A new method for antimony speciation in plant biomass and nutrient media using anion exchange cartridge

Author’s Accepted Manuscript A new method for antimony speciation in plant biomass and nutrient media using anion exchange cartridge Rujira Tisarum, Jing–Hua Ren, Xiaoling Dong, Hao Chen, Jason T. Lessl, Lena Q. Ma www.elsevier.com/locate/talanta

PII: DOI: Reference:

S0039-9140(15)30198-3 http://dx.doi.org/10.1016/j.talanta.2015.07.073 TAL15838

To appear in: Talanta Received date: 14 April 2015 Revised date: 22 July 2015 Accepted date: 28 July 2015 Cite this article as: Rujira Tisarum, Jing–Hua Ren, Xiaoling Dong, Hao Chen, Jason T. Lessl and Lena Q. Ma, A new method for antimony speciation in plant biomass and nutrient media using anion exchange cartridge, Talanta, http://dx.doi.org/10.1016/j.talanta.2015.07.073 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1 2 3 4

A new method for antimony speciation in plant biomass and nutrient media using anion exchange cartridge

5 6 7 8

Rujira Tisaruma,b, Jing–Hua Rena, Xiaoling Dongb, Hao Chenb, Jason T. Lessla,b and Lena Q.

9

Maa,b*

10 11 12

a

13

University, Jiangsu 210046, China

14

b

State Key Lab of Pollution Control and Resource Reuse, School of the Environment, Nanjing

Soil and Water Science Department, University of Florida, Gainesville, FL 32611 USA

15 16 17 18 19 20

________________________

21

*

22

Environment, Nanjing University, Jiangsu 210046, China; Tel./fax: +86 025 8969 0631,

23

[email protected]

24

Corresponding author. State Key Lab of Pollution Control and Resource Reuse, School of the

1

25

Abstract

26

A selective separation method based on anion exchange cartridge was developed to determine

27

antimony (Sb) speciation in biological matrices by graphite furnace atomic absorption

28

spectrophotometry (GFAAS). The selectivity of the cartridge towards antimonite [SbIII] and

29

antimonate [SbV] reversed in the presence of deionized (DI) water and 2 mM citric acid. While

30

SbV was retained by the cartridge in DI water, SbIII was retained in citric acid media. At pH 6,

31

SbIII and SbV formed SbIII- and SbV-citrate complexes, but the cartridge had higher affinity

32

towards the SbIII-citrate complex. Separation of SbIII was tested at various concentrations in fresh

33

and spent growth media and plant tissues. Our results showed that cartridge-based Sb speciation

34

was successful in plant tissues, which was confirmed by HPLC–ICP–MS. The cartridge retained

35

SbIII and showed 92–104% SbV recovery from arsenic hyperaccumulator Pteris vittata roots

36

treated with SbIII and SbV. The cartridge procedure is an effective alternative for Sb speciation,

37

offering low cost, reproducible results, and simple Sb analysis using GFAAS.

38 39

Keywords: antimony; speciation; antimonite; antimonate; solid phase extraction

40 41

2

42 43

Introduction Antimony (Sb) is a toxic element and is commonly associated with sulfur minerals [1].

44

Total Sb concentrations in the earth’s crust are 0.2–0.3 mg/kg whereas the background

45

concentrations in surface soil are 0.02–7.01 mg/kg [2]. Sb has been used in batteries, car brake

46

linings and in fire retardants and it is released to the environment at 38,000 tons per year from

47

both natural and anthropogenic sources [3]. A steady increase of Sb levels in snow and ice in the

48

Arctic and in the aerosol samples from Tokyo indicate that Sb contamination is a worldwide

49

environmental problem [4].

50

Sb is a possible carcinogen, with no known biological function [5]. It induces keratitis,

51

dermatitis, conjunctivitis and gastritis and has been shown to cause lung cancer in rats [6]. In

52

plants, Sb inhibits photosynthesis and decreases utilization of essential elements [7]. The toxicity

53

of Sb largely depends on its chemical species, with inorganic species being more toxic than

54

organic [8] and the reduced inorganic species SbIII being 10 times more toxic than the oxidized

55

SbV species [6].

56

The amount and type of Sb taken up by plants vary with species. For example, rye takes

57

up more SbV than SbIII whereas wheat and rice take up more SbIII than SbV [9, 10]. Sb in plant

58

tissues exists as both inorganic and organic forms, with inorganic Sb being the dominant species

59

[11, 12]. Thus, information on Sb speciation is important to understand Sb behavior in the

60

environment as well as Sb translocation and transformation in plants.

61

There are several analytical techniques to analyze total Sb, such as low cost equipment

62

graphite furnace atomic absorption spectrophotometry (GFAAS) and high cost equipment

63

inductively coupled plasma mass spectrometry (ICP–MS). In term of Sb speciation, the

64

commonly used analytical method couples HPLC with ICP–MS [13]. Other Sb speciation 3

65

methods focus on Sb separation based on the affinity of SbIII for sulfur group (–S). For example,

66

pyrrolidine dithiocarbamate (PDC) containing –S group has been used for SbIII separation with

67

non-polar organic solvents such as methyl isobuthyl ketone [14] or xylene [15]. However, this

68

technique has the disadvantage of requiring manual injection of the organic media into the

69

graphite furnace [15]. Using a column to separate SbIII–PDC complexes has been developed to

70

overcome liquid–liquid extraction methods. Separating SbIII–PDC complexes requires a solid

71

phase extraction (SPE) based on Chromosorb®self-packed resin column [16], which might cause

72

inaccurate results. The non-polar SPE Isolute® silica-based octyl cartridge retains SbIII–PDC

73

complex [17], but the cartridge is not globally available.

74

In addition to PDC chelating agent, there are several SPE techniques for Sb speciation.

75

For example, an ionic Amberlite® IRA 910 resin for adsorbing SbV[18], Dowex® 1x4 resin for

76

retaining SbIII–Cl complex [19], and modified columns with dimercaptosuccinic acid or

77

tetraethylenepentamine for retaining SbIII [20, 21]. However, those techniques have not been

78

successful for Sb speciation as they are limited by graphite furnace analysis, procedure

79

complication and commercial availability. For example, an ionic Amberlite® IRA 910 column

80

requires self-packing and the Dowex® 1x4 column requires HCl elution, which is impractical for

81

analysis by GFAAS. Furthermore, dimercaptosuccinic acid to chemically modifying mesoporous

82

TiO2 micro-columns or tetraethylenepentamine to bind silica gel columns are not commercially

83

available so they need be synthesized in a laboratory. The reproducibility of those results

84

remains uncertain.

85

The pre-packed strong anion exchange cartridges proposed in this study have the ability

86

to separate SbIII from SbV, which can be detected by GFAAS or ICP. The cartridges are

87

reasonably priced and ready for immediate use. In addition, they are reproducible and 4

88

commercially available worldwide. This work was the first application of SPE in separating Sb

89

speciation in plant biomass and hydroponic nutrient media. It also sheds light on Sb speciation in

90

plant tissues and media while offering an alternative method against expensive HPLC technique.

91

The aim of this study was to investigate a silica-based anion exchange cartridge for SbIII

92

and SbV separation under pH controlled conditions for Sb analysis by GFAAS. Furthermore, the

93

results from Sb speciation in the roots of As hyperaccumulator Pteris vittata were confirmed by

94

HPLC–ICP–MS.

95

Material and methods

96

All chemicals were of analytical reagent grade. Standard stock solutions of SbIII and SbV

97

(1 g/L) were prepared by dissolving potassium antimonyltartrate (C8H4K2O12Sb2·3H2O, Fisher)

98

and potassium hexahydroxyantimonate (KSb(OH)6, Sigma–Aldrich) in DI water and hot DI

99

water, respectively. The SbIII and SbV stock solutions were stored in a refrigerator at 4ºC.

100

Working solutions were freshly prepared daily by further dilution from these stock solutions to

101

the ranges needed. A citric acid solution was prepared by dissolving monohydrate citric acid

102

(HOC(COOH)(CH2COOH)2.H2O, J.T. Baker) in DI water. NaOH and HCl were used for pH

103

adjustments. All labware used in this study was washed and soaked in 1 M nitric acid for 24 h

104

and rinsed several times with tap water followed by DI water before use.

105

Instrument for Sb Analysis and Speciation

106

An AA240Z Zeeman atomic absorption spectrometer, equipped with a GTA 120 graphite

107

furnace was used together with an Agilent Technology Sb hollow cathode lamp as the light

108

source at 10 mA. The wavelength for Sb was 217.6 nm, with peak area mode being used. The

109

linear range was 10150 μg/L with the slit at 0.2 nm. Partition-coated graphite tubes (Agilent

110

Technologies) along with a programmable sample dispenser and background corrector were used. 5

111

An ultra-high purity Argon gas was used as the purge gas. A 10 μL of sample and 5 μL of 500

112

mg/L palladium nitrate modifier solution containing 0.25% citric acid in 2.5% nitric acid were

113

introduced with an auto-sampler. The graphite furnace temperature program for Sb

114

determination was given in Table 1.

115

In addition to GFAAS, HPLC–ICP–MS was used to verify the accuracy of the cartridge-

116

based method in 2 mM citric acid system. A Hamilton PRP–X100 AEC column (250x4.1 mm,

117

10 µM; Hamilton, UK) with a guard column combined with a Waters 2695 HPLC system to

118

separate Sb species. The mobile phase consisted of 20 mM EDTA–(NH4)2 and 2 mM potassium

119

hydrogen phthalate adjusted to pH 4.5 with ammonium hydroxide. The sample was sonicated

120

and filtered (0.22 µm), with 50 µL being injected with a flow rate at 1.2 mL/min at 40°C. A

121

Nexion 300X inductively coupled plasma–mass spectrometry (ICP–MS) (Perkin–Elmer, USA)

122

was connected to HPLC to analyze Sb concentration. SbV and SbIII standard solutions at 10 µg/L

123

were run to obtain retention time. Stock solutions of Sb species were prepared from

124

C8H4K2O12Sb2·3H2O (Sigma–Aldrich, >99%) and KSb(OH)6 (Fluka, >99%) with DI water. The

125

stock solution at 1 g/L was stored in the dark at 3°C until use. All standard solutions were diluted

126

from stock solution with 2 mM citric acid on the day of analysis. SbIII and SbV standard solutions

127

were calibrated on ICP–MS using a 100 mg/L multi-element environmental calibration standard

128

(PerkinElmer, in 5% HNO3). The detection limit of the instrument was 0.2 and 0.4 μg/L for SbIII

129

and SbV, respectively.

130

Optimization of Cartridge-Based Sb Separation

131

SPE is an extraction method common in liquid chromatography where solid and liquid

132

phases are interacted to separate chemical compounds of interest [22]. The sorbents for SPE

133

refer to the stationary phase used in the columns [22]. Anion exchange column such as 6

134

Hamiltion PRP–X100 and Synchropak Q300 have been used for Sb speciation by HPLC [3, 23].

135

However, the affinity of Sb species towards the solid phase depends on SbIII complex [3]. SbIII-

136

citrate complex has a higher affinity towards the solid phase than SbV-citrate complex whereas

137

without citrate complexing agent, the affinity of SbV is greater than SbIII [3].

138

The Sep-Pak AccellPlus QMA Plus Short cartridges used in this study was obtained from

139

Waters (WAT020545, Milford, MA). The cartridge contains a silica-based, hydrophilic, strong

140

anion exchanger with pore size of 300 Å, and sorbent particle size of 37–55 µm. The silica

141

backbone in the cartridge contains quaternary ammonium group as anion-charged site and Cl as

142

anion exchanger. Anionic compounds from the sample were exchanged to Cl .

143

–

–

To test the efficiency of the cartridge in SbV retention, 45 mL of SbIII or SbV solution was

144

passed through the cartridge at a flow rate of 10 mL/min. SbIII or SbV solution at 200 µg/L was

145

prepared in DI water at pH 4, 6, or 8. The filtrate from the cartridge was collected for Sb analysis

146

by GFAAS. The sample without cartridge filtration was used for total Sb concentration. The

147

SbIII or SbV concentrations were calculated from a difference between total Sb and Sb passed

148

through the cartridge.

149

To ensure better separation of SbV from SbIII, the discarded filtrate volume was

150

determined by applying 250 mL of 200 µg/L SbIII + SbV mixture containing 0, 25, 50, 75 and

151

100% SbIII. Sb speciation was determined every 50 mL aliquots of the filtrate. The results

152

indicated that 200 mL of filtrate needed to be discarded before collecting filtrate for effective

153

retention of SbV by the cartridge.

154

To reduce the discarded volume, cartridges were preconditioned by passing a certain

155

volume of Sb solution. 15 mL of solutions containing 1 mg/L of SbIII or SbV was used to

156

precondition the cartridge, with SbV solution showing a better result (data not shown). The 7

157

conditioning was determined by applying 15 mL of 1 mg/L SbV to the cartridge and analyzing the

158

discarded solution in 10 mL intervals after the first 50 mL of solution was discarded.

159

Preconditioning of the cartridge using SbV solution reduced the discarded filtrate volume from

160

200 to ~100 mL.

161

Discarding ~100 mL of filtrate was still too much so citric acid was used to condition the

162

cartridge. Citric acid reacts with both SbIII and SbV to form anionic complexes [23]. The

163

cartridge with citric acid media retained SbIII-citrate complex. SbV [Sb(OH)6 ] is present as

164

[Sb(OH)3(C6O7H5)]– whereas SbIII [Sb2(C4O6H2)2]2 is present as [Sb(C6O7H6)2]– in citric acid

165

solution [23]. An anion exchange cartridge adsorbs Sb from the samples according to Eq. 1:





–

–

166

RCl + [Sb(C6O7H6)2]  R–Sb(C6O7H6)2 + Cl

167

Where RCl stands for anion exchange cartridge saturated with Cl . This negatively-

168 169

Eq. 1 –

–

–

charged [Sb(C6O7H6)2] complex replaces Cl on the cartridge. Solution of 200 µg/L SbIII + SbV mixture containing 0, 25, 50, 75 and 100% SbIII in 2 mM

170

citric acid at pH 4, 6, and 8 was used to condition the cartridge. Using a cartridge that had

171

already been preconditioned with 15 mL of 1 mg/L SbV to reduce SbV + SbIII anion exchange

172

sites, 70 mL of the SbIII + SbV solution was passed through the cartridge. To determine optimal

173

pH, sample in every 10 mL aliquot was collected for Sb speciation. Using the optimal pH,

174

different SbIII proportions of 200 µg/L SbIII + SbV mixture in 2 mM citric acid were applied to the

175

column. To determine discarded volume, every 15 mL of filtrate was collected and analyzed for

176

Sb speciation.

177

Sb Speciation in Plant Tissues and Spent Growth Media

178 179

Since the As-hyperaccumulator P. vittata was efficient in Sb accumulation in the roots, it was used to test the efficiency of Sb speciation using the cartridge [24]. P. vittata was 8

180

propagated from spores in our laboratory. Uniform plants with 3–4 fronds were selected and

181

acclimatized hydroponically in 0.2X Hoagland solution (HS) with constant aeration under cool

182

and warm fluorescent lamps (the total light intensity~200 micromoles per m2) with temperature

183

of 23–28ºC and ~70% humidity for 4 weeks. After 4 weeks of acclimatization, P. vittata were

184

transferred to 0.5 mM CaCl2 (pH 6) for 24 h to desorb surface nutrients from the roots. Then P.

185

vittata was divided into 3 treatments: control (DI water), 8 mg/L SbIII, and 8 mg/L SbV, all

186

adjusted to pH 6. After 24 h, P. vittata roots were washed to desorb apoplastic Sb in an ice-cold

187

solution containing 1 mM Na2HPO4 and 0.5 mM Ca(NO3)2, pH 6 for 10 min. The plant tissues

188

were used for Sb extraction and speciation.

189

Plant tissues were extracted using a modified method by Okkenhaug et al. [25] with Sb

190

extraction efficiency of 70–92% in plants [24]. Briefly, plants were harvested and washed

191

thoroughly in DI water. They were separated into the roots and fronds, which were freeze-dried

192

for 2 d. Dry tissues were ground with liquid nitrogen to fine powder in a ceramic mortar and

193

freeze-dried for an additional 2 d. Powdered tissue samples (~50 mg) were shaken at 100 rpm

194

with 10 mL of 100 mM citric acid for 4 h and then sonicated at 42 kHz for 1 h (VWR DHT

195

Ultrasonic Cleaner B3500A). The 10 mL extracts containing 100 mM citric acid were diluted 5

196

times to 50 mL with DI water and filtered (45 m filter) before separation of Sb species. The

197

samples were further diluted 10 times with DI water. At this stage, the initial citric acid

198

concentration of 100 mM in the extracts was diluted down to 2 mM and the pH of the extracts

199

adjusted to 6.0 by use of NaOH and/or HCl.

200

In addition to plant biomass, we also tested Sb speciation in spent growth media. Spent

201

media were obtained after growing P. vittata in 0.2X HS for 8 d. To reduce interference from

202

other ions, the 0.2X HS was diluted 40 times with DI water to 0.005X HS. The diluted HS at 9

203

0.005X contained 14 µg/L H3BO3, 17µg/L MnSO4.H2O, 0.5µg/L CuSO4.5H2O, 1.1 µg/L

204

ZnSO4.7H2O, 0.5 µg/L (NH4)6MO7O24.4H2O, 44.5 nM H2SO4, 168 µg/L Na2EDTA, 139.5 µg/L

205

FeSO4, 4.7 mg/L Ca(NO3)2.4H2O, 2.6 mg/L MgSO4.7H2O, 3.3 mg/L KNO3 and 0.6 mg/L

206

NH4H2PO4 [26].

207

Cartridge-based Sb speciation in plant roots and growth media were tested. The extracted

208

plant samples and growth media were first passed through the cartridge, which were

209

preconditioned with SbV, and then determined for Sb by GFAAS. To verify the accuracy of the

210

cartridge method, the extractions from P. vittata roots treated with 8 mg/L SbIII or SbV for 24 h

211

were also analyzed for Sb speciation using HPLC–ICP–MS.

212

Results and discussion

213

SbV Retention on Cartridge in DI Water

214

The silica-based, hydrophilic, anion-exchange cartridge retains anionic SbIII or SbV

215

species via exchange with Cl , with non-negatively charged ions being passing through the

216

cartridge. However, the cartridge had a higher affinity to SbV as SbV was retained on the

217

cartridge instead of SbIII (Figure 1).

–

218

To assess impact of solution pH and discarded filtrate volumes (filtrate volume required

219

to condition the cartridge for Sb separation), solutions containing 100 µg/L SbIII or SbV at pH 4,

220

6, or 8 were applied to the cartridge and Sb retained on the cartridge was determined (Figure 1).

221

The pH of SbIII or SbV solutions had little impact on Sb retention on the cartridge. As expected,

222

SbV was rapidly retained on the cartridge whereas SbIII gradually released after 10 mL of filtrate

223

volume. The results showed that the cartridge had an ability to separate SbV from SbIII. Further

224

experiments on discarded filtrate volume were studied at pH 6.

10

225

Under pH 6, the cartridge required to discard 200 mL filtrate to achieve good SbV and

226

SbIII separation (Figure 2A). The amount of SbV retained on the cartridge after passing through

227

250 mL solution equaled 50 µg SbV. Thus the cartridge had an ability to retain at least 50 µg SbV

228

and after discarding the first 200 mL of filtrate, the cartridge effectively retained SbV while

229

passing SbIII (Figure 2A).

230

The 200 mL discard volume was too much, thus effort was to reduce the discarded filtrate

231

volume. Based on our data, preconditioning the cartridge with 15 mL of 1 mg/L SbV, it reduced

232

the discarded volume from 200 to 110 mL (Figure 2B).

233

SbIII Retention on Cartridge under Citric Acid

234

Due to its high Sb stability and strong negative charge complex, 2 mM citric acid was

235

used to condition the sample to better separate SbV from SbIII. Citric acid forms a complex with

236

either SbIII or SbV. In presence of citric acid, the SbIII-tartrate complex from potassium

237

antimonyltartrate is replaced by the SbIII-citrate complex ([Sb(C6O7H6)2] ) and SbV(Sb(OH) 6) is

238

transformed to [Sb(OH)3(C6O7H5)] ) [23]. Both forms of SbIII- and SbV-citrate complexes are

239

negatively charged and SbIII-citrate has a higher affinity to the cartridge, which allows it to be

240

retained on the cartridge. Citric acid was not only used for separation of SbIII from SbV, but it

241

also helped to stabilize SbIII in the solution [23].

–

–

–

242

Solutions containing different concentrations of SbIII and SbV at pH 4, 6 and 8 were

243

passed through the cartridge and analyzed for Sb. The pH impacted the Sb retention on the

244

cartridge in both SbIII and SbV solutions (Table 2). At pH 4 and 6, the cartridge retained 96 and

245

99% SbIII and 3 and 1% SbV whereas at pH 8, it retained 1% of either SbIII or SbV. The data

246

indicated that the cartridge accurately separated SbIII from SbV at pH 4 and 6 from the mixtures of

247

SbV and SbIII. 11

248

At pH 6, the best SbIII retention (~99%) was achieved while allowing ~99% SbV to pass

249

through the cartridge. We then determined the minimum discarded filtrate volume for accurate

250

results (Figure 2C). Sample mixtures of SbIII + SbV in 2 mM citric acid at pH 6 were passed

251

through cartridges preconditioned using SbV. Samples with >50% SbIII in the mixture showed

252

~99% retention of SbIII after discarding 45 mL of filtrate. On the other hand, with SbIII at 0 and

253

25%, 60 mL of filtrate needed to be discarded to achieve optimal retention. So 60 mL of filtrate

254

was discarded for Sb speciation in 2 mM citric acid (pH 6) using cartridges preconditioned with

255

15 mL of 1 mg/L SbV.

256

Sb Speciation in Plant Biomass and Growth Media

257

Cartridge-based Sb speciation was applied to plant biomass (P. vittata roots) and growth

258

media, including spent 0.005X HS media after growing P. vittata for 8 d and 0.005X HS spiked

259

with Sb (Table 3). SbIII was 100% retained by the cartridge, but its retention dropped slightly (4–

260

12%) in the presence of SbV in P. vittata roots (Table 3). The HS samples were spiked with

261

different concentrations of SbIII and SbV containing 2 mM citric acid at pH 6. The cartridge was

262

preconditioned with 15 mL 1 mg/L SbV and the first 60 mL of filtrate was discarded. The spent

263

HS growth media contained different nutrients and organic acids from root exudates [27],

264

however, they did not impact SbIII retention on the cartridge. Accurate recovery of SbIII from Sb

265

mixtures confirmed the ability of cartridge on Sb speciation (Table 3). For example, the cartridge

266

retained 2–3% Sb from solution containing 0% SbIII (100% SbV), 22–27% from solution

267

containing 25% SbIII, 47–50% from solution containing 50% SbIII, and 72–76% from solution

268

containing 75% SbIII (Table 3). Without SbV, the cartridge showed 100% SbIII retention from all

269

samples (Table 3).

12

270

To verify the ability of SbIII retention by the cartridge in 2 mM citric acid, Sb extraction

271

of P. vittata roots treated with 8 mg/L SbIII or SbV for 24 h were used. Using the optimized

272

parameters, the cartridge showed effective retention of SbIII in P. vittata roots extracts with 92

273

and 104% SbV recovery from SbIII and SbV treatments, respectively (Table 4). The same samples

274

were also analyzed using HPLC–ICP–MS. Sb speciation from HPLC–ICP–MS verified that no

275

SbIII from P. vittata root extractions passed through the cartridge as there was no detection of

276

SbIII (Table 4). Our data showed that cartridge-based Sb speciation can be used to separate SbIII

277

from SbV in P. vittata roots or plant growth medium.

278

Conclusions

279

This study showed the first application of a solid phase extraction using an anion

280

exchange resin for Sb speciation in plant biomass and hydroponic growth media. The

281

concentration of SbV was determined by GFAAS after retention of SbIII by the cartridge. The

282

separation of SbIII from SbV in the presence of citric acid was conducted at pH 6 with SbV-

283

preconditioned cartridges after discarding the first 60 mL of filtrate. Retention of SbIII on the

284

cartridge under citric acid condition was confirmed by HPLC–ICP–MS, with 100% SbIII

285

retention and 92–104% SbV recovery from P. vittata roots treated with SbIII and SbV. The

286

method described here is simple and efficient because of the availability of the cartridge used and

287

the avoidance of manual injection to a GFAAS and potential variations of self-pack resin

288

columns.

289

Acknowledgements

290 291

This research was supported in part by the National Natural Science Foundation of China (No. 21277070), UF/IFAS and the Royal Thai Government.

292 13

293

References

294 295 296

[1] F. Liu, X.C. Le, A. McKnight-Whitford, Y. Xia, F. Wu, E. Elswick, C. Johnson, C. Zhu, Antimony speciation and contamination of waters in the Xikuangshan antimony mining and smelting area, China, Environ. Geochem. Health, 32 (2010) 401-413.

297 298 299

[2] S.C. Wilson, P.V. Lockwood, P.M. Ashley, M. Tighe, The chemistry and behaviour of antimony in the soil environment with comparisons to arsenic: A critical review, Environ. Pollut., 158 (2010) 1169-1181.

300 301 302

[3] J. Zheng, M. Ohata, N. Furuta, Antimony speciation in environmental samples by using highperformance liquid chromatography coupled to inductively coupled plasma mass spectrometry, Anal. Sci., 16 (2000) 75-80.

303 304 305

[4] M. Krachler, J. Zheng, R. Koerner, C. Zdanowicz, D. Fisher, W. Shotyk, Increasing atmospheric antimony contamination in the northern hemisphere: snow and ice evidence from Devon Island, Arctic Canada, J. Environ. Monit., 7 (2005) 1169-1176.

306 307

[5] M. Krachler, H. Emons, J. Zheng, Speciation of antimony for the 21st century: promises and pitfalls, Trends Anal. Chem., 20 (2001) 79-90.

308 309

[6] P. Smichowski, Antimony in the environment as a global pollutant: A review on analytical methodologies for its determination in atmospheric aerosols, Talanta, 75 (2008) 2-14.

310 311

[7] R. Feng, C. Wei, S. Tu, Y. Ding, R. Wang, J. Guo, The uptake and detoxification of antimony by plants: A review, Environ. Exp. Bot., 96 (2013) 28-34.

312 313

[8] M.J. Nash, J.E. Maskall, S.J. Hill, Methodologies for determination of antimony in terrestrial environmental samples, J. Environ. Monit., 2 (2000) 97-109.

314 315 316

[9] Y. Huang, Z. Chen, W. Liu, Influence of iron plaque and cultivars on antimony uptake by and translocation in rice (Oryza sativa L.) seedlings exposed to Sb(III) or Sb(V), Plant Soil, 352 (2012) 41-49.

317 318

[10] I. Shtangeeva, E. Steinnes, S. Lierhagen, Uptake of different forms of antimony by wheat and rye seedlings, Environ. Sci. Pollut. Res., 19 (2012) 502-509.

319 320 321

[11] P.J. Craig, S.N. Forster, R.O. Jenkins, D. Miller, An analytical method for the detection of methylantimony species in environmental matrices: methylantimony levels in some UK plant material, Analyst, 124 (1999) 1243-1248.

322 323 324

[12] K. Müller, B. Daus, J. Mattusch, H.-J. Stärk, R. Wennrich, Simultaneous determination of inorganic and organic antimony species by using anion exchange phases for HPLC–ICPMS and their application to plant extracts of Pteris vittata, Talanta, 78 (2009) 820-826.

325 326

[13] H.R. Hansen, S.A. Pergantis, Analytical techniques and methods used for antimony speciation analysis in biological matrices, J. Anal. At. Spectrom., 23 (2008) 1328-1340. 14

327 328 329 330

[14] A.M. García, M.C.P. Rodríguez, J.E.S. Uria, A. Sanz-Medel, Sb(III) and Sb(V) separation and analytical speciation by a continuous tandem on-line separation device in connection with inductively coupled plasma atomic emission spectrometry, Fresen. J. Anal. Chem., 353 (1995) 128-132.

331 332 333

[15] J.M. Serafimovska, S. Arpadjan, T. Stafilov, Speciation of dissolved inorganic antimony in natural waters using liquid phase semi-microextraction combined with electrothermal atomic absorption spectrometry, Microchem. J., 99 (2011) 46-50.

334 335

[16] S. Saracoglu, M. Soylak, M. Dogan, L. Elci, Speciation of antimony using Chromosorb 102 resin as a retention medium, Anal. Sci., 19 (2003) 259-264.

336 337 338

[17] C. Yu, Q. Cai, Z.-X. Guo, Z. Yang, S.B. Khoo, Antimony speciation by inductively coupled plasma mass spectrometry using solid phase extraction cartridges, Analyst, 127 (2002) 1380-1385.

339 340 341

[18] F. Sánchez Rojas, C. Bosch Ojeda, J.M. Cano Pavón, An ion-exchange method for speciation of antimony by flow injection electrothermal atomic absorption spectrometry, Talanta, 71 (2007) 918-922.

342 343 344

[19] L. Łukaszczyk, W. Żyrnicki, Speciation analysis of Sb(III) and Sb(V) in antileishmaniotic drug using Dowex 1×4 resin from hydrochloric acid solution, J. Pharm. Biomed. Anal., 52 (2010) 747-751.

345 346 347 348

[20] C. Huang, B. Hu, Z. Jiang, Simultaneous speciation of inorganic arsenic and antimony in natural waters by dimercaptosuccinic acid modified mesoporous titanium dioxide microcolumn on-line separation and inductively coupled plasma optical emission spectrometry determination, Spectrochim. Acta B, 62 (2007) 454-460.

349 350 351

[21] D. Mendil, H. Bardak, M. Tuzen, M. Soylak, Selective speciation of inorganic antimony on tetraethylenepentamine bonded silica gel column and its determination by graphite furnace atomic absorption spectrometry, Talanta, 107 (2013) 162-166.

352 353

[22] A. Żwir-Ferenc, M. Biziuk, Solid Phase Extraction Technique – Trends, Opportunities and Applications, Pol. J. Environ. Stud., 15 (2006) 677-690.

354 355 356

[23] J. Zheng, A. Iijima, N. Furuta, Complexation effect of antimony compounds with citric acid and its application to the speciation of antimony(III) and antimony(V) using HPLCICP-MS, J. Anal. At. Spectrom., 16 (2001) 812-818.

357 358 359

[24] R. Tisarum, J.T. Lessl, X. Dong, L.M. de Oliveira, B. Rathinasabapathi, L.Q. Ma, Antimony uptake, efflux and speciation in arsenic hyperaccumulator Pteris vittata, Environ. Pollut., 186 (2014) 110-114.

360 361 362 363

[25] G. Okkenhaug, Y.-G. Zhu, J. He, X. Li, L. Luo, J. Mulder, Antimony (Sb) and arsenic (As) in Sb mining impacted paddy soil from Xikuangshan, China: differences in mechanisms controlling soil sequestration and uptake in rice, Environ. Sci. Technol., 46 (2012) 31553162. 15

364 365

[26] A.O. Fayiga, L.Q. Ma, B. Rathinasabapathi, Effects of nutrients on arsenic accumulation by arsenic hyperaccumulator Pteris vittata L., Environ. Exp. Bot., 62 (2008) 231-237.

366 367 368 369

[27] S. Tu, L. Ma, T. Luongo, Root exudates and arsenic accumulation in arsenic hyperaccumulating Pteris vittata and non-hyperaccumulating Nephrolepis exaltata, Plant Soil, 258 (2004) 9-19.

16

370 371 372

Figure Captions

373

Values are the mean of three replicates+ standard error.

374

Figure 2. Percentage of Sb retained on the cartridge in DI water at pH 6 after discarding 200 mL

375

of filtrates (A) and after cartridge being preconditioned with 15 mL of 1 mg/L SbV and discarding

376

110 mL of filtrates (B), and percentage of Sb retained on the cartridge in 2 mM citric acid at pH 6

377

after cartridge being preconditioned with 15 mL of 1 mg/L SbV and discarding 60 mL of filtrates

378

(C). Values are the mean of three replicates+standard error.

Figure 1. Percentage of SbIII and SbV retained on the cartridge in DI water at different pH.

379 380 381

17

382

Table 1. Graphite furnace temperature program for Sb determination. Step

Temperature (ºC)

Time (s)

Gas flow rate (mL/min)

1

85

5.0

300

2

95

40.0

300

3

120

10.0

300

4

700

5.0

300

5

700

1.0

300

6

700

2.0

0.0

7

2,000

0.7

0.0

8

2,000

2.0

0.0

9

2,000

2.0

300

383 384 385 386 387 388 389 390 391 392

18

393

Table 2. Sb retained on the cartridge after being preconditioned with 15 mL of 1 mg/L SbV in 2

394

mM citric acid at different pH and after discarding 70 mL of filtratea. Sb concentration(µg/L)

395

a

Sb expectedon cartridge (%)

Sb retained on cartridge (%)

SbIII 200

SbV 0

100

pH 4 96 ± 0.5

pH 6 99 ± 0.4

pH 8 1.0±1.7

150

50

75

74 ± 0.1

74 ± 0.1

41 ± 0.5

100

100

50

48 ± 1.4

50 ±0.5

27 ± 1.2

50

150

25

27 ± 1.0

23 ±0.9

11 ±0.7

0

200

0

3.0 ± 0.4

0.6± 0.2

1.0± 1.8

Values are the mean of three replicates + standard error.

396 397

19

398

Table 3. Sb retained on the cartridge from growth media (0.005X HS fresh and spent media) and

399

P. vittata roots after the cartridge being preconditioned with 15 mL of 1 mg/L SbV in 2 mM citric

400

acid and after discarding 60 mL of filtratea. Sb concentrations (µg/L) SbIII SbV

401 402

Sb expected on cartridge (%)

fresh HS media

100 200 0 100 ± 0.0 75 150 50 76 ± 0.6 50 100 100 50 ± 0.4 25 50 150 27 ± 1.7 0 0 200 3.0 ± 1.5 a Values are the mean of three replicates + standard error. b

Sb retained (%) spent HS media

P. vittata roots

100 ± 0.0 75 ± 0.2 47 ± 2.3 24 ± 2.2 2.0 ± 2.4

100 ± 0.0 72 ± 0.6 48 ± 1.0 22 ± 0.3 2.0 ± 2.1

HS = Hoagland solution.

403

20

404

Table 4. Sb speciation in P. vittata roots extraction after exposure to 8 mg/L SbIII or SbV for 24 h

405

by HPLC–ICP–MS before and after passing through the cartridge a. Treatment

406 407 408

Sb concentration (mg/kg) SbIII SbV

SbV recovery from the cartridge (%)

Control –Before 0.5 ± 0.1 µg/kg 0.2 ± 0.0 µg/kg b –After nd 0.2 ± 0.0 µg/kg III Sb –Before 289 ± 87.6 456 ± 43.2 a –After nd 421 ± 43.0 a 92 ± 2 V Sb –Before 1.4 ± 0.4 45.6 ± 3.6 b –After nd 47.6 ± 7.1 b 104 ± 5 a Values are the mean of three replicates+standard error, and columns with the same letters are not significantly different. b

nd =not detected (<0.2 μg/L SbIII).

409 410

21

III

Sb pH 4 III

Sb pH 6

411 412

Figure 1. Percentage of SbIII and SbV retained on the cartridge in DI water at different pH.

413

Values are the mean of three replicates+ standard error.

414

22

III

V

III

V

III

V

Sb :Sb =200:0 Sb :Sb =150:50 Sb :Sb =100:100 III

V

III

V

Sb :Sb =50:150 Sb :Sb =0:200

415

416

417 418

Figure 2. Percentage of Sb retained on the cartridge in DI water at pH 6 after discarding 200 mL

419

of filtrates (A) and after cartridge being preconditioned with 15 mL of 1 mg/L SbV and discarding

420

110 mL of filtrates (B), and percentage of Sb retained on the cartridge in 2 mM citric acid at pH 6

421

after cartridge being preconditioned with 15 mL of 1 mg/L SbV and discarding 60 mL of filtrates

422

(C). Values are the mean of three replicates+standard error. 23

423

SPE cartridge selectively retained SbV in DI water and SbIII in citric acid

424

425 426 427

Highlights

428



A SEP cartridge method was developed for antimony speciation in biological matrices.

429



While SbV was retained by cartridge in DI water, SbIII was retained in citric acid.

430



92-104% SbV recovery for As-hyperaccumulator P. vittata roots containing SbIII & SbV

431



The SEP cartridge procedure is effective and reproducible for Sb speciation.

432 433

24