Simple Method for High-Sensitivity Determination of Cosmogenic 35S

Subscriber access provided by University of Newcastle, Australia

Article

A simple method for high-sensitivity determination of cosmogenic 35S in snow and water samples collected from remote regions Mang Lin, Kun Wang, Shichang Kang, and Mark H. Thiemens Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b05066 • Publication Date (Web): 03 Mar 2017 Downloaded from http://pubs.acs.org on March 14, 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.

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

Page 1 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

1

A simple method for high-sensitivity determination of cosmogenic 35S in snow and

2

water samples collected from remote regions

3 4

Mang Lin1*, Kun Wang2,4,1, Shichang Kang2,3 and Mark H. Thiemens1

5 6

1. Department of Chemistry and Biochemistry, University of California San Diego, La Jolla,

7

California 92093, USA

8

2. State Key Laboratory of Cryosphere Sciences, Northwest Institute of Eco-Environment and

9

Resources, Chinese Academy of Sciences (CAS), Lanzhou 730000, China

10

3. CAS Center for Excellence in Tibetan Plateau Earth Sciences, Beijing 100101, China

11

4. University of CAS, Beijing 100049, China

12 13

* Email: [email protected]

14

1

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 28

15

Abstract

16

Cosmogenic 35S is useful in understanding a wide variety of chemical and physical processes in

17

the atmosphere, the hydrosphere and the cryosphere. The 87.4-day half-life and the ubiquity of

18

sulfur in natural environments renders it an ideal tracer of many phenomena. Measurements of

19

35

20

of sample (>20 L) due to their high analytical activity background and low counting efficiency.

21

Here, we present a new set of snow/water sample collecting and handling procedures for high-

22

sensitivity determination of cosmogenic

23

Laboratory experiments using diluted

24

minute) showed a

25

from the true value. Using this method, we successfully measured

26

sample collected from a glacier on the Tibetan Plateau to be 47±7 mBq/L. Based on 35S activities

27

in 9 natural samples measured in this study, a first proof-of-concept approximation for age

28

determinations and source attributions was presented. This new method will provide a powerful

29

tool in studying 35S in small volumes of snow and water samples, especially those from remote

30

but climatically important regions such as the polar regions and the Tibetan Plateau and

31

Himalayas. The measurements are particularly important as the radioactive sulfur provides an

32

actual clock of glacial melting processes. With the growing rate of glacial loss, the need for

33

measurements from remote locations becomes all the more important.

S in snow and water samples are scarce as existing analytical methods require a large volume

35

35

35

S using a low-level liquid scintillation spectrometer.

S standards (with activities of 20 L) is required. Such large volumes of water are difficult to obtain in the field,

57

especially in the remote but climatically important region such as the polar regions and the

58

Tibetean Plateau and Himalayas. For example, the Tibetean Plateau and Himalayas region,

59

which is also widely known as the third pole because it contains the largest ice store outside the

60

polar regions, provides fresh water supply to nearly 40% of the world’s populution.17-18 The

3



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 28

61

glacier retreat and snow melting in this region play an crucial role in the hydrological cycle in

62

Asia but are not well understood because the harsh environment restricts comprehensive field-

63

based measurements. Therefore, a high-sensitive analytical method for measuring

64

volumes of water sample is required to assist in such quantification.

35

S in small

65

Recently, Brothers et al.19 modified the sample preparation method for the low-level liquid

66

scintillation counting technique and suggested to convert sulfur samples to aqueous

67

Na2SO4/H2SO4. It was shown that counting aqueous Na2SO4/H2SO4 generated a high counting

68

efficiency (up to ~95%) and a low background activity (~0.9 counts per minute, cpm). This

69

optimizition made 35S measurements in small sizes of samples possible. Even though this method

70

has led to a growing number of

71

insights into the vertical and horizontal transport processes in the atmosphere,20-24 the utility of

72

35

73

preparation method for natural water. Although our previous studies successfully measured

74

in small volumes (~3 L) of glacier snow and runoff samples collected from the Tibetan Plateau

75

and Himalayas using ion exchange resins,25-26 the reliability of the sample handling method was

76

not unambigously demonstrated and reported in detail, which resulted in uncertainties in data

77

interpretation and restricted further comprehensive studies. In this study, a series of control

78

experiments were conducted to evaluate the accuracy of previous measurements25-26 and develop

79

a new, simple and economical method for collections of snow, ice and meltwater samples and

80

the highly sensitive and accurate analysis of comosgenic 35S activity. The 35S activities in natural

81

samples collected from a glacier on the Tibetan Plateau were determined using our newly

82

developed method.

35

S measurements in aerosol samples, which provide new

S in hydrologic studies remains limited, mainly due to a lack of sample collection and

83

4


35

S

Page 5 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

84


Experimental Sections

85

Ultra-Low-Level Liquid Scintillation Spectrometer and Counting Efficiency. In this study, an

86

ultra-low-level liquid scintillation spectrometer (Quantulus 1220, PerkinElmer) was used to

87

quantify cosmogenic

88

Shape Analyzer and the Pulse Amplitude Comparator) on the Quantulus 1220 followed the

89

default setting for

90

decay of 35S (EMax = 167 keV) and 14C (EMax = 156 keV).5, 19 The 35S counting window was set as

91

channel 1-450 to minimize the possible influence of

92

sample was counted for 12 times (2-h counting for each cycle) and the average was reported.

93

Potential outliers, if any, were identified by the Dixon’s Q-Test (99% confidence level) and

94

rejected. In all measurements made in this study, no outlier was identified. Using the

95

propagation-of-error method of the standard environmental counting statistics,27 the net counting

96

error (95% confidence level) can be calculated as follows:

14

35

S activity. The setting of two software selectable parameters (the Pulse

C because of the significant overlap between the energy spectra of the β-

= 1.96 × (

14

C (with peak at channel ~485). Each

+ )

97

where S and B represent sample and background counting rates, respectively, and tS and tB

98

represent elapsed counting times at which sample and background counting rates were measured,

99

respectively. In this study, the long elapsed counting time (24-h = 2-h × 12) yields a small error

100

(e.g., 0.09 cpm for a sample with a counting rate of 2.0 cpm). Given the standard environmental

101

counting statistics also recommends to divide the total counting time into several equal cycles to

102

check variability of counting rates,27 we carefully checked results for 12 counting cycles (see

103

examples in Figure 1) and found that the standard variation of 12 counting rates (e.g., 0.16 cpm

104

for a sample with an average counting rate of 2.0 cpm) was greater than the net counting error

5



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 28

105

estimated previously. Therefore, the larger error (i.e., one standard deviation for 12 counting

106

cycles) was reported in this study. This error was further propagated in subsequent arithmetic

107

calculations (e.g. corrections for the background activity and the decay time).27-28

108

Standard or sample was prepared in aqueous form (Na2SO4/H2SO4) instead of solid form 133

Ra,

228

(BaSO4) to eliminate the potential contamination of other radionuclides (e.g.

110

and their daughter products) in BaCl2 reagents.8,

111

quantitatively transfer to a 20-mL plastic scintillation vial (Fisherbrand) and mixed with 10 mL

112

of scintillation gel (Insta-Gel Plus, PerkinElmer). Plastic scintillation vials had a lower

113

background activity (~0.15 cpm) than traditional glass vials which contain

114

The mixed cocktail was shaken and vibrated vigorously prior to loading into the ultra-low-level

115

liquid scintillation spectrometer.

19

Ba,

226

109

Ra

The pure Na2SO4/H2SO4 solution was

40

K (~0.45 cpm).19

116

The background and counting efficiency for various volumes of Na2SO4/H2SO4 samples are

117

summarized in Table 1. Similar to our previous study,19 the counting efficiency of this method

118

was greater than 90% if 1 mL of aqueous solution was used. However, such small volume of

119

solution was potentially problematic in natural sample analysis because as little as ~0.1 M of

120

SO42- can impede the formation of a stable microemulsion and cause phase instability due to the

121

divalent charge and relative size of divalent anions.8, 29 In addition, if the aqueous solution was

122

not completely clean, trace amount of organics from natural samples may also lead to color

123

quench problems in such a small volume of solution.29 Table 1 shows that the counting

124

efficiency dropped down to 79.0±0.4% when 2 mL of aqueous solution were used, but the

125

further losses of counting efficiency due to using larger sample volume (5 mL) were marginal.

126

Therefore, we diluted the solution to a larger volume (5 mL) to eliminate the potential problems

127

of phase instability and color quench. It should be noted that the sample should be counted as

6


Page 7 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


128

soon as possible after the sample-gel mixture was made, because we found that the counting

129

efficiency for 5 mL of aqueous solution dropped down to 2 mmol of sulfate

152

was added to cocktail. Based on the fitted polynomial curve (Figure 2b), an upper limit of 1.5

153

mmol of sulfate is recommended for this method. If the analytical goal is to measure samples

154

with low 35S SA, e.g. groundwater and seawater, converting sulfate to BaSO4 is required and a

155

method recently developed by Uriostegui et al.8 is recommended.

156

Standard Preparation. The

35

S activities in environmental samples, which depend on the

157

sample size and storage time, are usually less than 5 dpm. A H235SO4 standard (PerkinElmer),

158

which was stored separately in a locker at the Radiation Division of the Environment, Health and

159

Safety Department (EH&S) at the University of California San Diego (UCSD), was diluted to

160

laboratory sub-standards with comparable activities (< 5 dpm) in the EH&S. These sub-standards

161

were used to develop the sample extraction and purification method for

162

sensitive, accurate and economical way. To prevent any possible cross-contamination in

163

experiments, these sub-standards were stored in a different laboratory (instead of our wet

164

chemistry laboratory).

35

S analysis in a

165

Concentrating SO42- in the Field Using Anion Exchange Resins. Previous methods using

166

Amberlite IRA 400 anion exchange resins for large volumes of samples (>20 L) require the

167

acidification of samples using concentrated hydrochloric acid (HCl) and the exchange efficiency

168

enhancement using a pump or an industrial spinner.7-8 Such complicated concentration processes

169

are conventionally performed in the laboratory. In this study, because the aqueous

170

Na2SO4/H2SO4 method has a lower background activity and higher counting efficiency (Table

171

1), only ~3 L of water is required. We utilized a Bio-Rad AG1-X8 anion exchange resin

172

(analytical grade, 100-200 mesh, chloride form) to concentration sulfate, which adsorption

173

efficiency and recovery percentage for anions in fresh water samples is greater than 98%.30 In

8


Page 9 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


174

this study, the focus is on the recovery percentage of 35S. This concentration method is relatively

175

simple and straight forward and easily done in the field. It eliminates transporting large volumes

176

of water (~3 L per sample) from the field to the laboratory. This method is especially convenient

177

and economical for collecting samples from remote regions. Before using, the resin (3-4 mL)

178

was filled into a plastic column (Bio-Rad; i.d. 0.8 cm, length 4 cm) and rinsed by 5 mL deionized

179

(DI) water (18 MΩ·cm) for three times. Sulfate was collected by gravity dripping samples (or

180

laboratory sub-standards with known

181

capacity of the AG1-X8 anion exchange resin is 1.2 meq/mL, and therefore 3-4 mL of resin can

182

handle most snow and melted water samples.

35

S activities) through the resin column. The exchange

183

Elution of SO42- from Anion Exchange Resins. Sulfate trapped in resins was eluted by gravity

184

dripping 15 mL of 1 M hydrobromic acid (HBr) through the column in 5 mL increments. To

185

quantitatively collect all samples trapped in resins, 15 mL of DI water was dripped through the

186

column in 5 mL increments and collected after elution with HBr. Positive air pressure was

187

applied to inject residual eluent/water using a clean 60-mL syringe. The 30 mL of eluent

188

containing SO42-, NO3-, Cl- and Br- was collected by a 50-mL plastic centrifuge tube. After

189

stripping, excess Br-/Cl- (~15 mmol) must be removed because such high concentrations of ions

190

cannot be handled by the scintillation gel.

191

Purification Using Ag2O and Preparation of Aqueous SO42 for Counting. Two sets of

192

experiments were performed to remove excess Br-/Cl- ions in two different ways. In this set of

193

experiments (set 1), excess Br-/Cl- were precipitated as AgBr/AgCl by slowly adding Ag2O (~2

194

grams) in successive increments of ~0.4 grams. The pH of the solution was tested by pH paper to

195

ensure that Br-/Cl- completed precipitation and the solution turned neutral (pH≈7). The

196

AgBr/AgCl slurry was removed by a disposable vacuum-driven filtration system (with 0.22 um

9



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 28

197

Millipore Express Plus Membrane, Steriflip, Millipore) and the filtrate was collected by another

198

clean 50-mL centrifuge tube. An additional ~10 mL of DI water was used to wash any trace

199

amount of sample left in the original 50-mL centrifuge tube and the filter membrane into the new

200

50-mL centrifuge tube in successive increments of ~5 mL, bringing the sample volume to ~40

201

mL. This large volume of solution must be concentrated for further processing. It is noted that

202

H2SO4 reacts with Ag2O to form Ag2SO4,31 which is minimally soluble in water (0.83 g / 100 mL

203

at 25°C). To minimize the possible losses of

204

sample, 5 mL of 1000 ppm Na2SO4 solution containing no

205

solution was evaporated and dried in a clean oven (80°C) overnight. After drying, the sample

206

was re-dissolved by adding 2 mL of DI water to the 50-mL centrifuge tube and was subjected to

207

sonication for 1 hour.

35

S during handling small volume (~1 mL) of 35

S was added to the solution. The

208

The subsequent cleaning procedure followed the protocol developed by Brothers et al.19: the 2

209

mL of solution was passed through a polyvinylpyrrolidone (PVP) column and a Ag-cartridge

210

(Dionex OnGuard II) to remove organics and any remaining Cl- (or Br-), respectively, and was

211

collected by a 15-mL centrifuge tube. An additional ~8 mL of DI water was used to wash trace

212

amount of sample left in the 50-mL centrifuge tube, the PVP column and the Ag-cartridge into

213

the 15-mL centrifuge tube in successive increments of ~2 mL. The 15-mL centrifuge tube

214

containing ~10 mL solution was put in a freeze-dryer overnight. After freeze-drying, the sulfate

215

sample (white crystalline solid) was re-dissolved again by added 1 mL of DI water and was

216

sonicated for 1 hour. If any remaining dirty residue was observed in this step, ~2 mL of

217

hydrogen peroxide (H2O2) was added to remove organics and PVP-cleaning and freeze-drying

218

should be repeated. Following this cleaning procedure, the sample was ready to transfer to a

219

plastic scintillation vial for preparing the sample-gel mixture. To quantitatively transfer, the

10


Page 11 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


220

“empty” 15-mL centrifuge tube was washed with 1 mL DI water and transferred into the

221

scintillation vial and this procedure was repeated for four times. The sample-gel mixture was

222

prepared and sent to the Quantulus 1220 following the procedure described in the previous

223

section (Ultra-Low-Level Liquid Scintillation Spectrometer and Counting Efficiency).

224

Our results show that the recovery percentage of

35

S was ~90% in this set of experiments

225

(Table 2). It is noted that the sulfur carrier (Na2SO4) is important because without adding Na2SO4

226

the recovery percentage was less than 50% (i.e., the measured 35S was more than -50% deviation

227

from the true value) (set 0) (Table 2). Since our previous pilot study25 did not notice the

228

importance of the sulfur carrier, the measured

229

processing) may be underestimated. Another concern is the cost of Ag2O (Fisher Scientific,

230

current price: ~900 US$ for 100 grams). Previous studies used 15 mL of 3 M HCl as eluent to

231

strip NO3- and SO42- samples from AG1-X8 resins, which required ~6.5 grams of Ag2O (~60

232

US$) to remove excess Cl- for each sample.30 Even though 15 mL of 1 M HBr was used as the

233

eluent in this study, which can significantly cut down the required amount of Ag2O without

234

reducing the recovery percentage of anion because the relative selectivity of Br- for AG1-X8

235

resin is more than 2 times greater than Cl- (AG1, AG MP-1 and AG2 Strong Anion Exchange

236

Resin

237

rad.com/webroot/web/pdf/lsr/literature/9114_AG_1.pdf), the cost for each sample remained high

238

(~18 US$). Therefore, we developed another set of experiments to remove Br-/Cl- without using

239

Ag2O.

35

Instruction

S activities (without the sulfur carrier during

Manual:

http://www.bio-

240

Improved Purification Method Using H2O2. In this set of experiments (set 2), 5 mL of 30%

241

H2O2 was added to eluent to oxidize HBr/HCl to Br2/Cl2. The color of solution immediately

242

changing from colorless to yellow characterizes the formation of Br2. The H2O2 can also help to

11



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 28

243

remove organics in the solution. Following adding H2O2, 5 mL of 1000 ppm Na2SO4 containing

244

no

245

vapor pressure:

Simple Method for High-Sensitivity Determination of Cosmogenic 35S

Recommend Documents