Sea Level Rise Induced Arsenic Release from ... - ACS Publications

Subscriber access provided by HKU Libraries

Article

Sea level rise induced arsenic release from historically contaminated coastal soils Joshua J. LeMonte, Jason W. Stuckey, Joshua Z. Sanchez, Ryan V Tappero, Jörg Rinklebe, and Donald L. Sparks Environ. Sci. Technol., Just Accepted Manuscript • Publication Date (Web): 04 May 2017 Downloaded from http://pubs.acs.org on May 5, 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 34

Environmental Science & Technology

1

Sea level rise induced arsenic release from historically contaminated coastal soils

2 3

Joshua J. LeMontea,b, Jason W. Stuckeya, Joshua Z. Sancheza, Ryan Tapperoc, Jörg

4

Rinklebed, and Donald L. Sparksa*

5 6

a

7

of Delaware, Newark, Delaware, 19711, USA.

8

b

9

Center, Vicksburg, MS, 39180, USA

Department of Plant and Soil Sciences, Delaware Environmental Institute, University

Currently: U.S. Army Corps of Engineers, Engineer Research & Development

10

c

11

11973, USA

12

d

13

Architecture and Civil Engineering, University of Wuppertal, Wuppertal, 42285,

14

Germany

Photon Sciences Division, Brookhaven National Laboratory, Bldg 743, Upton, NY,

Institute of Foundation Engineering, Water- and Waste-Management, School of

15 16

*Corresponding author - Delaware Environmental Institute, ISE Lab, 221 Academy

17

Street, Suite 250, University of Delaware, Newark, DE 19716; (302) 831-3436;

18

[email protected]

1

ACS Paragon Plus Environment


19 20

Abstract Climate change-induced perturbations in the hydrologic regime are expected to

21

impact biogeochemical processes, including contaminant mobility and cycling.

22

Elevated levels of geogenic and anthropogenic arsenic are found along many coasts

23

around the world, most notably in south and southeast Asia, but also in the United

24

States, particularly along the Mid-Atlantic coast. The mechanism by and the extent to

25

which arsenic may be released in contaminated coastal soils due to sea level rise are

26

unknown. Here we show a series of data from a coastal arsenic-contaminated soil

27

exposed to sea and river waters in biogeochemical microcosm reactors across field-

28

validated redox conditions. We find that reducing conditions lead to arsenic release

29

from historically contaminated coastal soils through reductive dissolution of arsenic-

30

bearing mineral oxides in both sea and river water inundations, with less arsenic

31

release from sea water scenarios than river water due to inhibition of oxide

32

dissolution. For the first time, we systematically display gradation of solid phase soil-

33

arsenic speciation across defined redox windows from reducing to oxidizing

34

conditions in natural waters by combining biogeochemical microcosm experiments

35

and X-ray absorption spectroscopy. Our results demonstrate the threat of sea level rise

36

stands to impact arsenic release from contaminated coastal soils by changing redox

37

conditions.

38 39

Keywords: sea level rise, arsenic, geochemistry, soil chemistry, XANES, redox

2


Page 2 of 34

Page 3 of 34


40 41

Introduction Earth’s climate is changing and as a result, frequency and intensity of storm

42

surges are increasing and sea levels are rising.1 It is estimated that the global mean sea

43

level will rise by 0.8 m over the next century and impact the 25% of the world’s

44

population inhabiting coastal zones.2,3 Increased storm surges and flooding have

45

garnered the bulk of research and media attention associated with sea level rise (SLR)

46

due to the physical and economic threats posed to coastal regions. However, chemical

47

and mineralogical changes in the subsurface that result from saltwater inundation and

48

prolonged flooding, and the concomitant impacts on contaminant fate and transport,

49

remain poorly understood.3–5

50

Climate change induced shifts in the hydrologic regime are expected to impact

51

biogeochemical processes including contaminant mobility and cycling.6 These

52

changes will alter soil redox conditions and introduce salinity and reducing zones in

53

areas that have long remained oxic or less brackish and thus threaten to release redox

54

sensitive toxic elements such as arsenic (As) to groundwater. Elevated levels of As,

55

from both geogenic and anthropogenic sources, are found along the world’s coastlines.

56

In south and southeast Asia, tens of millions of people have been chronically poisoned

57

by As-contaminated groundwater.7 Within the United States, As has been detected in

58

nearly 20% of public groundwater supplies, and elevated levels of As are present at

59

more than a third of the U.S. EPA’s superfund sites on the National Priorities List.8

60

Parts of the world are experiencing faster rates of SLR than the global average.9,10 The

61

Atlantic coastlines are particularly susceptible to near-future SLR, including the Mid-

3



62

Atlantic coast of the U.S. where SLR rates are higher than elsewhere in the world due

63

to the combination of SLR and coastal subsidence.9,11 It has been recommended that

64

states in this region prepare for SLR of at least 1 m by 2100.12 Situated in the center of

65

the Mid-Atlantic, Delaware is expecting at least 1 m of SLR, likely impacting many

66

coastal As-contaminated sites, making it an ideal study location.

67

The fate and transport of As in soils and sediments is intimately tied to

68

environmental conditions (e.g., Eh, pH, ionic competition) as well as the

69

biogeochemical redox cycling of other soil constituents such as Fe, Al, and Mn

70

(oxyhydr)oxides (referred to as “oxides” hereafter), S, and C.6,7,13 Anaerobic

71

conditions lead to reduction of surface bound AsV to the more weakly sorbed AsIII, and

72

thus increase As mobility.14 Reducing conditions also promote reductive dissolution of

73

As-bearing Fe oxides and favor dissimilatory metal reducing bacteria. Dissimilatory

74

reduction of Fe oxides is the suspected primary driver of As release under reducing

75

conditions in soils and sediments. Additionally, reduction of arsenate to arsenite by

76

As-respiring bacteria can be a driver of As release and be uncoupled from

77

dissimilatory Fe reduction.15 In high SO42- systems As release may be inhibited via

78

ligand complexation or energetic favorability of dissimilatory SO42- reduction in lieu

79

of dissimilatory reduction of Fe oxides.16,17

80

Whereas much research has been conducted on As cycling, the stability of As

81

in historically contaminated soils is uncertain.18 Furthermore, the extent to which SLR

82

will impact mobilization of existing contaminants in coastal soils is unknown. To fill

83

this knowledge gap, here we systematically evaluate the effects of inundation with

4


Page 4 of 34

Page 5 of 34


84

seawater and river water on As mobilization and speciation across pre-defined Eh

85

zones using a historically As contaminated soil from the densely populated Delaware

86

coastline. For the first time, measurements of As release to solution were coupled with

87

solid phase speciation under preset redox conditions, while simultaneously monitoring

88

factors controlling As cycling and redox chemistry (e.g. pH, Fe, DOC, S).

89 90

Experimental

91

Soil Sampling and Characterization

92

A series of experiments were conducted using automated biogeochemical

93

microcosm reactors (MC hereafter) to elucidate soil-As dynamics when inundated

94

with sea and river water across pre-defined redox windows. The soil used was

95

collected from a historically As-contaminated industrial site in Wilmington, Delaware,

96

USA, projected to be inundated by 1 m of SLR by 2100. The city of Wilmington, DE

97

has a long history of industrial activities including leather tanning, chemical

98

production, and ore processing.19 Each of these industries has contributed to the

99

widespread As contamination currently found along the banks of the tidally influenced

100

Christina River that flows through Wilmington. The soil was sampled from along the

101

banks of a tidal basin constructed as part of a remediation effort for an adjacent U.S.

102

EPA superfund site.

103

The soil was sampled from 0-20 cm from 10 random locations at the site and

104

combined into one composite sample.20 The composited sample was homogenized,

105

air-dried, ground to pass through a 2 mm sieve, and stored at 4°C in the dark until

5



106

experiments were conducted. Soil pHwater, texture, and organic matter were measured

107

using standard methods.21 Particle size was determined by laser diffraction (Beckman

108

Coulter LS 13 320). Total metal content of the homogenized soil samples was

109

determined by microwave-acid digestion (US EPA 1995 Method 3051), followed by

110

analysis via inductively coupled plasma atomic emission spectrometry (ICP-AES,

111

Thermo Elemental Intrepid II XSP Duo View).22 Total (“free”) iron oxide content was

112

determined using the citrate-bicarbonate-dithionite method.23 Poorly crystalline

113

(“active”) iron oxide content was determined using the acid ammonium oxalate in

114

darkness method.23

115

Microcosm Studies

116

An advanced automated biogeochemical microcosm (MC) system was used to

117

simulate oxidizing and reducing conditions. This system has been used successfully by

118

others; a detailed description can be found therein.24–26

119

At the initiation of each MC trial, an additional 10 g of “fresh” (moist) soil

120

collected from the respective field location was added to each MC as a microbial

121

inoculant. Additionally, 10 g of dried and ground reed (Phragmites australis) collected

122

at the site was used as a slow-release C source for microorganisms. Two 10 g doses of

123

glucose were also added to each MC to prime the system for reaching low redox

124

potentials by ensuring a C-limited system did not exist. Minimum Eh levels were

125

obtained through microbial reduction and flushing the MC with N2 gas.

126

After reaching a stable Eh minimum, the Eh was sequentially raised in 100 +/- 20 mV

127

increments by injecting O2 gas into the reactor then regulating at the target Eh using

6


Page 6 of 34

Page 7 of 34


128

O2 (to increase Eh) or N2 (to decrease Eh). Each targeted Eh window was maintained

129

for 72 h and then a slurry sample was taken. A total of six distinct Eh levels (-300 to

130

+200 mV) were set and sampled for each replicate. The experimental Eh range was

131

field-validated through installation and monitoring of multilevel in situ redox sensors

132

at the soil sampling site (Paleo Terra, Amsterdam, Netherlands).

133

Measurements of Eh, pH, and temperature of the microcosms were logged

134

every 10 minutes for the entirety of the incubation. Mean Eh values during the 6 h

135

prior to sampling was previously determined to give the best correlation coefficients

136

between Eh/pH and metal(loid)s in solution within the MC system and therefore will

137

be presented in this study.26 Slurry samples were sealed at sampling and immediately

138

centrifuged at 1000 g for 10 min. Once centrifuged, samples were moved into an

139

anaerobic glove box and the supernatant was filtered through a 0.45 µm Millipore

140

nylon membrane (Whatman, Inc). While under the 95% N2/5% H2 atmosphere,

141

aqueous subsamples were prepared for total elemental analyses. Subsamples analyzed

142

via ICP mass spectrometry (ICP-MS) for total metal content in solution were acidified

143

to 2% trace metal grade HNO3 and diluted to eliminate potential for C-induced As

144

signal amplification (Agilent Technologies, 7500cx ICP-MS).27 Total organic carbon

145

(TOC) was determined by high temperature catalytic oxidation (Vario TOC cube,

146

Elementar Americas, Mt. Laurel, NJ). Iron(II) was determined by the phenanthroline

147

method (Hach, Loveland, CO), sulfide was determined by the US EPA methylene blue

148

method (Hach), ammonium was determined by the high range salicylate method

7



149

(Hach), and alkalinity was determined by the bromcresol green-methyl red indicator

150

method using the Hach digital titrator, all within 2 h of sampling.

151

Correlations between solution geochemical data and arsenic release were

152

explored using multivariate Partial Least Squares Regression (PLSR).28

153

Bulk X-ray Absorption Spectroscopy

154

Samples from each defined redox window, from -300 mV to +200 mV in 100

155

mV increments, were selected from both the freshwater and seawater trials for As K-

156

edge X-ray absorption spectroscopy (XAS) analysis. Following centrifugation and

157

decanting, samples were dried under a 95%N2/5%H2 atmosphere and the dried soils

158

were mounted onto 1 mm thick aluminum between layers of 25 µm thick Kapton tape.

159

Analyses were performed at beamline 4-3 at the Stanford Synchrotron Radiation

160

Lightsource (SSRL). Fitting was performed with the refined manual linear

161

combination fitting tool available from beamline 10.3.2 at the Advanced Light Source

162

(ALS; https://sites.google.com/a/lbl.gov/als-beamline1032). Fitting was evaluated by

163

calculating the normalized sum of squares (NSS) residual of the fit involving inclusion

164

of the ith component [NSS = Σ(y – yfit)2)/Σ(y2)]. Inclusion of each component was

165

required to 1) improve the NSS by at least 10% and 2) represent at least 5% of the

166

measured signal.

167

Bulk S K-edge XANES analyses were performed on a subset of the same soils

168

analyzed for As XANES. Analyses were conducted at beamlines 14-3 at SSRL and

169

SXS at the Brazilian Synchrotron Light Laboratory. Least squares fitting of

170

normalized XANES fluorescence spectra was performed over a range of 2465–2490

8


Page 8 of 34

Page 9 of 34


171

eV using the LINEST function in Microsoft Excel, with details explained

172

previously.7,29

173

Micro X-ray Fluorescence (µXRF) Mapping and µXANES Spectroscopy

174

Samples from the most extreme redox windows (-300 mV and +200 mV) were

175

selected from both the freshwater and seawater trials. These samples were dried under

176

95%N2/5%H2 atmosphere and resin-embedded (EPOTEK 301_2FL) at low

177

temperature and low O2 (Spectrum Petrographics method X26A). A thin section was

178

then prepared and mounted onto a quartz slide. Samples were transported to the

179

beamline under inert atmosphere, where they were kept until analysis. To determine

180

elemental hot spots and colocation patterns, fine-scale µXRF mapping and µXANES

181

spectroscopy were performed at SSRL beamline 2-3 following a previously described

182

protocol. The beamline was equipped with KB focusing mirrors providing a spot size

183

of 2x2 µm. Elemental maps were generated using a single-element Si drift Vortex

184

detector; the sample was continuously rastered across the X-ray beam using a 5 µm

185

pixel size and dwell time of 50 ms per pixel. Windowed counts of each element were

186

isolated from the full X-ray fluorescence spectra and normalized to the intensity of the

187

incident X-ray beam (I0). Energy was selected using a Si (111) double crystal

188

monochromator and calibrated by assigning the whiteline position of Na3AsO4

189

standard to 11.874 keV.

190

Based on µXANES spectra of six As standards, representing the potential

191

chemical species within the reacted sediments, a 1.5 mm x 1.5 mm (2.25 mm2) region

192

of the thin section was mapped at 11.867, 11.870, 11.873, 11.877, and 11.881 keV

9



193

incident energy. Using the SMAK routine, principal component analysis (PCA) was

194

performed on dead time corrected maps to determine optimal locations for µXANES

195

spectral analysis.30 Least squares fitting of normalized µXANES spectra was

196

optimized at ±20 eV from E0 in Athena.31 Normalized µXANES spectra of the As

197

standards corresponding to the five incident energies were used to fit the multiple

198

energy µXRF maps in a non-negative linear least squares routine in SMAK.32

199

More experimental details can be found in the Supporting Information.

200

Results and Discussion

201

Arsenic Dynamics in Solution Across Eh Zones

202

The reacted soil was a loam with mean particle size of 737.1 µm (s.d. 641.6

203

µm) and pH = 5.9. Total As concentration in the soil was 1.3 g kg-1 and the total Fe

204

concentration was 9.7 wt %, 75% of which were amorphous hydrous Fe oxides (Table

205

1). Arsenic speciation in the soil collected and preserved for XANES analysis was

206

primarily AsIII (data not shown).

207

Inundation with both waters resulted in decreased Eh and increased pH, as has

208

been seen previously (Fig. S1).25,33,34 The river water had an initial pH of 7.2, and an

209

EC of 0.7 mmhos cm-1. The seawater had an initial pH of 7.8 and an EC of 45.6

210

mmhos cm-1. Arsenic was below detection (< 0.5 µg L-1) in both unreacted waters

211

(Table 1). Arsenic concentrations in solution increased as Eh decreased in both the

212

river water and seawater trials (Fig. 1). River water inundation resulted in

213

approximately twice as much As release than seawater (Fig. 1). Sequential extractions

214

revealed that 86% of total As is occluded in hydrous oxides, with the balance of As

10


Page 10 of 34

Page 11 of 34


215

sorbed (13% inner-sphere and 1% outer-sphere). These results suggest that As release

216

cannot be explained by sorption processes. Rather, dissolution (ligand, proton, and/or

217

reductive) of As-bearing mineral oxides may be driving As release.

218

Total SO42- in the seawater system was ten times that of the freshwater system,

219

with levels reaching nearly 70 mg L-1 at Eh levels greater than 100 mV. It is expected

220

a large amount of seawater SO42- sorbed to the soil, potentially stabilizing As-bearing

221

hydrous oxides by forming bidentate binuclear ligand complexes.17 Elevated SO42- in

222

solution can competitively inhibit dissimilatory reduction of AsV and FeIII via

223

dissimilatory reduction of SO42- at pH > 5.0, represented as:16,35 + → 2 +

224

Preferential or concurrent SO42- reduction may preserve occluded AsV and

225

result in decreased As release, which is a possible explanation for the decreased As in

226

solution in the seawater inundation scenario as compared to the river water. This is a

227

thermodynamic possibility if the Gibbs free energy (∆G0) of dissimilatory sulfate or

228

Fe reduction are equally favorable, as can be the case depending on which Fe oxide is

229

being reduced.36,37 It has been shown that following intrusion by water with high

230

salinity, mineralization of organic C can shift rapidly from Fe reduction to SO42-

231

reduction.38

232

Whereas As cycling can be strongly impacted by S cycling in reduced soils and

233

sediments, As was not correlated to sulfide (S2-) or sulfate in solution for either

234

inundation scenario (Fig. 1). Additionally, there was no pattern of S2- or SO42- in

235

solution across Eh and pH values (Fig. S3). Sulfide was below the method detection

11



236

limit (1 mg L-1) for both waters for the duration of the experiment. It is likely that S2-

237

was formed under the reducing conditions through the reduction of SO42-, but likely

238

rapidly transitioned to Fe-S, As-S, or Fe-As-S precipitates in the high AsIII and FeII

239

solution, resulting in the low S2- in solution throughout the duration of the

240

experiment.16 High rates of SO2- reduction, particularly in the seawater treatments may

241

have resulted in formation of aqueous As-S complexes (e.g. HAs3S62-).39 Thioarsenic

242

species can be found in environmental samples that have high levels of solution As

243

and S, but likely were not significant in this system because of the high levels of Fe

244

that would sequester As and S by coprecipitation.40

245

The difference in As release between treatments could also be partly attributed

246

to the higher imposed osmotic stress of increased salinity on the microbial community

247

by seawater. Alkalinity and As release were significantly correlated for both river

248

water and seawater, though the correlation was stronger with the river water (r2 = 0.82

249

and 0.59, respectively). Alkalinity generation was greater in the river water (5000 mg

250

L-1 at -225 mV) than the seawater (3500 mg L-1 at -187 mV), indicating decreased

251

microbial activity in the seawater, suggesting microbial inhibition.41 Salinity in soil is

252

a major source of stress on soil microbial communities, especially in those regions of

253

the world undergoing salinization due to climate change.42 Our results suggest that

254

introducing salinity to coastal soils through SLR and storm surges may similarly stress

255

the microbial communities in the inundated soils and have secondary effects on soil

256

chemistry.43 In the case of As mobilization, rapid introduction of increased salinity

257

proved beneficial in these soils by inhibiting dissimilatory metal reducing bacteria

12


Page 12 of 34

Page 13 of 34


258

activity. However, a slow shift from fresh water to more saline conditions, as in the

259

case of SLR, may allow the microbial community to adjust by accumulating

260

osmolytes.42

261

To ensure the experiment was not C limited, glucose and P. australis were

262

added and thus resulted in high concentrations of solution DOC. Concentrations of

263

DOC were similar for both inundation scenarios, ranging from approximately 1600

264

mg L-1 at the onset of the experiment (most reducing conditions) to 800 mg L-1 at the

265

conclusion of the experiment (most oxidizing conditions). The decrease in DOC in

266

solution with increasing Eh levels may be attributed to increased microbial C

267

consumption under more oxidizing conditions, or the microbial utilization of DOC

268

over the course of the experiment.34,44 The highest concentrations of As were present

269

under the same conditions that the highest concentrations of DOC were observed,

270

suggesting more available C can stimulate microbially driven reductive release of As

271

from As-bearing oxides. In addition to driving microbial reductive release, DOC can

272

also act as a carrier for As, thereby increasing As mobility.45 The DOC-As correlation

273

was stronger when inundated with river water (r2 = 0.69) than with seawater (r2 =

274

0.38).

275

Arsenic release increased notably at Eh values below 100 mV, corresponding

276

to the value at which reductive dissolution of Fe oxides begins.46 Reductive Fe-oxide

277

dissolution also began for both inundation scenarios near the critical redox level of

278

100 mV (Fig. S2). Similar to dissolved As, FeII in solution increased under reducing

279

conditions, with the lowest levels of FeII present at the highest Eh levels. Both the

13



280

river water and seawater trials reached maximum concentrations of FeII in solution of

281

approximately 1000 mg L-1 under the most reducing conditions (below -200 mV), at

282

which time As in solution was also at its peak. Though both inundation scenarios

283

demonstrated significant correlations between As and Fe(II), a stronger relationship

284

was observed in the freshwater (r2 = 0.82 for river water, r2 = 0.38 for seawater; Fig.

285

1). Total solution Fe displayed similar behavior (Fig. S2).

286

Arsenic release was correlated with a number of other measured factors,

287

including pH, total Fe, alkalinity, and DOC (Fig. 1). A two-factor partial least squares

288

model comprised of DOC, Eh, Fe(II), pH, total Fe, and Mn explains eighty percent of

289

the data variance (SI). Despite the seawater having a higher initial pH prior to

290

inundation, when reacted with the soil, the MC pH decreased below the river water

291

MC pH, and remained lower throughout the experiment (SI Fig. 1). The seawater

292

maximum pH levels were approximately 1 pH unit below the freshwater trials, likely

293

due to the “sea salt effect” (where cation exchange processes cause acidification

294

following sea salt-rich inundation) and the hydrolysis of exchangeable metal cations

295

(e.g., Al) in the soil.47–49 This slight decrease in pH may have contributed to less As in

296

solution in the seawater via increased AsV partitioning to the solid phase at lower

297

pH.50

298

Solid Phase As Speciation Dynamics Across Eh Zones

299

Although As speciation in the unreacted soil, collected and preserved for

300

XANES analysis, was primarily AsIII (data not shown), the soils were air dried and

301

ground prior to experimentation under oxidization-promoting conditions which led to

14


Page 14 of 34

Page 15 of 34


302

As oxidation prior to experimentation. Arsenic speciation in the reacted soils as

303

determined by bulk XANES showed As reduction at the lowest reductive potential for

304

both inundation scenarios, which corroborates the solution data. Both scenarios

305

produced partial reduction of AsV to AsIII, with AsV present under even the most

306

anoxic conditions (Fig. 2). In fact, mixed oxidation states were present at all Eh zones.

307

Additionally, authigenic As-S complexes formed in both water treatments (Table 2).

308

Under the most reducing conditions, As reduction was more pronounced in river water

309

as compared to the seawater, with the linear combination fitting (LCF) results showing

310

up to 79% AsIII contribution due to the river water treatment and a maximum of 66%

311

AsIII ascribed to the seawater treatment (Table 2). As Eh incrementally moved from

312

reducing to oxidizing, As was incrementally oxidized and the proportion of AsIII

313

decreased by a total of 18% in the river water systems, and 9% in the seawater

314

systems. At the most oxic Eh zone, As solid phase speciation was similar (60% AsIII,

315

40% AsV) for both waters (+200±20 mV) (Table 2).

316

Seawater trials led to more solid phase AsV across the entire Eh range (Fig. 2,

317

Table 2). As mentioned earlier, with As in solution, high levels of SO42- can inhibit

318

dissolution and/or dissimilatory reduction of hydrous Fe oxides.16,17 Further

319

supporting this, solid phase As speciation shows less reduction of As in the high SO42-

320

seawater than the river water.

321

Arsenic speciation as determined by µXANES showed the same pattern found

322

in the bulk XANES analyses, but provides additional insights into the heterogeneity

323

that leads to the mixed oxidation state results shown in the bulk XANES data. Multi-

15



324

energy µXRF mapping of the most reduced and oxic samples from each inundation

325

scenario suggests very little, if any, As oxidation states that are not some kind of

326

As(III)/As(V) mixture. It is possible that this is a limitation of the method, and perhaps

327

the use of a nanoprobe beamline would enable detection of distinct oxidation states.

328

Tri-color RGB µXRF maps used to determine elemental colocation of As, Fe, and S

329

indicate a stronger correlation between As and Fe in the solid phase with seawater

330

than river water, suggesting increased As-Fe sorption compounds (Figure 3). These

331

maps also show signs of deflocculation in the seawater with visibly smaller particle

332

sizes present in the maps, possibly due to the sodicity of the seawater.

333

The mean of all 10 points measured by µXANES at each Eh zone was similar

334

to that measured by bulk XANES at the same Eh zones, which indicates that the

335

microprobe data adequately represent the heterogeneity present in the bulk XANES

336

(SI Table 1). There were three to five components of mixed oxidation states at each Eh

337

zone as determined by principal component analysis (PCA) (Fig. 3). Similar to the

338

bulk XANES, the more oxic Eh zones were mostly AsV. However, some spots (river

339

water 184 mV spot A and seawater 183 mV spot A) were comprised of approximately

340

70% AsIII under these oxidizing conditions. This further demonstrates sample

341

heterogeneity and possible influence of microsites on overall As release.

342

A parallel, but inverse, phenomenon was found in the most reducing zones.

343

The solid phase in the river water treatment showed 4 components of mixed oxidation

344

state from the PCA, ranging from 69-85% AsIII, coupled with some As-S precipitation

345

at -324 mV. The largest contribution from AsV at this Eh zone was 32% (river water 16


Page 16 of 34

Page 17 of 34


346

324 mV, spot D). However, for the seawater microcosms at the lowest Eh zone (-254

347

mV), three mixed oxidation state PCA components were identified and none of the

348

µXANES indicated less than 28% contribution by AsV (seawater -254 mV spot A).

349

The maximum contribution of AsV for the reducing seawater microcosms reached up

350

to 59% (seawater -254 mV spot C).

351

These results further demonstrate decreased AsV reduction in the seawater

352

inundations as compared to the river water inundations. Arsenic release under

353

reducing conditions in soil is primarily a microbially mediated process through

354

dissimilatory reductive dissolution of As-bearing Fe oxides such as ferrihydrite,

355

goethite, lepidocrocite, magnetite, or hematite, but can also occur through direct

356

enzymatic reduction of AsV. The energy requisite for reducing these FeIII minerals,

357

their reactivity, surface area, and As sorption capacity are unique for each Fe oxide.51

358

Under certain environmental conditions, concurrent or preferential SO42- reduction is

359

possible when the Gibbs free energy of sulfate reduction is greater than that of more

360

crystalline phases of FeIII. Because reductive dissolution of Fe oxides is a primary

361

driver of As release from soils, it is therefore reasonable to assume AsV reduction and

362

release is limited in the seawater inundations as compared to the river water by

363

reduced microbial activity.

364

Solid Phase S Speciation Dynamics Across Eh Zones

365

Sulfur speciation, as determined by bulk XANES spectroscopy showed

366

considerable variation between the inundation scenarios (Fig. 4). Solid phase sulfur

367

speciation in the river water inundation was primarily comprised of sulfides and

17



368

pyrite, with their respective contributions being inversely related with changes in Eh –

369

organic sulfides increased with increasing Eh whereas pyrite decreased (Fig. 4, Table

370

S1). The pyrite contribution at low Eh suggests the presence of Fe-As-S coprecipitates

371

under reducing conditions and is in agreement with the solid phase As speciation.

372

Sulfate was only present at 184 mV, comprising 12% of the S speciation.

373

The S speciation in soils inundated with seawater was also predominately

374

organic sulfides and pyrite, but the behavior thereof with changing Eh differed from

375

the river water inundations – pyrite and organic sulfide contributions decreased

376

slightly with increasing Eh (Fig. 4, Table S1). Sulfate contribution to S speciation was

377

greater in the seawater than the river water, and increased with increasing Eh,

378

contributing 16% under low Eh and up to 26% at 184 mV. The lower solid-phase

379

sulfate levels under reducing conditions are likely due to dissimilatory sulfate

380

reduction. Although the amount of sulfate reduction is similar in both river and

381

seawater inundations by percent, the amount of total S in the seawater systems is 100

382

times that of the river water. This suggests that the amount of sulfate reduction

383

occurring in the seawater inundation is much greater by mass than in the river water.

384

This corroborates the hypothesis of concurrent or preferential dissimilatory sulfate

385

reduction in the high sulfate seawater systems, potentially resulting in less As release.

386

Implications for Sea Level Rise

387

Climate change induced seawater inundation, short and/or long term flooding,

388

and salinification will alter the biogeochemistry of these sensitive sites. Newly flooded

389

areas will see formerly oxic zones turn more reducing and thereby change the ability

18


Page 18 of 34

Page 19 of 34


390

of the soils and sediments to sequester contaminants. Our findings show that

391

introducing reducing conditions can lead to an increased As release from soils through

392

dissolution of As-bearing mineral oxides in both river water and seawater inundations

393

scenarios. Additionally, short-term inundation with seawater can lead to less As

394

release than inundation with freshwater due to inhibition of oxide dissolution.

395

Accordingly, the threat of SLR stands to impact release of As from contaminated

396

coastal soils primarily by changing the redox conditions and subsurface mineralogy.

397

Future work should include abiotic controls to further understand the mechanisms of

398

As release and the role of microbial communities under different degrees of

399

salinization on the biogeochemical cycling of As. The compounding issues of SLR

400

and its impacts on the cycling of contaminants in coastal soils is of merit for the

401

communities at coastal areas around the world and calls for an appropriate field

402

monitoring and modern lab-based process-orientated research to better understand the

403

underlying biochemical processes and the associated risks for humans and the

404

environment.

405 406

Acknowledgements

407

This work was made possible by a US Department of Defense SMART Fellowship

408

Program awarded to J.J.L., the National Science Foundation EPSCoR Grant No. IIA-

409

1301765, and the state of Delaware. We are appreciative to the Delaware

410

Environmental Institute (DENIN), Caroline Golt for her ICP expertise, DENIN

411

summer scholars Prian Esquivel and Benjamin Wendt for help in collecting and

19



412

preparing samples, and John Cargill and others of the Delaware Department of Natural

413

Resources and Environmental Control for assistance in gaining permissions for field

414

sampling. Thank you to the staff at the Stanford Synchrotron Radiation Lightsource

415

for support. Use of the Stanford Synchrotron Radiation Lightsource, SLAC National

416

Accelerator Laboratory is supported by the US Department of Energy, Office of

417

Science, Office of Basic Energy Sciences under Contract No. DE-AC02-76SF00515.

418 419

Supporting Information

420

Details of microcosm experimental methods, bulk X-ray absorption spectroscopy, and

421

figures detailing redox potential, iron, DOC and sulfate in solution, and bulk sulfur

422

XANES results.

423 424

References

425 426

(1)

Church, J. A.; White, N. J.; Aarup, T.; Wilson, W. S.; Woodworth, P. L.;

427

Domingues, C. M.; Hunter, J. R.; Lambeck, K. Understanding global sea levels:

428

Past, present and future. Sustain. Sci. 2008, 3, 9–22.

429

(2)

Church, J. A.; Clark, P. U.; Cazenave, A.; Gregory, J. M.; Jevrejeva, S.;

430

Levermann, A.; Merrifield, M. A.; Milne, G. A.; Nerem, R. S.; Nunn, P. D.; et

431

al. Sea level change. Clim. Chang. 2013 Phys. Sci. Basis. Contrib. Work. Gr. I

432

to Fifth Assess. Rep. Intergov. Panel Clim. Chang. 2013, 1137–1216.

433

(3)

Small, C.; Nicholls, R. J. A global analysis of Human Settlement in Coastal

20


Page 20 of 34

Page 21 of 34


434 435

Zones. J. Coast. Res. 2003, 19 (3), 584–599. (4)

Nicholls, R. J.; Tol, R. S. J. Impacts and responses to sea-level rise: a global

436

analysis of the SRES scenarios over the twenty-first century. Philos. Trans. A.

437

Math. Phys. Eng. Sci. 2006, 364 (1841), 1073–1095.

438

(5)

Yu, X.; Yang, J.; Graf, T.; Koneshloo, M.; Neal, M. A. O.; Michael, H. A.

439

Assessing the Impact of Topography on Groundwater Salinization Due to

440

Storm Surge Inundation References : Water Resour. Res. 2016, 52 (8), 5794–

441

5812.

442

(6)

Borch, T.; Kretzschmar, R.; Skappler, A.; Van Cappellen, P.; Ginder-Vogel,

443

M.; Voegelin, A.; Campbell, K. Biogeochemical redox processes and their

444

impact on contaminant dynamics. Environ. Sci. Technol. 2010, 44, 15–23.

445

(7)

Stuckey, J. W.; Schaefer, M. V.; Kocar, B. D.; Dittmar, J.; Pacheco, J. L.;

446

Benner, S. G.; Fendorf, S. Peat formation concentrates arsenic within sediment

447

deposits of the Mekong Delta. Geochim. Cosmochim. Acta 2015, 149, 190–205.

448

(8)

DeLemos, J. L.; Bostick, B. C.; Renshaw, C. E.; Stürup, S.; Feng, X. Landfill-

449

stimulated iron reduction and arsenic release at the Coakley Superfund Site

450

(NH). Environ. Sci. Technol. 2006, 40, 67–73.

451

(9)

Krasting, J. P.; Dunne, J. P.; Stouffer, R. J.; Hallberg, R. W. Enhanced Atlantic

452

sea-level rise relative to the Pacific under high carbon emission rates. Nat.

453

Geosci. 2016, 9 (3), 210–215.

454 455

(10)

Rietbroek, R.; Brunnabend, S.-E.; Kusche, J.; Schröter, J.; Dahle, C. Revisiting the contemporary sea-level budget on global and regional scales. Proc. Natl.

21



456 457

Acad. Sci. 2016, 113 (6), 1504–1509. (11)

458 459

Sallenger, A. H.; Doran, K. S.; Howd, P. a. Hotspot of accelerated sea-level rise on the Atlantic coast of North America. Nat. Clim. Chang. 2012, 2 (8), 1–5.

(12)

Titus, J. G. (Coordinating lead author); Anderson, K. E.; Cahoon, D. R.; Gesch,

460

D. B.; Gill, S. K.; Gutierrez, B. T.; Thieler, E. R.; Williams, S. J. (lead authors).

461

Coastal Cycle Sensitivity to Sea-Level Rise : Focus on Mid-Atlantic Region;

462

Washington, D.C., USA, 2009.

463

(13)

Kocar, B. D.; Herbel, M. J.; Tufano, K. J.; Fendorf, S. Contrasting effects of

464

dissimilatory iron (III) and arsenic (V) reduction on arsenic retention and

465

transport. Environ. Sci. Technol. 2006, 40 (21), 6715–6721.

466

(14)

Charlet, L.; Morin, G.; Rose, J.; Wang, Y.; Auffan, M.; Burnol, A.; Fernandez-

467

Martinez, A. Reactivity at (nano)particle-water interfaces, redox processes, and

468

arsenic transport in the environment. Comptes Rendus Geosci. 2011, 343, 123–

469

139.

470

(15)

Tufano, K. J.; Reyes, C.; Saltikov, C. W. Reductive Processes Controlling

471

Arsenic Retention : Revealing the Relative Importance of Iron and Arsenic

472

Reduction. Environ. Sci. Technol. 2008, 42 (22), 8283–8289.

473

(16)

Burton, E. D.; Johnston, S. G.; Kraal, P.; Bush, R. T.; Claff, S. Sulfate

474

availability drives divergent evolution of arsenic speciation during microbially

475

mediated reductive transformation of schwertmannite. Environ. Sci. Technol.

476

2013, 47 (5), 2221–2229.

477

(17)

Sparks, D. L. Soil Physical Chemistry, Second Edi.; Sparks, D. L., Ed.; CRC

22


Page 22 of 34

Page 23 of 34


478 479

Press: New York, 1999. (18)

Rajpert, L.; Kolvenbach, B. A.; Ammann, E. M.; Hockmann, K.; Nachtegaal,

480

M.; Eiche, E.; Scha, A.; Francois, P.; Corvini, X.; Sk, A.; et al. Arsenic

481

Mobilization from Historically Contaminated Mining Soils in a Continuously

482

Operated Bioreactor: Implications for Risk Assessment. Environ. Sci. Technol.

483

2016, 50, 9124–9132.

484

(19)

Landrot, G.; Tappero, R.; Webb, S. M.; Sparks, D. L. Arsenic and chromium

485

speciation in an urban contaminated soil. Chemosphere 2012, 88 (10), 1196–

486

1201.

487

(20)

488 489

Francis Group: New York, 2005. (21)

490 491

Sparks, D. L. Methods of soil analysis. Part 3 - Chemical Methods. SSSA Book Series No. 5; Sparks, D. L., Ed.; SSSA and ASA: Madison, WI, 1996.

(22)

492 493

Tan, K. H. Soil Sampling, Preparation, and Analysis, Second.; Taylor &

Agency, U. S. E. P. Method 3051a: Microwave Assisted Acid Digestion of Sediments, Sludges, Soils, and Oils; Washington, D.C., USA, 1998.

(23)

Loeppart, R. H.; Inskeep, W. P. Iron. In Methods of Soil Analysis. Part 3.

494

Chemical Methods; Sparks, D. L., Ed.; Soil Science Society of America:

495

Madison, WI, 1996; pp 384–411.

496

(24)

497 498 499

Yu, K.; Rinklebe, J. Advancement in soil microcosm apparatus for biogeochemical research. Ecol. Eng. 2011, 37 (12), 2071–2075.

(25)

Frohne, T.; Rinklebe, J.; Diaz-Bone, R. A.; Du Laing, G. Controlled variation of redox conditions in a floodplain soil: Impact on metal mobilization and

23



500 501

biomethylation of arsenic and antimony. Geoderma 2011, 160 (3–4), 414–424. (26)

Frohne, T.; Diaz-Bone, R. a.; Du Laing, G.; Rinklebe, J. Impact of systematic

502

change of redox potential on the leaching of Ba, Cr, Sr, and V from a riverine

503

soil into water. J. Soils Sediments 2015, 15 (3), 623–633.

504

(27)

Larsen, E. H.; Sturup, S. Carbon-enhanced inductively coupled plasma mass

505

spectrometric detection of arsenic and selenium and its application to arsenic

506

speciation. J. Agric. Food Chem. 1994, 9 (October), 1099–1105.

507

(28)

Rigol, A.; Camps, M.; De Juan, A.; Rauret, G.; Vidal, M. Multivariate soft-

508

modeling to predict radiocesium soil-to-plant transfer. Environ. Sci. Technol.

509

2008, 42 (11), 4029–4036.

510

(29)

Almkvist, G.; Boye, K.; Ingmar, P. K-edge XANES analysis of sulfur

511

compounds: An investigation of the relative intensities using internal

512

calibration. J. Synchrotron Radiat. 2010, 17 (5), 683–688.

513

(30)

Webb, S. M. The microanalysis toolkit: X-ray fluorescence image processing

514

software. In 10th International Conference on X-ray Microscopy; McNulty, I.,

515

Eyberger, C., Lai, B., Eds.; 2011; pp 196–199.

516

(31)

Ravel, B. A THENA U ser ’ s G uide. 2009.

517

(32)

Mayhew, L. E.; Webb, S. M.; Templeton, a. S. Microscale imaging and

518

identification of fe speciation and distribution during fluid-mineral reactions

519

under highly reducing conditions. Environ. Sci. Technol. 2011, 45 (10), 4468–

520

4474.

521

(33)

Frohne, T.; Rinklebe, J.; Diaz-Bone, R. a. Contamination of Floodplain Soils

24


Page 24 of 34

Page 25 of 34


522

along the Wupper River, Germany, with As, Co, Cu, Ni, Sb, and Zn and the

523

Impact of Pre-definite Redox Variations on the Mobility of These Elements.

524

Soil Sediment Contam. An Int. J. 2014, 23 (7), 779–799.

525

(34)

Shaheen, S. M.; Rinklebe, J.; Frohne, T.; White, J. R.; DeLaune, R. D. Redox

526

effects on release kinetics of arsenic, cadmium, cobalt, and vanadium in Wax

527

Lake Deltaic freshwater marsh soils. Chemosphere 2015, 150, 740–478.

528

(35)

529 530

lakes and their watersheds. Sci. Total Environ. 2006, 369 (1–3), 307–332. (36)

531 532

Blodau, C. A review of acidity generation and consumption in acidic coal mine

Bethke, C. M.; Sanford, R. A.; Kirk, M. F.; Jin, Q.; Flynn, T. M. The thermodynamic ladder in geomicrobiology. Am. J. Sci. 2011, 311 (3), 183–210.

(37)

Kocar, B. D.; Fendorf, S. Thermodynamic constraints on reductive reactions

533

influencing the biogeochemistry of arsenic in soils and sediments. Environ. Sci.

534

Technol. 2009, 43 (13), 4871–4877.

535

(38)

Weston, N. B.; Dixon, R. E.; Joye, S. B. Ramifications of increased salinity in

536

tidal freshwater sediments: Geochemistry and microbial pathways of organic

537

matter mineralization. J. Geophys. Res. 2006, 111 (G1), G01009.

538

(39)

Newman, D. K.; Beveridge, T. J.; Morel, F. M. M. Precipitation of Arsenic

539

Trisulfide By Desulfotomaculum Auripigmentum. Appl. Environ. Microbiol.

540

1997, 63 (5), 2022–2028.

541

(40)

Stucker, V. K.; Silverman, D. R.; Williams, K. H.; Sharp, J. O.; Ranville, J. F.

542

Thioarsenic species associated with increased arsenic release during

543

biostimulated subsurface sulfate reduction. Environ. Sci. Technol. 2014, 48

25



544 545

(22), 13367–13375. (41)

Postma, D.; Larsen, F.; Minh Hue, N. T.; Duc, M. T.; Viet, P. H.; Nhan, P. Q.;

546

Jessen, S. Arsenic in groundwater of the Red River floodplain, Vietnam:

547

Controlling geochemical processes and reactive transport modeling. Geochim.

548

Cosmochim. Acta 2007, 71 (21), 5054–5071.

549

(42)

Rath, K. M.; Rousk, J. Salt effects on the soil microbial decomposer community

550

and their role in organic carbon cycling: A review. Soil Biol. Biochem. 2015,

551

81, 108–123.

552

(43)

553 554

Egamberdieva, D.; Renella, G.; Wirth, S.; Islam, R. Secondary salinity effects on soil microbial biomass. Biol. Fertil. Soils 2010, 46 (5), 445–449.

(44)

Yu, K.; Böhme, F.; Rinklebe, J.; Neue, H.-U.; DeLaune, R. D. Major

555

Biogeochemical Processes in Soils-A Microcosm Incubation from Reducing to

556

Oxidizing Conditions. Soil Sci. Soc. Am. J. 2007, 71 (4), 1406.

557

(45)

Mladenov, N.; Zheng, Y.; Simone, B.; Bilinski, T. M.; McKnight, D. M.;

558

Nemergut, D.; Radloff, K. A.; Rahman, M. M.; Ahmed, K. M. Dissolved

559

Organic Matter Quality in a Shallow Aquifer of Bangladesh: Implications for

560

Arsenic Mobility. Environ. Sci. Technol. 2015, 49 (18), 10815–10824.

561

(46)

562 563

Reddy, R.; Delaune, R. D. Biogeochemistry of Wetlands: Science and Applications; CRC Press: Boca Raton, FL, 2008.

(47)

Wright, R. F.; Norton, S. A.; Brakke, D. F.; Frogner, T. Experimental

564

verification of episodic acidification of freshwaters by sea salts. Nature 1988,

565

334 (4), 422–424.

26


Page 26 of 34

Page 27 of 34


566

(48)

Wong, V. N. L.; Johnston, S. G.; Burton, E. D.; Bush, R. T.; Sullivan, L. A.;

567

Slavich, P. G. Seawater causes rapid trace metal mobilisation in coastal lowland

568

acid sulfate soils: Implications of sea level rise for water quality. Geoderma

569

2010, 160 (2), 252–263.

570

(49)

Dixit, S.; Hering, J. G. Comparison of arsenic(V) and arsenic(III) sorption onto

571

iron oxide minerals: Implications for arsenic mobility. Environ. Sci. Technol.

572

2003, 37 (18), 4182–4189.

573

(50)

574 575

Campbell, K. M.; Nordstrom, D. K. Arsenic Speciation and Sorption in Natural Environments. Rev. Mineral. Geochemistry 2014, 79 (1), 185–216.

(51)

Pedersen, H.; Postma, D.; Jakobsen, R. Release of arsenic associated with the

576

reduction and transformation of iron oxides. Geochim. Cosmochim. Acta 2006,

577

70 (16), 4116–4129.

578

27



579

TABLES

580

Table 1:

Soil Collection Site

Soil and water characterization

Depth

Texture

(cm) Wilmington, DE, USA

0-20

Loam

EC Water Type

Mean Particle Size (µm)

pH

737

5.9

DOC

Amorphous Hydrous Fe Oxides

Total Fe

Al

As

------------------------- (mg g-1) -------------------------

Ca

97.01

Mg

72.25

Na

As

pH

16.27

Total Fe

13.30

Mn

S

-------------------------- (mg L-1) --------------------------

(mmhos/cm)

581

Page 28 of 34

River

7.2

0.7

4.4

24

9.4

42.4

bd*

0.04

bd

7.8

Sea

7.8

45.6

5.5

369

1091

7153

bd

0.08

bd

789

*bd indicates values that were below the detection limit (0.5 µg L-1 for As and 3 µg L-1 for Mn)

28


Cr

Pb

Mn

S

------------ (mg kg-1) -----------62

360

558

4157

Page 29 of 34

582


Table 2:

Arsenic bulk XANES linear combination fit results..

583 Sea Water Eh6h (mV) 183 169 125 120 10 -25 -94 -127 -187 -188 -240 -254

As As5+ ------------------------(%)-----------------------0 57 43 0 57 43 7 56 37 6 56 38 7 59 34 7 59 34 9 63 28 9 61 30 9 62 29 9 61 30 8 60 32 8 66 26

del-E (eV) 0.4 0.3 -0.13 0.28 -0.11 0.31 -0.25 0.36 -0.43 0.25 0.2 -0.58

River Water As-S As As5+ ------------------------(%)-----------------------0 60 40 0 63 37 8 60 32 8 64 28 10 67 23 11 68 21 12 71 17 11 69 20 11 73 17 11 75 14 10 76 13 10 79 11

del-E (eV) 0.22 -0.81 0.38 -0.56 0.46 -0.62 0.22 -0.69 -0.5 -0.04 -0.2 0.13

As-S*

3+

NSS** 1.06E-03 9.95E-04 1.03E-03 8.80E-04 9.86E-04 8.29E-04 9.24E-04 7.83E-04 9.50E-04 8.01E-04 8.35E-04 8.89E-04

584 585 Eh6h (mV) 185 184 108 81 13 -19 -117 -118 -214 -225 -314 -318 586 587 588 589 590 591

3+

NSS 1.10E-03 1.13E-03 8.49E-04 9.71E-04 8.26E-04 8.86E-04 7.45E-04 8.29E-04 8.02E-04 7.31E-04 8.02E-04 7.12E-04

*The standard used for linear combination fitting that is represented by As-S was realgar (AsS or α-As4S4). **NSS = normalized sum of squares residual of the fit, determined by NSS = Σ(y – yfit)2)/Σ(y2). The standards used for As3+ and As5+ were sodium arsenite and sodium arsenate, respectively. Inclusion of each component was required to 1) improve the NSS by at least 10% and 2) represent at least 5% of the measured signal.

29 ACS Paragon Plus Environment


295x238mm (300 x 300 DPI)


Page 30 of 34

Page 31 of 34


Figure 1. Arsenic in solution as regressed to linked factors. For all bivariate linear regression analyses, n = 20, p value given by analysis of variance 261x187mm (300 x 300 DPI)



Figure 2. A. Arsenic bulk XANES for sea water and river water inundations scenarios across designated Eh windows following 72 h equilibration time at that Eh. The solid lines represent spectra for XANES data and fits obtained by linear combination fitting of the data are dotted lines. B. Weight percent As speciation contribution as determined by linear combination fitting. 262x351mm (300 x 300 DPI)


Page 32 of 34

Page 33 of 34


Figure 3. A. Solid phase As speciation changes via XRF and XANES spectroscopy. River water spectra are expressed in orange and sea water are in purple, with the darker shades indicating lower Eh values. Data are represented by the solid lines and fits obtained by linear combination fitting are dotted lines. B. In the tri-color RBG maps red represents As, blue represents Fe, and green represents S. 195x154mm (300 x 300 DPI)



Figure 4. A. Solid phase S speciation changes via bulk XANES. River water spectra are expressed in orange and sea water are purple, with the darker shades indicating lower Eh values. Data are represented by the solid lines and fits obtained by linear combination fitting are dotted lines. Linear combination fitting was performed using sulfate ester, pyrite, diphenylsulfoxide, and organic sulfides (methionine, cystine, and diphenyldisulfide). B. Weight percent As speciation contribution as determined by linear combination fitting. 193x127mm (300 x 300 DPI)


Page 34 of 34

Sea Level Rise Induced Arsenic Release from ... - ACS Publications

Recommend Documents