 2.00349(6) 2.00330(8) 2.00603(4) 2.00195(5) 2.005(2) 1.996(5) 1.999(5) 2.000(5) 2.001(4) 1.999 1.998 2.00537(4)

2.0470 2.030(5) 2.02407(1) 2.03760(6)

g2

g3

A1/geβ e (mT) [AsO3]2− 72.950(6) 76.457(9) 74.620(6) 79.976(9) 84.771(9) 63.906(5) 93.73(1) 72.5(1) 91.8 76.5 66.1 65.9(4) 85.23 94.03 69.65(1)

2.00399(4) 2.00298(7) 2.0034(1) 2.00322(6) 2.00287(9) 2.00538(4) 2.00195(5) 2.005(2) 1.996(5) 1.999(5) 2.000(5) 1.998(4) 1.999 1.993 2.00371(5)

2.00154(4) 2.00122(6) 2.00161(5) 2.00110(6) 2.00104(8) 2.00218(4) 2.00162(5) 2.004(2) 1.996(5) 1.999(5) 2.000(5) 1.998(4) 1.999 1.993 2.00134(3)

2.0122 2.023(5) 2.00982(1) 2.00325(6)

[AsO4]2− 2.00070 2.30 2.016(5) 3.1(1) 2.00326(1) 2.158(1) 2.00281(6) 3.020(6)

A2/geβ e (mT)

A3/geβ e (mT)

Unpaired spin (%) 4s 4pz 4px

Ref.

56.581(5) 60.210(9) 58.271(9) 62.975(6) 67.64(1) 49.391(5) 75.24(1) 56.7(2) 75 60.8 50 51.6(4) 67.26 75.63 53.26(1)

56.576(5) 60.059(9) 58.172(9) 62.937(7) 67.62(1) 49.016(6) 75.24(1) 56.7(1) 74.2 60.4 49.6 51.6(4) 67.26 75.63 53.12(1)

11.9 12.5 12.2 13.1 14.0 10.3 15.6 11.9 15.4 12.6 10.6 10.8 13.9 15.6 11.2

45.8 45.7 45.9 47.7 48.0 41.1 51.8 44.3 49.3 45.1 46.0 40.1 53.2 51.6 46.2

0.3

TS TS TS TS TS TS 48 47 49 49 49 50 51 51 33

1.9 1.7(1) 1.861(1) 2.332(7)

1.87 0.7(1) 1.678(1) 1.405(8)

0.39 0.35 0.36 0.43

1.0 3.3 0.7 3.2

2.2

52 53 54 TS

0.3 0.2

0.8

2.1 0.9 0.9

[AsO2]2− boussingaultite 2.0115(1) 1.9996(1) 1.9946(1) 16.86(1) -8.14(1) -8.44(1) 0.01 70.4 0.7 gypsum 2.01484(2) 1.99962(2) 1.9958(1) 16.96(1) -7.53(1) -7.95(2) 0.09 69.2 0.9 377 Reference TS is this study. The distribution of unpaired spins in the 4s and 4p orbitals of arsenic has been calculated 378 to the atomic a0 and b0 values of 523.11 and 11.905 mT, respectively.55

379 380


TS 46


Page 24 of 28

24 381 382

AUTHOR INFORMATION

383

*Phone: 306 966 5699, email: [email protected]

384

Author contributions: The manuscript was written through contributions of all authors. All

385

authors have given approval to the final version of the manuscript.

386

Notes: The authors declare no competing financial interest.

387 388

ACKNOWLEDGMENT

389 390 391 392 393 394 395

We thank two reviewers and Prof. Jorge L. Gardea-Torresdey for incisive criticisms and helpful suggestions, Prof. George Demopoulos of McGill University for provision of synthetic scorodite and the Natural Sciences and Engineering Research Council (NSERC) for financial support. Canadian Light Source is supported by NSERC, the National Research Council Canada, the Canadian Institutes of Health Research, the Province of Saskatchewan, Western Economic Diversification Canada, and the University of Saskatchewan.

396

ABBREVIATIONS

397 398 399 400

ICPMS, inductively coupled plasma mass spectrometry; µ-SXRF, microbeam synchrotron X-ray flurorescence imaging; XAS, X-ray absorption spectroscopy; XANES, X-ray absorption near edge structure; EXAFS, extended X-ray absorption fine structure; EPR, electron paramagnetic resonance spectroscopy.

401 402

ASSOCIATED CONTENT

403 404 405 406 407 408 409 410

Supporting Information Available Available supporting information includes Figure S1 showing representative PXRD patterns of synthetic struvite and boussingaultite, Figure S2 illustrating a prismatic crystal of synthetic struvite, Figure S3 showing the arsenic contents in synthetic struvite as a function of pH, Figures S4 to S8 depicting EPR line-position data (roadmaps) Centers I to VI, and Table S1 summarizing complete Spin Hamiltonian parameters of five [AsO3]2− radicals in struvite at 295 K. This material is available free of charge via the Internet at http://pubs.acs.org/.

411 412 413 414

(1) Parsons, S. A.; Smith, J. A. Phosphorus removal and recovery from municipal wastewaters. Elements 2008, 4 (2), 109-112.

REFERENCES


Page 25 of 28


25 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464

(2) de-Bashan, L. E.; Bashan, Y. Recent advances in removing phosphorus from wastewater and its future use as fertilizer (1997-2003). Water Resour. 2004, 38 (19), 4222-4246. (3) Perera, P. W. A.; Han, Z. Y.; Chen, Y. X.; Wu, W. X. Recovery of nitrogen and phosphorous as struvite from swine waste biogas digester effluent. Biomed. Environ. Sci. 2007, 20 (5), 343-350. (4) Zhang, W.; Lau, A. Reducing ammonia emission from poultry manure composting via struvite formation. J. Chem. Technol. Biotechnol. 2007, 82 (6), 598-602. (5) Frost, R. L.; Weier, M. L.; Erickson, K. L. Thermal decomposition of struvite implications for the decomposition of kidney stones. J. Therm. Anal. Calorim. 2004, 76 (3), 1025-1033. (6) Ferraris, G.; Fuess, H.; Joswig, W. Neutron diffraction study of MgNH4PO4·6H2O (struvite) and survey of water molecules donating short hydrogen bonds. Acta Crystallogr. 1986, B42, 253-258. (7) Doyle, J. D.; Parsons, S. A. Struvite formation, control and recovery. Water Res. 2002, 36 (16), 3925-3940. (8) El Diwani, G.; El Rafie, S.; El Ibiari, N. N.; El-Aila, H. I. Recovery of ammonia nitrogen from industrial wastewater treatment as struvite slow releasing fertilizer. Desalination 2007, 214 (1-3), 200214. (9) Jaffer, Y.; Clark, T. A.; Pearce, P.; Parsons, S. A. Potential phosphorus recovery by struvite formation. Water Res. 2002, 36 (7), 1834-1842. (10) Le Corre, K. S.; Valsami-Jones, E.; Hobbs, P.; Parsons, S. A. Phosphorus recovery from wastewater by struvite crystallization: A review. Crit. Rev. Env. Sci. Technol. 2009, 39 (6), 433-477. (11) Negrea, A.; Lupa, L.; Negrea, P.; Ciopec, M.; Muntean, C. Simultaneous removal of ammonium and phosphate ions from wastewaters and characterization of the resulting product. Chem. Bull. "POLITEHNICA" Univ. (Timisoara) 2010, 55 (69), 2, 136-142. (12) Rahaman, M. S. Phosphorus recovery from wastewater through struvite crystallization in a fluidized bed reactor: Kinetics, hydrodynamics and performance. Ph.D. Dissertation, University of British Columbia, Vancouver, BC, 2009. (13) Tünay, O.; Kabdasli, I.; Orhon, D.; Kolçak, S. Ammonia removal by magnesium ammonium phosphate precipitation in industrial wastewaters. Water Sci. Technol. 1997, 36 (2-3), 225-228. (14) Weil, M. The struvite-type compounds M [Mg(H2O)6](XO4), where M = Rb, Tl and X = P, As. Cryst. Res. Technol. 2008, 43 (12), 1286-1291. (15) Sohel, N.; Kanaroglou, P. S.; Persson, L. A.; Haq, M. Z.; Rahman, M.; Vahter, M. Spatial modelling of individual arsenic exposure via well water: Evaluation of arsenic in urine, main water source and influence of neighbourhood water sources in rural Bangladesh. J. Environ. Monit. 2010, 12 (6), 13411348. (16) Arai, Y.; Lanzirotti, A.; Sutton, S.; Davis, J. A.; Sparks, D. L. Arsenic speciation and reactivity in poultry litter. Environ. Sci. Technol. 2003, 37 (18), 4083-4090. (17) Hunger, S.; Sims, J. T.; Sparks, D. L. Evidence for struvite in poultry litter: Effect of storage and drying. J. Environ. Qual. 2008, 37 (4), 1617-1625. (18) Aguilera, I.; Daponte, A.; Gil, F.; HernÃ¡ndez, A. F.; Godoy, P.; Pla, A.; Ramos, J. L. Urinary levels of arsenic and heavy metals in children and adolescents living in the industrialised area of Ria of Huelva (SW Spain). Environ. Int. 2010, 36 (6), 563-569. (19) Weil, M. Redetermination of the cubic struvite analogue Cs[Mg(OH2)6](AsO4). Acta Crystallogr. 2009, E65 (1), i2. (20) Chowdari, B. V. R.; Ravisekhar, Y.; Muralikrishna, C. Radiation damage of Ni2+-doped NH4MgPO4 · 6H2O. Phys. Status Solidi B 1982, 109 (1), 315-324. (21) Agarwal, O. P.; Chand, P. Absorption spectra of vanadyl ion doped in MgNH4PO4·6H2O (struvite) crystal. Solid State Commun. 1984, 52 (4), 417-421. (22) Sivaprasad, P.; Ramesh, K.; Reddy, Y. P. Optical and EPR studies of Cu2+ in MgNH4PO4·6H2O. Phys. Status Solidi A 1990, 118 (2), K103-K106. (23) Chand, P.; Agarwal, O. P. Electron paramagnetic resonance study of doped synthetic crystals of struvite and its zinc analogue. Spectrochim. Acta, Part A 1991, 47 (6), 775-783.



Page 26 of 28

26 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515

(24) Rao, S. N.; Sivaprasad, P.; Reddy, Y. P. Optical absorption studies on cobalt doped struvite. Radiat. Eff. Defects Solids 1991, 116 (4), 369-373. (25) Ravikumar, R. V. S. S. N.; Madhu, N.; Reddy, B. J.; Reddy, Y. P.; Rao, P. S. Spectroscopic investigations on vanadyl doped cadmium struvite. Phys. Scr. 1997, 55 (5), 637. (26) Sougandi, I.; Venkatesan, R.; Rajendiran, T. M.; Rao, P. S. Identification of VO(II) at substitutional and interstitial locations in magnesium ammonium phosphate hexahydrate: A single crystal EPR study. Phys. Scr. 2003, 67 (2), 153. (27) Uysal, A.; Yilmazel, Y. D.; Demirer, G. N. The determination of fertilizer quality of the formed struvite from effluent of a sewage sludge anaerobic digester. J. Hazard. Mater. 2010, 181 (1-3), 248-254. (28) Ronteltap, M.; Maurer, M.; Gujer, W. The behaviour of pharmaceuticals and heavy metals during struvite precipitation in urine. Water Res. 2007, 41 (9), 1859-1868. (29) Pan, Y.; Fleet, M. E. Compositions of the apatite-group minerals: Substitution mechanisms and controlling factors. Rev. Mineral. Geochem. 2002, 48 (1), 13-49. (30) Lee, Y. J.; Stephens, P. W.; Tang, Y.; Li, W.; Phillips, B. L.; Parise, J. B.; Reeder, R. J. Arsenate substitution in hydroxylapatite: Structural characterization of the Ca5(PxAs1–xO4)3OH solid solution. Am. Mineral. 2009, 94 (5-6), 666-675. (31) Biagioni, C.; Pasero, M. The crystal structure of johnbaumite, Ca5(AsO4)3OH, the arsenate analogue of hydroxylapatite. Am. Mineral. in press, 2013. (32) Bothe, J. V.; Brown, P. W. Arsenic immobilization by calcium arsenate formation. Environ. Sci. Technol. 1999, 33 (21), 3806-3811. (33) Hektisch, H. K.; Dennis, J.; Hanoka, J. I. Crystal growth in gels. J. Phys. Chem. Solids 1965, 26 (3), 493-496. (34) Henisch, H. K. Crystal Growth in Gels. The Pennsylvania State University Press, University Park: London, 1970. (35) Patel, A. R.; Rao, A. V. Crystal growth in gel media. Bull. Mater. Sci. 1982, 4 (5), 527-548. (36) Feng, R.; Gerson, A.; Ice, G.; Reininger, R.; Yates, B.; McIntyre, S. VESPERS: A beamline for combined XRF and XRD measurements. AIP Conf. Proc. 2007, 879, 872-874. (37) Lin, J.; Pan, Y.; Chen, N.; Mao, M.; Li, R.; Feng, R. Arsenic incorporation in colemanite from borate deposits: Data from ICP-MS, micro-S-XRF, XAS and EPR analyses. Can. Mineral. 2011, 49 (3), 809-822. (38) Jiang, D. T.; Chen, N.; Demopoulos, G. P.; Rowson, J. W. Response to the comment by D. Paktunc on "Structural characterization of poorly-crystalline scorodite, Iron(III)–arsenate co-precipitates and uranium mill neutralized raffinate solids using X-ray absorption fine structure spectroscopy". Geochim. Cosmochim. Acta 2010, 74 (15), 4597-4602. (39) Rehr, J. J.; Albers, R. C. Theoretical approaches to X-ray absorption fine structure. Rev. Mod. Phys. 2000, 72 (3), 621-654. (40) Ressler, T. WinXAS: A new software package not only for the analysis of energy-dispersive XAS data. J. Phys. IV 1997, 7 (C2), 269-270. (41) Kitahama, K.; Kiriyama, R.; Yoshihisa, B. Refinement of the crystal structure of scorodite. Acta Crystallogr. 1975, B31, 322-324. (42) Foster, A. L.; Brown, G. E.; Tingle, T. N.; Parks, G. A. Quantitative arsenic speciation in mine tailings using X-ray absorption spectroscopy. Am. Mineral. 1998, 83, 553-568. (43) Moldovan, B. J.; Jiang, D. T.; Hendry, M. J. Mineralogical characterization of arsenic in uranium mine tailings precipitated from iron-rich hydrometallurgical solutions. Environ. Sci. Technol. 2003, 37 (5), 873-879. (44) Kim, E.; Chen, N.; Arthur, Z.; Warner, J.; Demopoulos, G. P.; Rowson, J. W.; Jiang, D. T. Multiple scattering Debye-Waller factors for arsenate. J. Phys. Conf. Ser. 2013, 430. (45) Chen, N.; Jiang, D. T.; Cutler, J.; Kotzer, T.; Jia, Y. F.; Demopoulos, G. P.; Rowson, J. W. Structural characterization of poorly-crystalline scorodite, iron(III)–arsenate co-precipitates and uranium mill neutralized raffinate solids using X-ray absorption fine structure spectroscopy. Geochim. Cosmochim. Acta 2009, 73 (11), 3260-3276.


Page 27 of 28


27 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552

(46) Lin, J.; Chen, N.; Nilges, M. J.; Pan, Y. Arsenic speciation in synthetic gypsum (CaSO4·2H2O): A synchrotron XAS, single-crystal EPR, and pulsed ENDOR study. Geochim. Cosmochim. Acta 2013, 106, 524-540. (47) Lin, W. C.; McDowell, C. A. Electron spin resonance of an X-ray irradiated single crystal of disodium hydrogen arsenate, Na2HAsO4·7H2O. Mol. Phys. 1964, 7 (3), 223 - 234. (48) Serway, R. A.; Marshall, S. A. Electron spin resonance absorption spectrum of the AsO32molecule ion in γ-irradiated single-crystal calcite. J. Chem. Phys. 1966, 45 (6), 2309-2314. (49) Dalal, N. S.; Dickson, J. R.; McDowell, C. A. Electron paramagnetic resonance studies of Xirradiated KH2AsO4, KD2AsO4, RbH2AsO4, RbD2AsO4, CsH2AsO4, NH4H2AsO4, and ND4D2AsO4 (ferroelectrics and antiferroelectrics). J. Chem. Phys. 1972, 57 (10), 4254-4265. (50) Pöppl, A; Tober, O.; Völkelel, G. ESR study of the AsO32– radical in γ-irradiated betaine arsenate. Phys. Status Solidi B 1994, 183, K63-K66. (51) Gilinskaya, L. G.; Afanas'eva, V. I. Two forms of the AsO32− radical in γ-irradiated heidingerite CaHAsO4·H2O. J. Struct. Chem. 1981, 22 (4), 640-642. (52) Edwards, P. R.; Subramanian, S.; Symons, M. C. R. Electron spin resonance studies of vanadate, niobate, phosphate, and arsenate in γ-irradiated calcium molybdate and calcium tungstate. Chem. Commun. (London) 1968, 799-799. (53) Subramanian, S.; Murty, P. N.; Murty, C. R. K. EPR studies of some arsenic and phosphorus oxyradicals. J. Phys. Chem. Solids 1977, 38, 825-829. (54) Mao, M.; Lin, J.; Pan, Y. Hemimorphite as a natural sink for arsenic in zinc deposits and related mine tailings: Evidence from single-crystal EPR spectroscopy and hydrothermal synthesis. Geochim. Cosmochim. Acta 2010, 74, 2943-2956. (55) Weil, J. A.; Bolton, J. R. Electron Paramagnetic Resonance: Elementary Theory and Practical Applications, 2rd, ed.; John Wiley & Sons: New York, 2007. (56) Pan, Y. Arsenic speciation in rock-forming minerals determined by EPR spectroscopy. In Arsenic: Sources, Environmental Impacts, Toxicity and Human Health: A Medical Geology Perspective; Masotti, A., Ed.; Nova Science Publishers Inc: 2013; pp 39-52. (57) Xu, R. EPR studies of AsO32− centers in MgNH4PO4·6H2O (struvite) crystals. Phys. Status Solidi B 1992, 172 (1), K15-K18. (58) Smedley, P. L.; Kinniburgh, D. G. A review of the source, behaviour and distribution of arsenic in natural waters. Appl. Geochem. 2002, 17 (5), 517-568. (59) Reeder, R. J.; Schoonen, M. A. A.; Lanzirotti, A. Metal speciation and its role in bioaccessibility and bioavailability. Rev. Mineral. Geochem. 2006, 64 (1), 59-113. (60) Schwendtner, K.; Kolitsch, U. Octahedral As in M+ arsenates-architecture and seven new members. Acta Crystallogr. 2007, B63 (2), 205-215. (61) Groat, L. A.; Evans, J.; J, C.; McCammon, C.; Houzar, S. Fe-rich and As-bearing vesuvianite and wiluite from Kozlov, Czech Republic. Am. Mineral. 2013, 98, 1330-1337.

553 554




Page 28 of 28

Arsenic Incorporation in Synthetic Struvite - American Chemical Society

Recommend Documents