Monothioarsenate occurrence in Bangladesh groundwater and its

4 hours ago - The Altmetric Attention Score is a quantitative measure of the attention that a research article has received online. Clicking on the do...
2 downloads 8 Views 524KB Size
Subscriber access provided by UNIV OF DURHAM

Remediation and Control Technologies

Monothioarsenate occurrence in Bangladesh groundwater and its removal by ferrous and zero-valent iron technologies Britta Planer-Friedrich, Jörg Schaller, Fabian Wismeth, Judith Mehlhorn, and Stephan J. Hug Environ. Sci. Technol., Just Accepted Manuscript • Publication Date (Web): 19 Apr 2018 Downloaded from http://pubs.acs.org on April 19, 2018

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 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 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.

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 24

Environmental Science & Technology

1

Journal: Environmental Science and Technology

2 3

Monothioarsenate occurrence in Bangladesh groundwater and its removal by ferrous

4

and zero-valent iron technologies

5 6

Britta Planer-Friedricha*, Jörg Schallera, Fabian Wismetha, Judith Mehlhorna, and Stephan J.

7

Hugb

8 9 10

a

11

(BayCEER), Bayreuth University, Universitaetsstrasse 30, 95440 Bayreuth, Germany

Environmental Geochemistry, Bayreuth Center for Ecology and Environmental Research

12 13

b

14

Ueberlandstrasse 133, 8600 Duebendorf, Switzerland

Eawag,

Swiss

Federal

Institute

for

Environmental

Science

and

Technology,

15 16 17 18

* Corresponding author: phone +49 921 553999, fax +49 921 552334, b.planer-

19

[email protected]

20

ACS Paragon Plus Environment

1

Environmental Science & Technology

21

Page 2 of 24

TOC Graph

22

23 24 25

ACS Paragon Plus Environment

2

Page 3 of 24

Environmental Science & Technology

26

Abstract

27

In most natural groundwaters, sulfide concentrations are low and little attention has been

28

paid to potential occurrence of thioarsenates (AsVSn-IIO4-n3- with n=1-4). Thioarsenate

29

occurrence in groundwater could be critical with regard to the efficiency of iron (Fe)-based

30

treatment technologies because previous studies reported less sorption of thioarsenates to

31

preformed Fe-minerals compared to arsenite and arsenate. We analyzed 273 groundwater

32

samples

33

monothioarsenate (MTA), likely formed via solid-phase zero-valent sulfur, in almost 50% of

34

all samples. Concentrations ranged up to >30 µg L-1 (21% of total As). MTA removal by

35

locally used technologies in which zero-valent or ferrous Fe is oxidized by aeration and As

36

sorbs or co-precipitates with the forming Fe(III)hydroxides was indeed lower than for

37

arsenate. The presence of phosphate required up to three times as much Fe(II) for

38

comparable MTA removal. However, in contrast to previous sorption studies on preformed

39

Fe minerals, MTA removal, even in the presence of phosphate, was still higher than that of

40

arsenite. The more efficient MTA removal is likely caused by a combination of co-

41

precipitation and adsorption rendering the tested Fe-based treatment technologies suitable

42

for As removal also in the presence of MTA.

taken

from

different

wells

in

Bangladesh

over

a

year

and

detected

43 44

Introduction

45

Arsenic (As) is a ubiquitous element and even at trace concentrations carcinogenic to

46

humans. Many aquifers worldwide contain naturally high As concentrations and drinking

47

water is one of the main exposure routes for As to humans. The As problem is especially

48

severe in Southeast Asian countries

49

concentrations up to several mg L-1 As have been reported for groundwater 5. The local

50

drinking water limit is 50 µg L-1 As, which is exceeded in one third of all tubewells

51

leads to a 1% additional cancer risk

52

limit to the 10 µg L-1 recommended by the World Health Organization 8. The two As species

1-4

. In Bangladesh, one of the worst affected countries,

6, 7

5

and still

. Worldwide, many countries have lowered their As

ACS Paragon Plus Environment

3

Environmental Science & Technology

Page 4 of 24

53

typically determined in groundwater are the reduced species arsenite (non-charged below pH

54

9.2 9, H3AsIIIO30,) and the oxidized species arsenate (typically negatively charged, pK

55

(H2AsVO4-/HAsVO42-) = 7.2 9).

56 57

Many of the technologies used for As removal from drinking water rely on iron (Fe)-based

58

treatment strategies. Thereby, As is either adsorbed to pre-formed Fe(III) (hydr)oxide phases

59

or the Fe(III) mineral is formed “concurrently”, i.e. in the presence of As, from added zero-

60

valent iron (ZVI) or ferrous iron (Fe(II)). Arsenic, if present as arsenite, is first oxidized to

61

arsenate and then removed by a combination of sorption and co-precipitation. The major

62

challenge in developing countries like Bangladesh without central water supply and central

63

water treatment facilities is to provide a low-cost and simple technology to affected

64

households that does not require the use of potentially hazardous chemicals. Two examples

65

for concurrent removal technologies are the SORAS method 10 and SONO filters which are

66

more widely used in Bangladesh

67

oxidation of arsenite to arsenate promoted by UV-light and co-precipitation with Fe(III)

68

formed form naturally occurring Fe(II) in polyethylenterephthalate (PET) bottles

69

filters are buckets filled with layers of sand and ZVI, pre-treated with acid to enhance

70

formation of Fe(III) phases and thereby As removal. Total As removal efficiency in all

71

methods might be negatively affected by competition with other groundwater constituents.

72

Silicate hereby mainly affects arsenite, while phosphate has a strong effect on arsenate and

73

arsenite 13.

11, 12

. SORAS (solar oxidation and removal of As) uses

10

. SONO

74 75

A group of As species, typically not considered in the context of natural groundwater

76

occurrence, are thioarsenates. Thioarsenates (AsVS-IInO4-n3-, with n=1-4) are structural

77

analogues to arsenate. At near-neutral pH, thioarsenates do not form directly by addition of

78

sulfide to arsenate, but in two steps via As(III) species

79

exchange for hydroxyl groups first forms thioarsenites (AsIIIS-IInO3-n3-, with n=1-3). Addition of

80

zero-valent sulfur then leads to formation of thioarsenates; As(III) is oxidized to As(V) and

14, 15

ACS Paragon Plus Environment

. Starting from arsenite, sulfide

4

Page 5 of 24

Environmental Science & Technology

81

S(0) is reduced to S(-II) 14, 15. The only thioarsenate species that does not require free sulfide

82

for formation is monothioarsenate (MTA) which forms from arsenite and zero-valent sulfur. It

83

also is the most stable thioarsenate species with very slow transformation rates upon

84

acidification 16, aeration

85

environments, such as in hot springs

86

be the dominant As species. For groundwaters, the only evidence for thioarsenate

87

occurrence so far comes from contaminated sites

88

for groundwater As remediation

89

sulfide concentrations such as in Southeast Asia (1-2 µM 28).

17

, heating 17, or in the absence of zero-valent sulfur 18, 19

22-27

or stagnant terminal lakes

15

20, 21

14

. In sulfide-rich

, thioarsenates can

or sites where sulfate has been applied

but not from natural groundwaters with typically low

90 91

If thioarsenates occur in groundwater used as drinking water their removal must be

92

addressed besides that of arsenite and arsenate because of similar toxicological risk

93

potential concern could be that thioarsenates are removed less efficiently by Fe-based As

94

removal technologies. Thioarsenates were reported to decrease As removal on concurrently

95

formed mackinawite (Fe(II)S)

96

arsenite and arsenate

97

32, 33

98

arsenate, but these experiments were conducted at higher As concentrations than typically

99

encountered in groundwater (0.1 mM

32

31

29, 30

.A

and showed less sorption to pre-formed mackinawite than

. With regard to pre-formed Fe(III) phases, monothioarsenate (MTA)

and trithioarsenate

32

showed slower and less sorption to goethite than arsenite and

32

and 0.5 mM

32, 33

). To our knowledge, there are to

100

date no studies on thioarsenate removal with concurrently formed Fe(III) phases and

101

competition with silicate and phosphate is unknown.

102 103

The present study focuses on MTA, the only thioarsenate species that can form in the

104

absence of free sulfide and the most stable thioarsenate species. It addresses two questions:

105

1) Does MTA form in natural groundwaters of a field site in Bangladesh and if so, does its

106

occurrence correlate with a certain well depth, season, or other hydrogeochemical

107

parameters? 2) How is MTA in comparison to arsenite and arsenate removed with Fe(II) and

108

ZVI in the absence and presence of competitors, such as silicate and phosphate?

ACS Paragon Plus Environment

5

Environmental Science & Technology

Page 6 of 24

109 110

Materials and Methods

111

Field sampling groundwater in Bangladesh. Groundwater samples were collected from

112

seven monitoring wells drilled to different depths (9, 15, 21, 26, 27, 35, 85 m) at a field site in

113

the Titas district in Bangladesh (N23°35,809’, E90°47,857’). Sampling was done almost

114

weekly from the middle of the dry season in February 2007, through the rainy season which

115

peaks in August, to the beginning of the following dry season in December 2007 (31

116

sampling times). There is a gap in our monitoring from July, 25th and September, 2nd where

117

for logistic reasons we could not obtain samples. Additionally, 48 irrigation and drinking water

118

wells in an area of approximately 10 km2 around the field site were sampled between

119

February and May 2007; eight wells were sampled a second time in July 2007. For a detailed

120

description on well installation at the field site, sedimentology, climate, as well as methods

121

and results of general hydrogeochemistry with

122

publication 34.

2

H/18O isotope data see our previous

123 124

Samples for As speciation by ion chromatography coupled to inductively-coupled plasma

125

mass spectrometry (IC-ICP-MS) were filtered (0.2 µm cellulose-acetate), acidified to 2% HCl

126

and stored refrigerated until analysis at the latest 3-4 weeks later following our routine

127

method

128

thioarsenate species. Higher thiolated species, if they were present (see discussion below),

129

would have transformed to arsenite upon acidification

130

that also preserve higher thiolated species such as flash-freezing

131

filtered samples in nitrogen-purged bottles

132

remoteness of the field site, political instability, and daily power cuts. Total Fe, Fe(II), and

133

sulfide were determined photometrically on-site by the HACH ferrozine, phenanthroline, and

134

methylene-blue method, respectively 36.

19

. The use of acid for sample stabilization limited us to detecting MTA as the only

35

16

. Alternative stabilization methods 19

or ethanol addition to

could unfortunately not be realized due to the

135 136

Lab experiments.

ACS Paragon Plus Environment

6

Page 7 of 24

Environmental Science & Technology

137

Experiments included 1) fixed-time batch experiments with varying concentrations of Fe(II)

138

with and without phosphate and silicate, or with and without UV light and addition of citrate,

139

2) time-resolved batch experiments at a fixed Fe(II) concentration with and without

140

phosphate and silicate, 3) time-resolved batch and 4) column experiments using ZVI with

141

and without phosphate addition.

142 13

was used. CaCO3 and MgO (both Sigma-

143

For all experiments, synthetic groundwater

144

Aldrich p.a.) were dissolved in nano-pure water (Barnstead nano-pure TOC-UV) by stirring

145

and purging with CO2 (resulting in a pH of 5-6 and final concentrations of 2.5 mM Ca and

146

1.65 mM Mg). Silicate was added under rapid stirring as a stock solution of Na2SiO3 (Aldrich

147

>98%) to final concentrations of 0.71 mM (20 mg L-1). Selected experiments were conducted

148

without silicate addition. Stirring under CO2 was continued for at least 5 h. The water was

149

then aerated and brought to pH 7.0 by purging with air. Phosphate (if applicable) was added

150

from a stock solution of NaH2PO4 or KH2PO4 (both Fluka puriss p.a.) with a final

151

concentration of 67.7 µM P (3 mg L-1), which did not change the pH. The selected

152

concentration of 3 mg L-1 reflects typical concentrations in As-affected groundwater in

153

Bangladesh (1.9±1.8 mg L-1)

154

sampling (data in Supporting Information, Table S1 & S2; data from the 35 m deep well were

155

excluded since this well is influenced by the overlying peat layer 34). Arsenate or arsenite was

156

added from Na2HAsO4 (Fluka puriss. p.a.) or AsNaO2 (Fluka, purum p.a.) stock solutions,

157

respectively, and MTA was added from a stock solution prepared according to Suess et al. 37.

158

Standard experimental concentrations were 6.7 µM (500 µg L-1 As). Selected experiments

159

were conducted with mixtures of all three As species in equal percentages at total

160

concentrations of 6.7 and 20 µM (500 and 1500 µg L-1), respectively.

1

as well as concentrations observed during our own field

161 162

Fe(II) addition and SORAS batch experiments. For the fixed-time Fe(II) batch

163

experiments, Fe(II) was added from a stock solution of FeSO4*7H2O (Fluka analytics) to the

164

As-containing synthetic groundwater to yield initial Fe(II) concentrations of 1, 2, 3, 5, 7, 10,

ACS Paragon Plus Environment

7

Environmental Science & Technology

Page 8 of 24

165

and 15 mg L-1. Solutions (250 mL) were allowed to react for 4 hours in closed PET bottles

166

(330 mL) with ambient air headspace and shaken intermittently. For the time-resolved Fe(II)

167

batch experiments, a fixed Fe(II) concentration of 15 mg L-1 was chosen and samples were

168

taken after 10, 20, 30, 45, 60, 90, 140, 200, and 300 min. To induce photooxidation as used

169

in the SORAS method, 200 mL samples with varying Fe(II) concentrations were filled into

170

330 mL PET bottles with ambient air headspace and exposed for 5 hours to UV-light in an

171

air-ventilated box with 8 Philips TL20W/05 (actinic blue) lamps. The lamps emit UV-light with

172

a similar spectral composition as UV in sunlight and the intensity in the range of 320-400 nm

173

was roughly equivalent to sunlight at 24° latitude in summer between 9 am and 4 pm (70-95

174

W m-2)

175

to an initial concentration of 50 µM. Citrate was tested because in previous experiments

176

photolysis of Fe(III) citrate complexes had been reported to lead to formation of oxidants and

177

promote Fe(III) flocculation and precipitation 10.

10

. In further experiments, citrate (C6H6Na2O7*1.5H2O, Fluka puriss. p.a.) was added

178 179

MTA oxidation by Fe(III). To test whether Fe(III) formed from Fe(II) acts as oxidizing agent

180

and determines MTA species transformations in solution, we conducted two additional

181

experiments using a lower MTA concentration (100 µg L-1 instead of 500 µg L-1) with 5 mg L-1

182

Fe(III) (molar ratio Fe/As = 67) and 50 mg L-1 Fe(III) (molar ratio Fe/As = 670). Sampling was

183

done after 0.1, 1, 2, 3, and 4 h reaction time. To avoid precipitation, these experiments were

184

conducted at pH 2.

185 186

Zero-valent iron (SONO) experiments. To simulate the patented composite Fe matrix used

187

in SONO-filters

188

and unspecified composition, Fe turnings were produced in Eawag’s mechanics shop from a

189

clean cylinder of CK-45 steel (98.5 % Fe, 0.42 % C, 0.49 % Mn, 0.26 % Si, 0.15 % Cu, all

190

other elements < 0.1 %), without cutting fluid, resulting in clean turnings of 1 mm width,

191

0.1 mm thickness and 50-100 mm length. The turnings were broken up to short pieces

192

logKarsenite 4.5. Mixtures with equal percentages of arsenate, MTA, and arsenite at final

312

concentrations of 500 and 1500 µg L-1 reflect the overall behavior expected from the

313

contributing individual species; at 15 mg L-1 Fe(II), 90 % and 88.5 % of initial total As was

314

removed (see Supporting Information, Figure S2).

13, 43

, arsenate removal by Fe(II) was much more efficient

Fe(II) was required to decrease the initial concentration of arsenate by 90% to less than 50

315 316

ACS Paragon Plus Environment

13

Environmental Science & Technology

% of initial concentration

100

Page 14 of 24

Arsenate MTA

80

Arsenite

60 40 20 0 0.0

2.5

5.0

7.5

10.0

12.5

15.0

Fe(II) mg L-1

317 318

Figure 2 Removal of arsenate, MTA, and arsenite with increasing concentrations of Fe(II) (all

319

experiments were done in the absence of phosphate and the presence of silicate); symbols

320

represent experimental data points, lines represent the output from the model calculation

321

(see Supporting Information, Table S3).

322 323

Influence of phosphate and silicate on monothioarsenate removal by Fe(II)

324

The presence of phosphate significantly increased both the required amount of Fe(II) and

325

kinetics of MTA removal with Fe(II) (Figure 3 & Supporting Information, Figure S3). To

326

remove 90% of initial MTA, about three times as much Fe(II) was required in the presence of

327

phosphate (15 mg L-1 Fe(II)) than in its absence (5 mg L-1 Fe(II)) and it took about five times

328

as long (Figure 3). The logKMTA 5.1 obtained in the presence of phosphate (fit and data

329

shown in Supporting Information, Figure S3 and Table S3, respectively) was lower than in its

330

absence.

331 11, 13

332

Similar effects of phosphate have been described for arsenate and arsenite before

333

the presence of phosphate, the amount of Fe(II) required to remove 90% of initial arsenate

334

and arsenite increased by a factor of five from 2-2.5 to 10-12 mg L-1 Fe(II) and by a factor of

335

1.5 from 30-40 to 50-55 mg L-1 Fe(II), respectively

336

constants for arsenate (logKarsenate 5.6) and arsenite (logKarsenite 3.7) obtained in these

337

previous studies under similar silicate and phosphate conditions

338

determined in the present study (logKMTA 5.1). The observed higher amount of Fe(II) in

339

MTA+P compared to MTA-P experiments is approximately what is needed to form the

11, 13

. In

. The modeled sorption equilibrium

ACS Paragon Plus Environment

11, 13

frame that of MTA

14

Page 15 of 24

Environmental Science & Technology

44

(at 3 mg P L-1 or 96.9 µM =

340

previously described P-Fe precipitate with a P/Fe ratio of 0.7

341

138 µM or 7.7 mg L-1 Fe). The different shape of the removal curves (Fig. 3a & S3) with a

342

steep initial decline in the absence of phosphate but an initial lag phase with very little

343

removal in the presence of phosphate also is a clear indication that, first, phosphate is

344

removed and only after its depletion, MTA follows.

345 Si+ P+

Si- P+

Si+ P-

Si- P-

(a)

80 60 40 20

100

% of initial monothioarsenate

% of initial monothioarsenate

100

0

80 60 40 20 0

0.0

346

(b)

2.5

5.0

7.5

10.0

12.5

15.0

0

50

Fe(II) (mg L-1)

100

150

200

250

300

Time (min)

347

Figure 3 a) Effect of phosphate and silicate on required amount of Fe(II) (after a reaction time

348

of 4 hours) and b) kinetics (at 15 mg L-1 Fe(II) addition) of MTA removal by Fe(II); Si+ reflects

349

addition of 20 mg Si L-1 and P+ addition of 3 mg L-1 phosphate to solution; no silicate or

350

phosphate was added to the Si- and P- treatments

351 352

Silicate, in contrast to phosphate, had almost no effect on MTA removal efficiency (Figure 3),

353

which is in line with previous observations for arsenate

354

charged arsenite, but not as effectively with anions such as arsenate or, like in our case,

355

MTA. In case of As removal with Fe(II), previous studies have even shown only a minor

356

effect of silicate on the removal of arsenite because there is little direct arsenite sorption but

357

mainly oxidation of arsenite to arsenate before removal by Fe(III) phases

358

consistent effect of silicate was, however, observed in the treatments without phosphate.

359

Both at high Fe(II) concentrations (> 7 mg L-1) and for longer reaction times (> 90 min), MTA

360

concentrations in solution re-increased in the absence of silicate, but not in its presence. A

361

similar silicate effect was reported for arsenate removal before 45. The authors showed that in

362

the absence of silicate the observed As re-increase in solution is explained by a

363

transformation from amorphous Fe phosphate and arsenate phases (with high sorption site

13

. Silicate competes with the non-

ACS Paragon Plus Environment

13

. A small, but

15

Environmental Science & Technology

Page 16 of 24

364

densities) to ferrihydrite and poorly crystallized lepidocrocite (with low sorption site densities)

365

45, 46

366

reaction times, no re-increase in As could be observed. However, overall in contrast to the

367

large effect of phosphate on MTA removal, silicate only played a minor role.

. Silicate slows down this transformation rate which explains why for comparable

368 369

Monothioarsenate species transformation during removal by Fe(II)

370

Chromatographic analysis of the MTA standard used for the As removal experiments showed

371

that its purity was 95 % (besides 3.5 % arsenite and 1.5 % arsenate). Speciation analysis

372

showed only little transformation without Fe(II) addition during 4 hours reaction times with

373

ambient air headspace (purity 93% at 0 mg L-1 Fe(II) in Figure 4a). This observation is in

374

accordance with previous reports on the stability of MTA upon exposure to atmospheric

375

oxygen

376

largest part this is removal by sorption to Fe(III) phases as discussed above, there is also

377

some species transformation leading to increases in absolute concentrations of arsenite and

378

(despite better removal compared to MTA) also of arsenate. Relative shares at 5 mg L-1

379

added Fe(II) were 74% MTA, 6% arsenite, and 20% arsenate (Figure 4a).

17

. With increasing Fe(II) addition, MTA concentrations decrease. While for the

100

(a)

100

10

0.0

2.5

5.0

7.5

10.0

12.5

15.0

As concentration in µg L-1

As concentration in µg L-1

1,000

(b)

10

1

0

50

100

Fe(II) (mg L-1)

380

MTA

Arsenite

150

200

250

Time (min) Arsenate

MTA

Arsenite

Arsenate

381 382

Figure 4 a) Monothioarsenate species transformation after a reaction time of 4 hours at

383

Fe(II)/As molar ratios of 0, 5, 13, and 40 at pH 7 and b) after a reaction time of 0.1, 1, 2, 3, 4

384

hours for an Fe(III)/As molar ratio of 67 at pH 2 (the low pH in this experiment was chosen to

385

avoid Fe(III) precipitation/sorption which would have affected As species distribution in

386

solution due to the different removal efficiencies for each species); note the logarithmic y-axis

387 ACS Paragon Plus Environment

16

Page 17 of 24

Environmental Science & Technology

388

In a separate experiment we tested whether Fe(III) formed from Fe(II) acts as oxidizing agent

389

and determines MTA species transformations in solution. Figure 4b shows that the first step

390

of Fe(III)-driven MTA transformation is disproportionation to arsenite (and zero-valent sulfur,

391

not measured here, but confirmed in previous reports

392

disproportionation is likely further oxidation of zero-valent sulfur by Fe(III). The sulfur

393

oxidation product was not determined in this experiment, but in a previous study we

394

confirmed that oxidation of zero-valent sulfur by Fe(III) yields thiosulfate

395

oxidation of zero-valent sulfur leads to a shift in equilibrium between MTA and arsenite/zero-

396

valent sulfur to the side of arsenite. Arsenite thereby only accumulates temporarily, seen e.g.

397

with a maximum share of 44 % after 1 hour at a Fe/As ratio of 67 (Figure 4b). The extent of

398

MTA dissociation vs. onward oxidation of arsenite to arsenate depends on the Fe/As ratio

399

and time. Higher Fe/As ratios (Supporting Information, Figure S4) and longer reaction times

400

(Figure 4a) promote the second step of further Fe(III)-driven oxidation of arsenite to

401

arsenate, as has been observed before

402

oxidation apparently was so fast that arsenite only appeared in minor shares as intermediate

403

product and sorption of MTA and arsenate determined the overall As removal.

12, 13, 48

14

). The reason for the initiation of

47

. The onward

. For all our sorption experiments, secondary

404 405

The influence of UV light and citrate addition on monothioarsenate removal by Fe(II)

406

(SORAS)

407

Including UV light and citrate addition in the Fe(II) removal experiments to mimic exactly the

408

SORAS method, did not change total MTA removal efficiency (Supporting Information, Figure

409

S5a) nor MTA species transformation (Supporting Information, Figure S5b). This observation

410

is in line with the known stability of MTA under ambient conditions even upon exposure to

411

atmospheric oxygen and hours of heating

412

oxidizing agents like H2O2

413

double-bond sulfur, leading to MTA transformation to arsenite. For the overall removal

414

efficiency by the SORAS method, this means UV light and citrate addition can be applied as

17

17

, or acidification to pH 2

9, 16

.

Only strong

or the above described Fe(III) are capable of oxidizing the

ACS Paragon Plus Environment

17

Environmental Science & Technology

10

Page 18 of 24

to oxidize arsenite to arsenate without yielding any additional undesired

415

described before

416

arsenite from MTA transformation.

417 418

Monothioarsenate removal by ZVI – batch experiments

419

Removal rates in batch experiments with ZVI followed the same order as observed for the

420

Fe(II) batch experiments with the highest removal rate for arsenate, closely followed by MTA,

421

and the lowest removal rate for arsenite (Figure 5a). Phosphate, again, decreased the

422

removal rates, stronger for arsenate and MTA than for arsenite. Still, despite significant

423

differences in removal kinetics, independent of As species and absence or presence of

424

phosphate, the high amount of reactive Fe applied (6.7 g L-1 ZVI compared to max. 15 mg L-1

425

Fe(II)) was sufficient to remove more than 99% of the initially applied As in a maximum

426

reaction time of 4 hours. 1,000

(a) As concentration in µg L-1

% of initial concentration

120 100 80 60 40 20

(b)

100

10

1

0 0

50

100

150

200

250

0

50

100

Time (min)

427

150

200

250

Time (min)

MTA P+

Arsenite P+

Arsenate P+

MTA P-

Arsenite P-

Arsenate P-

MTA

DTA

Arsenite

Arsenate

428

Figure 5 a) Removal of arsenate, MTA, and arsenite by ZVI (6.7 g L-1) in the absence (P-)

429

and presence (P+) of phosphate and b) species distribution in the MTA solution in the

430

absence of phosphate; note the logarithmic y-axis. All experiments with 20 mg L-1 Si.

431 432

During reaction with ZVI, a slight increase in total arsenite concentrations was observed both

433

in the absence (Figure 5b) and with a slight temporal delay, also in the presence of

434

phosphate (Supporting Information, Figure S6b). Interestingly, minor amounts (2-10 µg L-1

435

corresponding to 3-5% of total As) of dithioarsenate (AsVS-II2O23-, DTA) were observed in the

436

first 30 min. DTA can only form if some of the zero-valent sulfur liberated during MTA ACS Paragon Plus Environment

18

Page 19 of 24

Environmental Science & Technology

437

transformation to arsenite was further reduced to sulfide, reacted with arsenite to form

438

monothioarsenite (AsIIIS-IIO23-) which in turn reacted with zero-valent sulfur to form the DTA.

439

Dissolved arsenate in comparison to arsenite contributed little (at maximum 6%) and was not

440

detectable from 60 min on. In comparison to the Fe(II) experiments, a much higher share of

441

arsenite remained in solution. It contributed almost 50% to total dissolved As after 30 min,

442

80% after 60 min and was the only detectable dissolved species from 90 min on. The reason

443

could be either the significantly better sorption of MTA and arsenate versus arsenite or less

444

on-ward oxidation of arsenite to arsenate owed to the fact that more oxygen is required to

445

transform ZVI first to Fe(II) before it is further oxidized to Fe(III) phases. Overall, however,

446

direct sorption of MTA (and arsenate) was so fast that the formation of arsenite did not

447

significantly influence the overall As removal efficiency (when it became dominant > 30 min,

448

total As concentrations had already dropped to about 10% of the original concentration).

449 450

Monothioarsenate removal by ZVI – column experiments (SONO filters)

451

The ZVI column experiments mimicking the SONO filter confirmed the order observed in all

452

previous batch experiments. In total, the column was loaded 20 times in 10 days. As reported

453

before

454

oxidation to Fe(II) and various mixed-valence Fe(II/III) and Fe(III)-phases. After some initial

455

variations, As concentrations at the outflow stabilized and were lowest for arsenate (6.2 ± 2.3

456

µg L-1), followed by MTA (28 ± 2.7 µg L-1), and highest for arsenite (76 ± 5.2 µg L-1) (all

457

values are averages over the last six column loadings in Figure 6). The MTA breakthrough in

458

the presence of phosphate was initially very high (maximum values of 146 ± 4.5 µg L-1), but

459

dropped after 15 column loadings to average values of 53 ± 11.4 µg L-1. The same delayed

460

removal of MTA in the presence of phosphate has been observed in the ZVI batch

461

experiments (Figure 5).

12

, the periodic change between flowing and stagnant conditions promotes ZVI

462 463

Speciation analyses of the outflow solutions of each experiment after 2, 7, 14, and 20 column

464

loadings (Supporting Information, Table S4) confirmed the high removal efficiency for MTA ACS Paragon Plus Environment

19

Environmental Science & Technology

Page 20 of 24

465

and arsenate compared to low removal efficiency for arsenite. From the 2nd to the 21st

466

column loading with MTA (original purity 95%), absolute outflow concentrations of MTA

467

decreased from 55 µg L-1 to 8 µg L-1, while arsenite remained almost constant (11 and 14 µg

468

L-1). The preferential removal of MTA led to a relative increase in the share of arsenite from

469

14% to 54% in total As remaining in solution, while MTA dropped from 72% to 30%. Some

470

non-adsorbed arsenate (3 µg L-1 = 4% of total As) and minor reduction to arsenite (2 µg L-1 =

471

19% of total As) was also observed when applying a pure arsenite or a pure arsenate

472

solution, respectively (Supporting Information, Table S4).

473 180

Arsenate PMTA PArsenite PMTA P+

160

As in µg L-1

140 120 100 80 60 40 20 0 0

5

10

15

20

Column loadings

474 475 476

Figure 6 Concentrations of arsenate, MTA, and arsenite in the absence of phosphate (P-)

477

and MTA in the presence of phosphate (P+) over time at the outflow of the ZVI column; one

478

column loading equals approximately 90 mL synthetic groundwater solution with the

479

respective As spike (first column loading was discarded due to dilution effects). Inflow As-

480

concentrations were 500 µg L-1. All experiments with 20 mg L-1 Si.

481 482

Implications of MTA occurrence in groundwater

483

The present study has shown that MTA can occur in significant quantities in natural

484

groundwaters even when there is only low or no detectable free sulfide. The reason is that it

485

does not form from arsenate and sulfide but from arsenite and zero-valent sulfur which is

486

most likely bound to Fe phases or organic matter surfaces (solid phase vs. aqueous phase

487

driven formation). Clearly, sulfide is not an appropriate proxy to exclude existence of all

488

thioarsenates and more studies are needed to understand occurrence and fate of MTA in

489

natural groundwaters worldwide. Knowing about MTA occurrence in groundwater is ACS Paragon Plus Environment

20

Page 21 of 24

Environmental Science & Technology

490

important for choosing the appropriate As removal technology. For groundwaters containing

491

MTA, removal technologies should be chosen where the Fe sorbent is formed in the

492

presence of As because co-precipitation seems to be much more efficient than sorption for

493

MTA removal.

494 495

Acknowledgement

496

This work was funded by the German Research Foundation Grant PL 302/2-1 and PL 302/3-

497

1 (Emmy Noether Program).

498 499 500

Associated Content – Supporting Information

501

Hydrogeochemistry of wells in Bangladesh; modelled sorption equilibrium constants, species

502

distribution at the outflow of a SONO filter column; PCA analysis for hydrogeochemistry;

503

removal efficiency for mixtures of arsenate, MTA, and arsenite with increasing concentrations

504

of Fe(II), removal of arsenate, MTA, and arsenite in the absence of phosphate, of MTA in the

505

absence and presence of phosphate and of phosphate (in the presence of MTA); MTA

506

species transformation over time for an Fe/As ratio of 670; MTA removal efficiency and

507

speciation with UV light and citrate addition; MTA species transformation by ZVI in the

508

absence and presence of phosphate. The Supporting Information is available free of charge

509

on the ACS Publications website at ###.

510 511

ACS Paragon Plus Environment

21

Environmental Science & Technology

Page 22 of 24

512

References

513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561

1. Hug, S. J.; Leupin, O. X.; Berg, M., Bangladesh and Vietnam: Different groundwater compositions require different approaches to arsenic mitigation. Environmental Science & Technology 2008, 42, (17), 6318-6323. 2. Chowdhury, U. K.; Biswas, B. K.; Chowdhury, T. R.; Samanta, G.; Mandal, B. K.; Basu, G. C.; Chanda, C. R.; Lodh, D.; Saha, K. C.; Mukherjee, S. K.; Roy, S.; Kabir, S.; Quamruzzaman, Q.; Chakraborti, D., Groundwater arsenic contamination in Bangladesh and West Bengal, India. Environ. Health Persp. 2000, 108, (5), 393-397. 3. Nickson, R.; McArthur, J.; Burgess, W.; Ahmed, K. M.; Ravenscroft, P.; Rahman, M., Arsenic poisoning of Bangladesh groundwater. Nature 1998, 395, (6700), 338-338. 4. Smith, A. H.; Lingas, E. O.; Rahman, M., Contamination of drinking-water by arsenic in Bangladesh: a public health emergency. Bulletin of the World Health Organization 2000, 78, (9), 1093-103. 5. Gaus, I.; Kinniburgh, D. G.; Talbot, J. C.; Webster, R., Geostatistical analysis of arsenic concentration in groundwater in Bangladesh using disjunctive kriging. Environ. Geol. 2003, 44, (8), 939-948. 6. Smith, A. H.; Hopenhayn-Rich, C.; Bates, M. N.; Goeden, H. M.; Hertz-Picciotto, I.; Duggan, H. M.; Wood, R.; Kosnett, M. J.; Smith, M. T., Cancer risks from arsenic in drinking water. Environ. Health Persp. 1992, 97, 259-267. 7. Goyer, R. A.; Aposhian, H. V.; Brown, K. G.; Cantor, K. P.; G.P., C.; Cullen, W.; Daston, G. P.; Fowler, B. A.; C.D., K.; Kosnett, M. J.; Mertz, W.; Preston, R. J.; Ryan, L. M.; Smith, A. H.; Vather, M.; Wiencke, J. K., Arsenic in Drinking Water. The National Academies Press: Washington, D.C., 1999; p 330. 8. World_Health_Organization, Guidelines for drinking-water quality. 4th ed.; 2011; p 564. 9. Thilo, E.; Hertzog, K.; Winkler, A., Über Vorgänge bei der Bildung des Arsen(V)-sulfids beim Ansäuern von Tetrathioarsenatlösungen. Z. anorg. allg. Chem. 1970, 373, 111-121. 10. Hug, S. J.; Canonica, L.; Wegelin, M.; Gechter, D.; von Gunten, U., Solar Oxidation and Removal of Arsenic at Circumneutral pH in Iron Containing Waters. Environ. Sci. Technol. 2001, 35, 2114-2121. 11. Hussam, A.; Munir, A. K. M., A simple and effective arsenic filter based on composite iron matrix: Development and deployment studies for groundwater of Bangladesh. Journal of Environmental Science and Health Part a-Toxic/Hazardous Substances & Environmental Engineering 2007, 42, (12), 1869-1878. 12. Neumann, A.; Kaegi, R.; Voegelin, A.; Hussam, A.; Munir, A. K. M.; Hug, S. J., Arsenic Removal with Composite Iron Matrix Filters in Bangladesh: A Field and Laboratory Study. Environmental Science & Technology 2013, 47, (9), 4544-4554. 13. Roberts, L. C.; Hug, S. J.; Ruettimann, T.; Billah, M.; Khan, A. W.; Rahman, M. T., Arsenic removal with iron(II) and iron(III) waters with high silicate and phosphate concentrations. Environmental Science & Technology 2004, 38, (1), 307-315. 14. Planer-Friedrich, B.; Haertig, C.; Lohmayer, R.; Suess, E.; McCann, S. H.; Oremland, R., Anaerobic Chemolithotrophic Growth of the Haloalkaliphilic Bacterium Strain MLMS 1 by Disproportionation of Monothioarsenate. Environmental Science & Technology 2015, 49, (11), 6554– 6563. 15. Stauder, S.; Raue, B.; Sacher, F., Thioarsenates in sulfidic waters. Environ. Sci. Technol. 2005, 39, (16), 5933-5939. 16. Planer-Friedrich, B.; Wallschläger, D., A critical investigation of hydride generation-based arsenic speciation in sulfidic waters. Environmental Science & Technology 2009, 43, 5007-5013. 17. Planer-Friedrich, B.; Fisher, J. C.; Hollibaugh, J. T.; Süß, E.; Wallschläger, D., Oxidative transformation of trithioarsenate along alkaline geothermal drainages – abiotic versus microbially mediated processes. Geomicrobiology 2009, 26, 339-350.

ACS Paragon Plus Environment

22

Page 23 of 24

562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613

Environmental Science & Technology

18. Guo, Q.; Planer-Friedrich, B.; Liu, M.; Li, J.; Zhou, C.; Wang, Y., Arsenic and thioarsenic species in the hot springs of the Rehai magmatic geothermal system, Tengchong volcanic region, China. Chemical Geology 2017, 453, (Supplement C), 12-20. 19. Planer-Friedrich, B.; London, J.; McCleskey, R. B.; Nordstrom, D. K.; Wallschläger, D., Thioarsenates in geothermal waters of Yellowstone National Park: determination, preservation, and geochemical role. Environ. Sci. Technol. 2007, 41, (15), 5245-5251. 20. Fisher, J. C.; Wallschlaeger, D.; Planer-Friedrich, B.; Hollibaugh, J. T., A new role for sulfur in arsenic cycling. Environmental Science & Technology 2008, 42, 81-85. 21. Hollibaugh, J. T.; Carini, S.; Gürleyük, H.; Jellison, R.; Joye, S. B.; LeCleir, G.; Meile, C.; Vasquez, L.; Wallschläger, D., Arsenic speciation in Mono Lake, California: Response to seasonal stratification and anoxia. Geochim. Cosmochim. Ac. 2005, 69, (8), 1925-1937. 22. Keimowitz, A. R.; Mailloux, B. J.; Cole, P.; Stute, M.; Simpson, H. J.; Chillrud, S. N., Laboratory investigations of enhanced sulfate reduction as a groundwater arsenic remediation strategy. Environmental Science & Technology 2007, 41, (19), 6718-6724. 23. Lee, M. K.; Saunders, J. A.; Wilkin, R. T.; Mohammad, S., Geochemical modeling of arsenic speciation and mobilization: Implications for bioremediation. Advances in Arsenic Research 2005, 915, 398-413. 24. Lee, M.-K.; Saunders, J. A.; Harrington, J.; Lutes, C.; Wilkin, R. In Strategy for in situ bioremediation of arsenic in groundwater: Field and modeling studies, Proceedings of the Fourth International Conference on Remediation of Chlorinated and Recalcitrant Compounds, (eds.), Monterey, CA, 2004; A.R., G.; Chen, A. S. C., Eds. Battelle Press, Columbus, OH: Monterey, CA, 2004; pp Paper 4C-06. 25. Xie, X.; Wang, Y.; Li, J.; Yu, Q.; Wu, Y.; Su, C.; Duan, M., Effect of irrigation on Fe(III)–SO42− redox cycling and arsenic mobilization in shallow groundwater from the Datong basin, China: Evidence from hydrochemical monitoring and modeling. Journal of Hydrology 2015, 523, (Supplement C), 128-138. 26. Stucker, V. K.; Williams, K. H.; Robbins, M. J.; Ranville, J. F., Arsenic geochemistry in a biostimulated aquifer: An aqueous speciation study. Environ. Toxicol. Chem. 2013, 32, (6), 1216-1223. 27. Stucker, V. K.; Silverman, D. R.; Williams, K. H.; Sharp, J. O.; Ranville, J. F., Thioarsenic Species Associated with Increased Arsenic Release during Biostimulated Subsurface Sulfate Reduction. Environmental Science & Technology 2014, 48, (22), 13367-13375. 28. Zheng, Y.; Stute, M.; van Geen, A.; Gavrieli, I.; Dhar, R.; Simpson, H. J.; Schlosser, P.; Ahmed, K. M., Redox control of arsenic mobilization in Bangladesh groundwater. Appl. Geochem. 2004, 19, (2), 201-214. 29. Hinrichsen, S.; Geist, F.; Planer-Friedrich, B., Inorganic and methylated thioarsenates pass the gastrointestinal barrier. Chem. Res. Toxicol. 2015, 28, (9), 1678-1680. 30. Hinrichsen, S.; Lohmayer, R.; Zdrenka, R.; Dopp, E.; Planer-Friedrich, B., Effect of sulfide on the cytotoxicity of arsenite and arsenate in human hepatocytes (HepG2) and human urothelial cells (UROtsa). Environ. Sci. Pollut. R. 2014, 121, (17), 10151-10162. 31. Vega, A. S.; Planer-Friedrich, B.; Pastén, P. A., Arsenite and arsenate immobilization by preformed and concurrently formed disordered mackinawite (FeS). Chemical Geology 2017, 475, 6275. 32. Couture, R. M.; Rose, J.; Kumar, N.; Mitchell, K.; Wallschlager, D.; Van Cappellen, P., Sorption of arsenite, arsenate, and thioarsenates to iron oxides and iron sulfides: a kinetic and spectroscopic investigation. Environmental Science & Technology 2013, 47, (11), 5652-5659. 33. Suess, E.; Planer-Friedrich, B., Thioarsenate formation upon dissolution of orpiment and arsenopyrite. Chemosphere 2012, 89, (11), 1390-1398 34. Planer-Friedrich, B.; Haertig, C.; Lissner, H.; Steinborn, J.; Suess, E.; Hassan, M. Q.; Zahid, A.; Alam, M.; Merkel, B., Organic carbon mobilization in a Bangladesh aquifer explained by seasonal monsoon-driven storativity changes. Appl. Geochem. 2012, 27, 2324–2334. 35. Suess, E.; Mehlhorn, J.; Planer-Friedrich, B., Anoxic, ethanolic, and cool - an improved method for thioarsenate preservation in iron-rich waters. Appl. Geochem. 2015, 62, 224–233.

ACS Paragon Plus Environment

23

Environmental Science & Technology

614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648

Page 24 of 24

36. Cline, J. D., Spectrophotometric determination of hydrogen sulfide in natural waters. Limnology and Oceanography 1969, 14, 454-458. 37. Suess, E.; Scheinost, A. C.; Bostick, B. C.; Merkel, B. J.; Wallschläger, D.; Planer-Friedrich, B., Discrimination of thioarsenites and thioarsenates by X-ray Absorption Spectroscopy. Anal. Chem. 2009, 81, 8318-8326. 38. Braun, W.; Herron, J. T.; Kahaner, D. K., Acuchem - a Computer-Program for Modeling Complex Chemical-Reaction Systems. Int J Chem Kinet 1988, 20, (1), 51-62. 39. R, R. D. C. T., A Language and Environment for Statistical Computing. R Foundation for Statistical Computing: Vienna, Austria, 2008. 40. Wan, M. L.; Shchukarev, A.; Lohmayer, R.; Planer-Friedrich, B.; Peiffer, S., Occurrence of Surface Polysulfides during the Interaction between Ferric (Hydr)Oxides and Aqueous Sulfide. Environmental Science & Technology 2014, 48, (9), 5076-5084. 41. ThomasArrigo, L. K.; Mikutta, C.; Lohmayer, R.; Planer-Friedrich, B.; Kretzschmar, R., Sulfidization of Organic Freshwater Flocs from a Minerotrophic Peatland: Speciation Changes of Iron, Sulfur, and Arsenic. Environmental Science & Technology 2016, 50, (7), 3607-3616. 42. Langner, P.; Mikutta, C.; Kretzschmar, R., Arsenic sequestration by organic sulphur in peat. Nature Geoscience 2012, 5, (1), 66-73. 43. Meng, X. G.; Korfiatis, G. P.; Bang, S. B.; Bang, K. W., Combined effects of anions on arsenic removal by iron hydroxides. Toxicology Letters 2002, 133, (1), 103-111. 44. Senn, A. C.; Kaegi, R.; Hug, S. J.; Hering, J. G.; Mangold, S.; Voegelin, A., Composition and structure of Fe(III)-precipitates formed by Fe(II) oxidation in water at near-neutral pH: Interdependent effects of phosphate, silicate and Ca. Geochim. Cosmochim. Ac. 2015, 162, 220-246. 45. Senn, A. C.; Hug, S. J.; Kaegi, R.; Hering, J. G.; Voegelin, A., Arsenate co-precipitation with Fe(II) oxidation products and retention or release during precipitate aging. Water Res 2018, 131, 334345. 46. Senn, A. C.; Kaegi, R.; Hug, S. J.; Hering, J. G.; Mangold, S.; Voegelin, A., Effect of aging on the structure and phosphate retention of Fe(III)-precipitates formed by Fe(II) oxidation in water. Geochim. Cosmochim. Ac. 2017, 202, 341-360. 47. Lohmayer, R.; Kappler, A.; Lösekann-Behrens, T.; Planer-Friedrich, B., Role of sulfur species as redox partners and electron shuttles for ferrihydrite reduction by Sulfurospirillum deleyianum. Appl. Environ. Microb. 2014, 80, (10), 3141-3149. 48. Leupin, O. X.; Hug, S. J., Oxidation and removal of arsenic (III) from aerated groundwater by filtration through sand and zero-valent iron. Water Res. 2005, 39, (9), 1729-1740.

ACS Paragon Plus Environment

24