Effect of Humic Acid & Bacterial Exudates on Sorption-Desorption Interactions of

90Sr with Brucite

Hollie Ashwortha, Liam Abrahamsen-Millsb, Nick Bryanb, Lynn Fosterc, Jonathan R.

Lloydc, Simon Kelletd, Sarah Heathe

aCentre for Radiochemistry Research, School of Chemistry, University of Manchester, M13 9PL,

bNational Nuclear Laboratory, Chadwick House, Birchwood Park, Warrington, WA3 6AE,

cSchool of Earth and Environmental Science, University of Manchester, M13 9PL;

dSellafield Ltd, Sellafield Site, Seascale, Cumbria, CA20 1PG;

eDalton Nuclear Institute, University of Manchester, M13 9PL.

Abstract

One of the nuclear fuel storage ponds at Sellafield is open to the air, and has contained a

significant inventory of corroded Magnox fuel and sludge for several decades. As a

result, some fission products have also been released into solution. 90Sr is known to

constitute a small mass of the radionuclides present in the pond, but due to its solubility

and activity, it is at risk of challenging effluent discharge limits. The sludge is

predominantly composed of brucite (Mg(OH)2), and organic molecules are known to be

present in the pond liquor with occasional algal blooms restricting visibility.

Understanding the chemical interactions of these components is important to inform

ongoing sludge retrievals and effluent management. Additionally, interactions of

radionuclides with organics at high pH will be an important consideration for the

evolution of cementitious backfilled disposal sites in the UK. Batch sorption-desorption

experiments were performed with brucite, 90Sr and natural organic matter (NOM)

(humic acid (HA) and Pseudanabaena catenata cyanobacterial growth supernatant) in

both binary and ternary systems at high pH. Ionic strength, pH and order of addition of

1

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

components were varied. 90Sr was shown not to interact strongly with the bulk brucite

surface in binary systems under pH conditions relevant to the pond. HA in both binary

and ternary systems demonstrated a strong affinity for the brucite surface. Ternary

systems containing HA demonstrated enhanced sorption of 90Sr at pH 11.5 and vice

versa, likely via formation of strontium-humate complexes regardless of the order of

addition of components. The distribution coefficients show HA sorption to be reversible

at all pH values studied, and it appeared to control 90Sr behaviour at pH 11.5. Ternary

systems containing cyanobacterial supernatant demonstrated a difference in 90Sr

behaviour when the culture had been subjected to irradiation in the first stages of its

growth.

1. Introduction

The contents of one of the nuclear fuel storage ponds at Sellafield need to be retrieved

imminently for safer storage as a matter of high priority, but the removal and

reprocessing of such a complex undefined inventory poses a unique challenge.1–3

Brucite (Mg(OH)2), derived from corrosion of the Magnox fuel cladding, is a major

component of the fine particulate sludge that has accumulated at the bottom of the pond.

Organic molecules are known to be present in the pond also, due to it being open to the

air. These organics are expected to comprise chemical degradation products of natural

organic matter (NOM), and algal or cyanobacterial organisms that cause occasional

algal blooms. These blooms can further restrict visibility, as well as complicate the

chemical interactions of the pond components. 90Sr (t1/2 = 29.1 years, beta emitter) is

known to constitute a small mass of the radionuclides present in the pond, but due to its

solubility and activity, it is at risk of challenging effluent discharge limits. Retrievals of

sludge have already begun, and it is essential to understand the chemistry of these

2

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

systems (particularly that of 90Sr) to ensure the safe management of these legacy wastes.

In addition, studies of radionuclides in alkaline environments are important for

understanding their behaviour during the evolution of near surface and deep geological

disposal sites. Disposal containers in the Low Level Waste Repository surface disposal

site close to the village of Drigg are grouted prior to disposal,4 and the UK geological

disposal facility is intended to have a cementitious backfill.5 This will create a highly

alkaline environment for a significant period of time6, and interactions with organic

material will occur in the far field range of the sites.7

The pH of the pond is usually between 11 and 11.5, and brucite typically buffers

solutions towards pH 10.8-11, the value of its pHpzc.8,9 Brucite has relatively simple

surface speciation; at a pH up to 8 the dominant surface species is expected to be

Mg(OH)2+, above which there will be gradual replacement by MgOH0 and MgO-

between pH 10-12.8 However, the concentration of MgO- species does not become

significant until pH 12 and above.8

Strontium forms the +2 ion in solution and is typically hydrated by 8-12 water

molecules.10,11 Its behaviour in solution depends mostly on pH, ionic strength, and the

sorption capacity of any minerals in the system of interest.12 Sorption of strontium has

mostly been investigated with sediments, where the predominant components have been

aluminosilicate feldspars, iron oxide and clay minerals.13–16 These studies have been

conducted over a wide pH range (between 2-10) within a few different studies,13,14,16 but

they did not explicitly identify the presence of NOM in these experiments. Sorption of

Sr is commonly observed to increase with increasing pH,12,13,15 and is described as outer-

sphere.12–15 In one study with goethite, Sr demonstrated behaviour as both an inner and

3

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

outer-sphere complex,16 whilst both inner-sphere complexation, and incorporation into

secondary silicate minerals has been determined in hyperalkaline cement leachate.17

Humic substances (HS) are a significant component of NOM and are formed through

chemical degradation of natural organic material. Humic substances form strong

complexes with cations18 and therefore are an important consideration for understanding

the chemistry of 90Sr within the pond. Complexation of strontium with humic substances

has been well researched.19–23 Several studies note increasing strontium concentrations

will increase precipitation of strontium-humate complexes, and that Mg2+ is a

competitor for Sr2+, which may substitute for it in the formation of humate

complexes.19,20,24 With increasing pH, association of strontium with humics is also

expected to increase as a result of an increased negative charge on the humic acid (HA)

molecule from increased dissociation of carboxylic acid and phenolic groups.19,25 The

apparent stability constant for strontium humates has been investigated for the pH range

4-10, and has been shown to increase with pH but level off at higher pH values.26 The

concentrations of HA investigated by Samadfam et al. are at least 10 times higher than

those in this work. Humic-metal interactions are often roughly divided into two

fractions; those that bind but are rapidly exchangeable, and those that are slowly

exchangeable.7 Slow metal ion dissociation kinetics have been demonstrated for

dissolved organic matter (DOM) at high pH in a number of studies.22,27–29 In the presence

of metal ions, humic acid can form bridged aggregates also.30,31 In ternary systems, pre-

equilibration of mineral surfaces with NOM has been investigated, and shown to

suppress the sorption of the metal in a number of studies.28,32–35 This may be a result of

organics physically blocking metal sorption sites, or it could be due to complexation of

Sr(II) by solution phase humic species.35

4

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

Visual observations of the Sellafield outdoor pond have reported the presence of green

algal blooms on the water’s surface obscuring water clarity.36 Algal samples from pond

samples at Sellafield have been found to contain a range of microbial species, including

the photosynthetic cyanobacterium Pseudanabaena catenata.37 Cyanobacteria, or blue-

green algae, are prokaryotic unicellular or filamentous ‘primary colonising’

microorganisms with no cell sub-structures such as chloroplasts or mitochondria. Their

photosynthetic pigments contain mostly chlorophyll-a (green) and phycocyanin (blue),

creating the blue-green colour from which they get their name. P. catenata is a long

narrow or ‘single trichome’ bacterium in shape38 but detailed information in the

literature regarding this particular type of cyanobacteria is extremely limited.

Investigation of these systems provides an understanding of the effect of humic acid,

and supernatants from cultures of P. catenata, on 90Sr partitioning to brucite at high pH,

extending the range of conditions and interactions under which 90Sr behaviour is known.

Beyond the Sellafield storage ponds, this information is relevant to 90Sr behaviour in

other alkaline environments, such as the near fields of cementitious disposal facilities,

both deep and near-surface. They also give an indication of Group II ion behaviour

under alkaline conditions and in the presence of natural organic matter.

2. Experimental

All materials were of analytical reagent grade. Commercial Mg(OH)2 was used as

brucite to ensure a homogeneous sorbent material (Fluka; >99% pure). Humic acid was

supplied by Sigma Aldrich. Adjustments of initial pH were made using 0.1 M NaOH

solution made with NaOH pellets provided by Fisher Scientific. Ionic strength

adjustments were made using NaClO4 stock solutions of 0.01 and 1 M. The scintillant

5

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

cocktail Scintisafe 3 was provided by Fisher Chemical. Deionised water purified in a

Millipore system was used throughout the experiments. All batch experiments were

performed in triplicate in a CO2-free atmosphere glove box to avoid the presence of

unwanted large concentrations of carbonate species in solution and to reduce the

likelihood of significant pH drift due to the length of the experiments. pH adjustments

were made before addition of the solid and the pH was not controlled thereafter.

Vivaspin ultracentrifugation membranes made from polyethersulfone (PES) with pore

cut off sizes of 3 and 100 kDa were used for ultrafiltration of solutions prior to analysis

for Mg and Sr content. All error bars represent 2σ from the mean.

Binary experiments were run until apparent steady state was achieved. This was

established as 1-2 days for the brucite-90Sr binary systems (see results section 3.1,

Figures 1 and 2a, and Figure SI 4), but between 4 and 7 days for the brucite-HA system

(see results section, Figure 3 and 4a, and supporting information Figure SI 5). For

simplicity, in the ternary systems to allow easy comparison, the two components

combined first were left to equilibrate for a standard period of 7 days prior to addition

of the third component, since all of the binary systems would be at apparent equilibrium

by that stage. Due to the length of the experiments and the subsequent three-week

period to achieve secular equilibrium of 90Sr before analysis could be conducted, some

ternary experiments were run for different lengths of time after addition of the third

component. The minimum reaction time was three weeks.

2.1. Growth of Pseudanabaena catenata Cultures

BG11 medium was used to grow cultures of the cyanobacteria Pseuanabaena catenata.

BG11 contains a combination of different stock solutions (Table SI 1), which are

combined to form the final BG11 growth medium, as shown in (Table SI 2).

6

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

2.2. Irradiation of P. catenata Cultures

The cultures were exposed to 1 Gy/min of X-radiation for19 min using a Faxitron CP-

160 Cabinet X-radiator (160 kV; 6 mA; tungsten target). All irradiation treatments were

repeated daily for 5 days giving a total absorbed dose of 95 Gy.

2.3. Analysis of Chlorophyll-a Concentration in Cyanobacterial Cultures

In order to measure the growth of P. catenata in batch cultures the concentration of

chlorophyll-a (Chl-a) was determined. 1 mL samples of P. catenata culture were

centrifuged at 14,000 g for 10 minutes, the supernatant discarded, and the cells re-

suspended in 1 mL of 70 % ethanol. These suspensions were incubated at room

temperature for 2 hours, with the resulting samples centrifuged at 14,000 rpm for 10

minutes. The supernatant was then removed and analysed using a Jenway 6700 UV/Vis

spectrophotometer. The absorbance was measured at 665 nm (Chl-a) and at 750 nm to

correct for turbidity.39 The concentration of Chl-a was then calculated using the formula

of Jespersen and Christoffersen (Equation 1).40

Chl-a (μg L-1) = (V e ∙ f ∙ A)(Vs ∙ L)

Equation 1. Determination of the concentration of Chl-a, where Ve = total volume of solvent

(ml); A = Absorbance at 665nm - Absorbance at 750nm; Vs= Total volume of sample filtered

(litres), L = Cell path length (cm); f= (1/specific extraction coefficient)*1000; where the

specific extinction coefficient for Chl-a in 96% (v/v) ethanol is 83.41 g-1 cm-1.

7

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

2.4. Analysis of Brucite Starting Material

Brucite starting material was analysed by a Bruker D8 X-Ray Diffractometer with a Cu

Kα X-Ray source to confirm its composition. Surface area analysis was performed by N2

gas adsorption using the BET method.

2.5. Analysis of Humic Acid Starting Material

Humic acid was analysed by a Thermo iCap 6300 Inductively Coupled Plasma Optical

Emission Spectroscopy (ICP-OES) to determine trace metal content.

2.6. Preparation of 90 Sr Standards

297 mL deionised water was added to 3 mL of 90Sr stock solution (1 kBq mL-1) and left

to equilibrate for at least 24 hours prior to separation into 3 standards of 10 Bq mL-1.

2.7. Preparation of Humic Acid Stock Solution

0.1 g of humic acid was dissolved in NaOH (2 mL, 0.1M) and made up to 100 mL with

deionised water. The same humic acid stock solution was used for standards and across

all batch experiments.

2.8. Binary Sorption-Desorption Experiments

Aqueous solutions were prepared containing 90Sr tracer (initial concentration, 10

Bq mL-1, 2.17 x 10-11 M), or humic acid (initial concentration, 5 ppm). For pH variance

investigations, the initial pH was adjusted to either 10.5, 11, 11.5 or 12.5. This range

was chosen to cover the range exhibited by the pond, and to assess whether behaviour at

8

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

an increased alkaline pH had any effect. NaClO4 solution was used to adjust the ionic

strength (2.5 x 10-3, 0.01 and 0.1 M) and the initial pH was subsequently adjusted to

10.5. These solutions were added to bottles containing pre-weighed brucite powder (0.1

g, 1 g L-1), where the brucite was left to settle. All experiments were performed in

triplicate, and samples were taken at regular intervals for analysis of 90Sr and humic acid

content, as appropriate.

Desorption experiments were conducted by carefully removing the supernatant from a

system (care was taken to ensure minimal disturbance of the bulk solid). The

supernatant was then refreshed with the same volume of blank pH adjusted water as that

of the original. Samples were taken at regular intervals, to determine the 90Sr and humic

acid content released back into solution.

2.9. Ternary Sorption-Desorption Experiments

Aqueous solutions were prepared containing 90Sr tracer (initial concentration,

10 Bq mL-1; 2.17 x 10-11 M), and/or humic acid (initial concentration, 5 ppm), and the

initial pH was adjusted to 10.5, 11 or 11.5. These solutions were added to bottles

containing pre-weighed brucite powder (0.1 g; 1 g L-1) in the case of brucite-humic acid

pre-equilibration, and left for 7 days prior to addition of the 90Sr (note: a ternary system

in which brucite was pre-equilibrated with 90Sr before addition of HA was not

conducted due to time constraints). For the HA-90Sr-brucite ternary experiment where

brucite was added later, HA and 90Sr were left in solution together for 7 days before the

powder was carefully added, gently mixed and allowed to settle. None of the systems

were stirred, and samples were taken at regular intervals for analysis of 90Sr, Mg2+, and

humic acid content, as appropriate. Desorption experiments were conducted in the same

9

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

way as for binary experiments. In figures concerning ternary systems, dashed lines

indicate the point at which the third component was added.

For the cyanobacterial growth medium supernatant-90Sr-brucite ternary experiments a

1 mL cyanobacterial supernatant sample was added in place of the humic acid for each

experiment. The cyanobacterial supernatant samples were taken at the peak

concentration of chlorophyll-a in a batch culture, with the initial concentrations of total

organic carbon (TOC) as 324 mg L-1 for the unirradiated samples, and 162 mg L-1 for

the irradiated samples. All other steps were the same, with only 90Sr content analysed in

solution.

2.10. Analysis of 90 Sr Content

90Sr content was analysed using liquid scintillation counting. Samples of 0.2 mL were

acidified with 2 mL HCl (2%) and left for 3 weeks to reach secular equilibrium before

addition of 5 mL scintillant cocktail. Samples were left to equilibrate with the

scintillation fluid for at least 12 hours before counting on a Wallac Quantulus 1220

Ultra Low Level liquid scintillation spectrometer.

2.11. Analysis of Humic Acid Content

The absorbance of 1 mL samples was measured in 2 mL polymethylmethacrylate

(PMMA) cuvettes at 254 nm41 using a Shimadzu UV-1800 spectrometer.

2.12. Assessment of Mg-humate precipitation

An excess of brucite powder was mixed with 100 mL deionised water adjusted to pH

10.5 with 0.1 M NaOH, then left for over 1 week to saturate the solution with Mg2+. The

solution was decanted into a separate bottle, leaving the brucite solid in the original

10

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

bottle. A 1 ppm spike of HA was added to the bottle with the saturated solution in, and

left for over 1 week. No precipitate was observed.

2.13. Ultrafiltration of Supernatant Samples

0.5 mL samples of supernatant were centrifuged for 30 minutes at 14,000 rpm using

Vivaspin ultracentrifugation membranes with filter cut-offs at 3 and 100 kDa.

Subsequent analysis of Mg content in the filtrate was carried out as detailed below.

2.14. Analysis of Magnesium Content

Samples of 0.1 mL were taken from the supernatant and diluted to 5 mL with HNO3

(2%) to <0.4 Bq mL-1. Magnesium content was analysed using a Perkin Elmer Optima

5300 Dual View ICP-AES.

2.15. Analysis of Total Organic Carbon (TOC) in Cyanobacterial Growth

S upernatant

TOC measurements of both non-irradiated and irradiated sample of growth medium

supernatant were conducted using a Shimadzu TOC-L instrument to quantify the

difference in extracellular organics produced between the two cultures. Samples were

diluted up to 10x using 18 MΩ water purified with a Millipore system.

2.16. The Distribution Coefficient

Distribution coefficients can be used to determine the binding strength of a component

to solid phases, and the reversibility of those interactions. They are defined as a ratio of

the component in solution phase to that associated with the solid phase. Distribution

coefficients for 90Sr and HA were calculated at the end of the sorption and desorption

11

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

steps in the HA-90Sr-brucite ternary system. The distribution coefficient, Kd, is defined

as;

K d,s(mL/g)= sorbed concentration (Bq/g)solution concentration (Bq/mL)

Equation 2. The sorption distribution coefficient.

As the distribution coefficient is a ratio, the sorption Kd can be calculated from

measurements of activity in solution. The desorption Kd is a little more complex as

activity remaining in the residual volume needs to be accounted for. The desorption Kd

can be calculated using Equation 5.3.

K d,d = {( [RNsorb ]M ) + ( [RNfree,s ] VR ) - ( [RNfree,d ] VT ) }

( [RNfree,d ] M)

Equation 3. The desorption coefficient calculation.

where [RN,sorb] is the counts per second (CPS) associated with the solid; [RN,free,s] is

the CPS in solution at the end of the sorption step; [RN,free,d] is the CPS in solution at

the end of the desorption step; M is the mass of solid (g); VR is the interstitial volume

remaining in preparation for desorption step (mL) and; VT is the total volume.

2.17. PHREEQC Modelling

The PHREEQC v.3.3.3 model and LLNL database was used to calculate the surface

charge density, σ, versus pH for brucite. A site density of 10 sites nm-2 was used from

Pokrovsky and Schott.8 The software was also used to calculate the speciation of Sr in

12

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

solution between pH 10-13 (see supporting information for full details of the input

code).

3. Results and Discussion

3.1. Brucite- 90 Sr Behaviour

Brucite-90Sr binary systems were investigated with varying ionic strength between 2.5 x

10-3 and 0.1 M (Figure 1) and pH between 10.5 and 12.5 (Figures 2a and SI 4). The

content of 90Sr in solution was monitored over time after addition to brucite, with no

sorption seen in either scenario. The variation of ionic strength or pH had no impact on

the behaviour of 90Sr.

Figure 1. Percentage of 90Sr in solution after addition to brucite, with variance of ionic strength

at pH 10.5.

13

298

299

300

301

302

303

304

305

306

307

308

309

310

311

The point of zero charge (pHpzc) of the brucite surface is high at pH 10.8-11.8,9 LSC data

for pH 10.5 and 11 (see Figure SI 4) displayed larger variations than those at pH 11.5

and 12.5 (Figure 2a), but no significant sorption of 90Sr was identified, which is

probably due to the proximity of these pH values to the pHpzc. At the pHpzc the surface

groups will be in dynamic equilibrium between OH2+ and O-, meaning localised sorption

sites will be changing between negative and positive, hence the surface charge is low, as

will any electrostatic attraction between the surface and the +2 cation. As a result, the

sorption is relatively weak, which suggests that the intrinsic chemical interaction

between the brucite surface and Sr(II) is relatively weak. Increasing the pH to 12.5 still

did not show a significant effect of pH. It may have been expected as the PHREEQC

model confirmed a net negative charge on the surface at this pH value (see supporting

information section 8.3 and Figure SI 2), however this was still only relatively small at

5.90 x 10-2 C m-2. Note: the pH in the 10.5 and 11 systems demonstrated buffering

towards pH 10.8-10.9 within 7 days.

14

312

313

314

315

316

317

318

319

320

321

322

323

324

325

Figure 2. 90Sr in solution, as a percentage of total initial 90Sr for a) brucite-90Sr sorption binary

system at pH 11.5 and 12.5; and b) brucite-90Sr desorption binary system at pH 11.5 and 12.5.

The desorption behaviour of 90Sr (Figure 2b) confirmed that the interaction with brucite

was negligible. The 90Sr seen in solution can be attributed to the few millilitres of

a)

b)

15

326

327

328

329

330

331

332

residual supernatant that could not be removed without disturbance of the bulk solid,

and remains constant throughout the desorption period.

3.2. Brucite-HA Behaviour

Brucite-HA binary systems were investigated with varying ionic strength (Figure 3).

The concentration of HA in solution was monitored over time after addition to brucite,

with significant sorption seen (70-80%), and steady state achieved within 4-7 days. This

time to achieve steady state was confirmed by additional experiments (Figure SI 5). The

variation in ionic strength had a negligible impact on the behaviour on HA in the range

considered.

Figure 3. Concentration of HA in solution after addition to brucite, with variance of ionic

strength.

Similarly to investigation with varying ionic strength, sorption of HA to brucite at

differing pH values appeared to achieve steady state by 7 days (Figure SI 5). In Figure

16

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

SI 5 it is clear that changing the pH does cause an initial transient effect on sorption

behaviour, although not on the final equilibrium state.

Figure 4 compares the behaviour of HA in a brucite-HA binary system (Figure 4a) with

the behaviour of HA and 90Sr in the ternary brucite-HA-90Sr system (Figure 4b), where

brucite and HA were equilibrated for 7 days before 90Sr was added. Addition of 90Sr

produced no immediate significant effect on HA. Although it appears there could be a

small difference in the sorption of HA between the binary and ternary systems, it cannot

be attributed to the presence of Sr, as the atom concentration of Sr is too small. The

decrease in 90Sr in the ternary system suggests that the sorbed fraction of HA on the

brucite surface is able to remove 90Sr from solution as ternary Sr-humate complexes.

a)

17

349

350

351

352

353

354

355

356

357

358

359

360

361

362

Figure 4. a) Percentage of HA in brucite-HA binary system at pH 11.5 and; b) percentage of

HA and 90Sr in solution in the brucite-HA-90Sr ternary system at pH 11.5, where brucite and HA

were pre-equilibrated for 7 days. Dashed line at day 7 indicates the point at which 90Sr was

added.

The brucite surface only has a small net negative charge at the pH values studied 8, but

HA will have a significant negative charge.42 Such extensive sorption of HA could be

due to interaction with the positive sites remaining on the brucite surface. Alternatively,

it could be due to ligand exchange processes,7 whereby the OH- groups are displaced by

oxygen donor sites on the humic molecule. Yet, such significant sorption may only be

relative to the concentration of HA used in these studies. Large fractions of sorbed HA

have been observed at high pH in previous studies, but increasing the concentration of

HA at constant pH significantly reduced the sorbed fraction in the previous studies.43,44

Precipitation of Mg-humates in this case was ruled out by a separate test: no precipitate

was observed after spiking a brucite saturated magnesium solution with HA, without the

bulk brucite present.

b)

18

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

3.3. HA- 90 Sr-Brucite Ternary System

A ternary system where HA and 90Sr were pre-equilibrated prior to addition of brucite

was investigated at pH 10.5 and 11.5. When considering the 90Sr in solution in the

brucite-90Sr binary system, there was no observable sorption (Figure 2a), but after

addition of brucite at day 7 in the ternary system, a greater proportion of 90Sr is removed

from solution in the pH 11.5 system (Figure 5a). This difference in behaviour between

the binary and ternary systems suggests that some 90Sr is forming Sr-humate complexes

that sorb to the brucite surface, forming ternary complexes. Removal of HA is seen at

both pH values upon brucite addition, with a greater percentage removed from the pH

10.5 system, possibly due to the greater number of positive surface sites available at that

pH, and the associated greater electrostatic attraction between the surface and the

humic. The percentage of 90Sr sorbed appears constant at pH 10.5 after the addition of

brucite, suggesting that sorption of Sr-humate complexes does not occur. The decrease

in HA concentration in solution at pH 10.5 following the addition of brucite (whilst 90Sr

remains constant) seems to be due to the sorption of free HA molecules, as reported for

the binary system.

19

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

Figure 5. Percentage of total initial HA and 90Sr remaining in solution at pH 10.5 and 11.5, for

the ternary HA-90Sr-brucite a) during the sorption; and b) desorption steps. HA and 90Sr were

pre-equilibrated for 7 days prior to addition of brucite.

a)

b)

20

397

398

399

400

401

402

403

There was a small release of HA from the brucite surface by day 7 of the desorption

period (Figure 5b). The majority of 90Sr was removed with the old supernatant. The 90Sr

seen in solution at the start of desorption can be largely attributed to the residual

supernatant. With time, a significant increase of 90Sr in solution is seen at pH 11.5,

indicating some of the Sr has dissociated from the Sr-humate complexes bound to the

surface of the brucite, and a small amount is released. Considering no 90Sr sorbs to the

brucite surface in the absence of HA (Figure 2a), HA appears to be influencing both the

sorption and desorption behaviour of the 90Sr in this ternary system, particularly at pH

11.5.

It took approximately 2 weeks for the Mg in solution to reach equilibrium (Figure SI

6a). The concentration of Mg2+ in solution decreased significantly with the increase

from pH 10.5 to 11.5, a trend previously reported by Pokrovsky and Schott.8 Upon

ultrafiltration of the supernatant at day 28 (Figure SI 6b), there was no significant

difference between Mg concentrations that passed through either the 3 or the 100 kDa

filters, suggesting that no fine colloidal Mg was produced in this system.

Distribution coefficients for HA and 90Sr at the end of the sorption and desorption

periods are shown in Figure 6. Those for HA were within error of each other at all pH

values, and indicate the sorption of HA is reversible. A t-test was performed on the

triplicate measurements at the end of sorption and desorption which confirmed that the

two data sets were not significantly different for both pH values. It is important to note

that the concentration of HA is greatly in excess of 90Sr, and therefore the proportion of

HA-90Sr ternary complexes being formed at the brucite surface is very small compared

with simple HA-brucite species. The reversible nature of HA sorption coupled with the

fact that the relative release of Sr from the solid phase is greater than that of the HA

21

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

itself at pH 11.5 (Figure 5b), suggests that some strontium humates bound via the HA as

ternary complexes dissociate, releasing some 90Sr back into solution.

The 90Sr sorption and desorption coefficients were significantly different across the pH

range. They are closer together at pH 11.5 compared with 10.5, but still different. A t-

test was performed on the triplicate measurements at the end of sorption and desorption,

which demonstrated that the two data sets were significantly different at the 98%

confidence limit for both pH values. As the 90Sr is only bound through the HA, this

difference may be due to some minor fractionation of the humic acid, meaning that the

HA remaining bound to the brucite following desorption, may be chemically different to

that removed in the supernatant at the end of sorption and/or that which desorbs during

the desorption experiment. At pH 11.5 the two distribution coefficients are much closer

in magnitude, and this may indicate that such chemical fractionation is less important at

the higher pH. At both pH values, the 90Sr distribution coefficient following desorption

is higher, showing that the Sr interaction is not reversible. It has been shown

previously22 that the interaction of metal ions with humic substances may be

irreversible, and so given that all Sr sorption in the ternary systems takes place via the

formation of Sr-HA-brucite ternary complexes, the observed irreversibility is to be

expected.

22

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

Figure 6. Distribution coefficient values for HA and 90Sr at the end of the adsorption and

desorption steps in ternary HA-90Sr-brucite system.

3.4. Cyanobacterial Growth Supernatant- 90 Sr-Brucite Ternary System

Cyanobacterial growth medium supernatant removed from both irradiated and non-

irradiated P. catenata cultures was used for ternary experiments with 90Sr and brucite.

Figure 7 compares the two systems; in both cases the growth supernatant was

equilibrated with 90Sr for 7 days prior to addition of brucite. For the non-irradiated

culture, close to 100% of 90Sr remained in solution during this period, but for the

irradiated growth culture some (up to 10%) 90Sr was removed from solution. The

addition of brucite at 7 days did not show any alteration in the content of 90Sr in either

system. Total organic carbon measurements of the two culture supernatants were 324

mg/L and 162 mg/L for the non-irradiated and irradiated cultures respectively, and as

such the concentration of extracellular products cannot be identified as the cause of

increased 90Sr removal. This suggests it is species released as a result of radiation stress

and their interactions with 90Sr that are different. Bacterial species such as these may

represent those in pond liquors and, although there was no large impact on 90Sr

23

450

451452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

behaviour, the effect of a radiation dose is noticeable and should be explored further. A

t-test was performed on the triplicate measurements for the data point at day 28, which

demonstrated that the two data sets were significantly different at the 98% confidence

limit for both pH values.

Figure 7. Comparison of 90Sr in solution in cyanobacterial growth supernatant-90Sr-brucite

ternary system, where the growth supernatant and 90Sr were equilibrated for 7 days prior to the

addition of brucite. The dashed line indicates addition of brucite after the sample was taken for

that time point.

24

468

469

470

471

472

473

474

475

476

477

478

479

480

4. Conclusions

Brucite alone does not have a significant impact on removing 90Sr from solution under

conditions relevant to the Sellafield fuel storage pond. However, brucite will strongly

sorb humic acid under these same conditions. Only 5-10% remains in solution at the end

of sorption. Therefore brucite surfaces in the presence of humic acid are liable to have

an organic coating. Evidence for the association of organic material with sludge samples

taken from the pond has been observed previously45 and the work presented here may

explain that. Changes of ionic strength or pH do not cause significant differences in the

behaviour of HA or 90Sr with brucite within the range studied (pH = 10.5 – 12.5). 90Sr

forms humate complexes, which with subsequent addition of brucite in a ternary

system, will remain in solution at pH 10.5, but sorb to brucite at pH 11.5, forming

ternary complexes. Hence, HA aids the sorption of 90Sr.

No fine colloidal Mg was present in the HA based ternary system after 28 days

suggesting that 90Sr is not seen in solution as a result of its sorption to a colloidal

fraction, but as a combination of the humate complexes and free species (Sr2+ and

SrOH+). In the context of the pond, 90Sr sorption-desorption behaviour should be

investigated with other relevant phases, such as uranium corrosion products to identify

those that are perhaps more significant in controlling its distribution.

Ternary systems containing cyanobacterial growth medium supernatant, 90Sr and brucite

suggest that irradiation of the microbial culture impacts on the behaviour of 90Sr, with a

small fraction of Sr removed from solution. To our knowledge these are the first

reported ternary systems involving bacterial cultures and provide the basis for an

important avenue of further work.

25

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

497

498

499

500

501

502

503

504

505

506

5. Conflicts of Interest

There are no conflicts to declare.

6. Acknowledgements

Sellafield Ltd are acknowledged for funding for this work. We also extend our thanks to

Paul Lythgoe for assisting with ICP-AES analyses, and Jonathan Yarwood for TOC

measurements.

26

507

508

509

510

511

512

513

514

7. References

1. A. Pitois, P. I. Ivanov, L. G. Abrahamsen, N. D. Bryan, R. J. Taylor, and H. E.

Sims, J. Environ. Monit., 2008, 10, 315–24.

2. S. A. Parry, L. O’Brien, A. S. Fellerman, C. J. Eaves, N. B. Milestone, N. D.

Bryan, and F. R. Livens, Energy Environ. Sci., 2011, 4, 1457–1464.

3. Sellafield Ltd and H. Ashworth, Sellaf. Mag., 2016, 27–29.

4. LLW Repository Ltd., Wsc-Wac-Low, 2016, 5, 1–27.

5. Radioactive Waste Management, Geological Disposal: Generic Disposal Facility

Design, 2016.

6. P. Bots, K. Morris, R. Hibberd, G. T. W. Law, J. F. W. Mosselmans, A. P.

Brown, J. Doutch, A. J. Smith, and S. Shaw, Langmuir, 2014, 30, 14396–14405.

7. A. Stockdale and N. D. Bryan, Earth-Science Rev., 2013, 121, 1–17.

8. O. S. Pokrovsky and J. Schott, Geochim. Cosmochim. Acta, 2004, 68, 31–45.

9. H. Schott, J. Pharm. Sci., 1981, 70, 486–489.

10. P. D’Angelo, V. Migliorati, F. Sessa, G. Mancini, and I. Persson, J. Phys. Chem.

B, 2016, 120, 4114–4124.

11. I. Persson, Pure Appl. Chem., 2010, 82, 1901–1917.

12. C. L. Thorpe, J. R. Lloyd, G. T. W. Law, I. T. Burke, S. Shaw, N. D. Bryan, and

K. Morris, Chem. Geol., 2012, 306–307, 114–122.

13. S. H. Wallace, S. Shaw, K. Morris, J. S. Small, A. J. Fuller, and I. T. Burke,

Appl. Geochemistry, 2012, 27, 1482–1491.

14. J. Chorover, S. Choi, P. Rotenberg, R. J. Serne, N. Rivera, C. Strepka, A.

Thompson, K. T. Mueller, and P. A. O’Day, Geochim. Cosmochim. Acta, 2008,

72, 2024–2047.

15. N. Sahai, S. Carroll, S. Roberts, and P. O’Day, J. Colloid Interface Sci., 2000,

27

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

222, 198–212.

16. R. Rahnemaie, T. Hiemstra, and W. H. van Riemsdijk, J. Colloid Interface Sci.,

2006, 297, 379–388.

17. S. H. Wallace, S. Shaw, K. Morris, J. S. Small, and I. T. Burke, Environ. Sci.

Technol., 2013, 47, 3694–3700.

18. V. Moulin, Nato Sci. Ser. IV Earth Environ. Sci., 2005, 52, 155–173.

19. A. A. Helal, D. M. Imam, and H. F. Aly, J. Radioanal. Nucl. Chem., 1998, 237,

7–10.

20. A. A. Helal, H. F. Aly, D. M. Imam, and S. M. Khalifa, J. Radioanal. Nucl.

Chem., 1998, 227, 49–53.

21. O. N. Aleksandrova, M. Schulz, and M. Matthies, Water Air Soil Pollut., 2008,

194, 287–299.

22. S. J. King, P. Warwick, A. Hall, and N. D. Bryan, Phys. Chem. Chem. Phys.,

2001, 3, 2080–2085.

23. J. Monsallier, W. Schüssler, G. Buckau, T. Rabung, J. Kim, D. Jones, R. Keepax,

and N. Bryan, Anal. Chem., 2003, 75, 3168–3174.

24. A. A. Helal, D. M. Imam, S. M. Khalifa, and H. F. Aly, J. Radioanal. Nucl.

Chem., 1998, 232, 159–161.

25. O. Celebi, A. Kilikli, and H. N. Erten, J. Hazard. Mater., 2009, 168, 695–703.

26. M. Samadfam, T. Jintoku, S. Sato, and H. Ohashi, J. Nucl. Sci. Technol., 1998,

3131, 579–583.

27. P. Reiller, Radiochim. Acta, 2005, 93, 43–55.

28. R. Buda, N. L. Banik, J. V. Kratz, and N. Trautmann, Radiochim. Acta, 2008, 96,

657–665.

29. K. Schmeide and G. Bernhard, Appl. Geochemistry, 2010, 25, 1238–1247.

30. R. Baigorri, M. Fuentes, G. González-Gaitano, and J. M. García-Mina, J. Phys.

28

540

541

542

543

544

545

546

547

548

549

550

551

552

553

554

555

556

557

558

559

560

561

562

563

564

565

Chem. B, 2007, 111, 10577–10582.

31. R. Baigorri, M. Fuentes, G. González-Gaitano, and J. M. García-Mina, Colloids

Surfaces A Physicochem. Eng. Asp., 2007, 302, 301–306.

32. P. Reiller, F. Casanova, and V. Moulin, Environ. Sci. Technol., 2005, 39, 1641–8.

33. Y. O. Takahashi, Y. O. M. Inai, S. H. A. Mbe, Y. O. M. Akide, and F. U. A.

Mbe, Geochim. Cosmochim. Acta, 1999, 63, 815–836.

34. A. Křepelová, S. Sachs, and G. Bernhard, Radiochim. Acta, 2006, 94, 825–833.

35. S. Staunton, C. Dumat, and A. Zsolnay, J. Environ. Radioact., 2002, 58, 163–

173.

36. Sellafield Ltd, sellafieldsites.com, 2015.

37. L. Foster, Unpublished work, 2017.

38. British Phycological Society and The Natural History Museum London, in The

Freshwater Algal Flora of the British Isles: An Identification, eds. D. M. John, B.

A. Whitton, and A. J. Brook, Cambridge University Press, 2002, pp. 84, 88.

39. E. G. Bellinger and D. C. Sigee, Freshwater Algae; Identification and use as

Bioindicators, Wiley-Blackwell, 2010.

40. A. M. Jespersen and K. Christoffersen, Arch. fur Hydrobiol., 1987, 109, 445–

454.

41. A. Rodrigues, A. Brito, P. Janknecht, M. F. Proença, and R. Nogueira, J.

Environ. Monit., 2009, 11, 377–82.

42. M. Jones, Adv. Colloid Interface Sci., 1998, 78, 1–48.

43. Y. Niitsu, S. Sato, H. Ohashi, Y. Sakamoto, S. Nagao, T. Ohnuki, and S.

Muraoka, J. Nucl. Mater., 1997, 248, 328–332.

44. K. Schmeide, S. Pompe, M. Bubner, K. H. Heise, G. Bernhard, and H. Nitsche,

Radiochim. Acta, 2000, 88, 723–728.

45. C. R. Gregson, D. T. Goddard, M. J. Sarsfield, and R. J. Taylor, J. Nucl. Mater.,

29

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

2011, 412, 145–156.

30

592

31

593