Upload
others
View
5
Download
0
Embed Size (px)
Citation preview
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