Uranium Retention in a Bioreduced Region of an Alluvial Aquifer

Subscriber access provided by UNIV OF DURHAM

Characterization of Natural and Affected Environments

Uranium Retention in a Bioreduced Region of an Alluvial Aquifer Induced by the Influx of Dissolved Oxygen Donald Pan, Kenneth H. Williams, Mark J. Robbins, and Karrie Weber Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.8b00903 • Publication Date (Web): 11 Jul 2018 Downloaded from http://pubs.acs.org on July 12, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

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

Page 1 of 38

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

Environmental Science & Technology

Uranium Retention in a Bioreduced Region of an Alluvial Aquifer Induced by the Influx of Dissolved Oxygen Donald Pan1†, Kenneth H. Williams2, Mark Robbins2, Karrie A. Weber1,3* 1

School of Biological Sciences, University of Nebraska-Lincoln, Lincoln, NE, USA Lawrence Berkeley National Laboratory, Berkeley, CA, USA 3 Department of Earth and Atmospheric Sciences, University of Nebraska-Lincoln, Lincoln, NE, USA 2

†Present address: Department of Subsurface Geobiological Analysis and Research, Japan

Agency for Marine-Earth Science and Technology, 2-15 Natsushima-cho, Yokosuka, Kanagawa, 237-0061, Japan.

*Corresponding author Karrie A. Weber 232 Manter Hall School of Biological Sciences University of Nebraska—Lincoln Lincoln, NE 68588-0118 [email protected] phone: (402)472-2739 fax: (402)472-2083

ACS Paragon Plus Environment


32 33

ABSTRACT Reduced zones in the subsurface represent biogeochemically active hotspots enriched in

34

buried organic matter and reduced metals. Within a shallow alluvial aquifer located near Rifle,

35

CO, reduced zones control the fate and transport of uranium (U). While an influx of dissolved

36

oxygen (DO) would be expected to mobilize U, we report U immobilization. Groundwater U

37

concentrations decreased following delivery of DO (21.6 mg O2/well/hr). After 23 days of DO

38

delivery, injection of oxygenated groundwater was paused and resulted in the rebound of

39

groundwater U concentrations to pre-injection levels. When DO delivery resumed (day 51)

40

groundwater U concentrations again decreased. The injection was halted on day 82 again and

41

resulted in a rebound of groundwater U concentrations. DO delivery rate was increased to 54 mg

42

O2/well/hr (day 95) whereby groundwater U concentrations increased. Planktonic cell

43

abundance remained stable throughout the experiment, but virus-to-microbial cell ratio increased

44

1.8-3.4 fold with initial DO delivery, indicative of microbial activity in response to DO injection.

45

Together these results indicate that the redox-buffering capacity of reduced sediments can

46

prevent U mobilization, but could be overcome as delivery rate or oxidant concentration

47

increases, mobilizing U.


Page 2 of 38

Page 3 of 38

48


ABSTRACT ART

49



50 51

INTRODUCTION Subsurface sediments are chemically and physically heterogeneous due to deposition and

52

burial of soil horizons and surface derived organic material. 1, 2 These organic-rich deposits

53

represent an important facies type of subsurface sedimentary systems that generate reduced

54

zones. The high concentrations of sediment-associated organic matter in the reduced zones

55

generate “biogeochemical hotspots” distinct from the surrounding sediment matrix 3-5 and may

56

result in diagentic retention of reduced chemical species including iron (Fe(II)) and contaminants

57

such as uranium (U(IV)). 4-6 The reduction and oxidation of uranium plays a significant role in

58

controlling uranium mobility. 5, 6 The control on uranium mobility is primarily controlled by the

59

very low solubility of solid-phase U(IV) minerals. 7 As such, biostimulation of uranium-reducing

60

bacteria was conducted to immobilize uranium in subsurface systems. 8-11

61

The stability of the immobilized uranium in these reduced regions depends on

62

maintaining the immobile, reduced state rather than forming U(VI) which is highly soluble and

63

complexes with carbonate. 12 The influx of oxidants, such as dissolved oxygen (DO) or nitrate,

64

into reduced subsurface systems threatens long-term sequestration of U as U(IV)-bearing

65

minerals by oxidizing and thus dissolving U(IV)-bearing minerals rendering U mobile in

66

groundwater. 13 While the influx of nitrate is often attributed to anthropogenic inputs, 14, 15 DO

67

infiltrates into reduced sediments through advective oxic groundwater flow as well as DO

68

intrusion and transport from the capillary fringe.16, 17 Thus, the transitions between oxic and

69

anoxic conditions can result in U geochemical changes as well as stimulation of microbial

70

activity. 18-20 Organic carbon, H2, or reduced minerals in these sediments can scavenge

71

molecular oxygen and serve as a redox buffer. 21 The redox buffering capacity in reduced

72

environments is not limited to abiotic reactions. Microbial activity has also been demonstrated


Page 4 of 38

Page 5 of 38


73

to play a significant role in redox buffering in sedimentary environments scavenging oxidants. 22

74

The effect of an influx of oxidants on re-oxidation and remobilization of U in situ is not yet well

75

understood. However, the prevailing hypothesis describes that an influx of oxidants (such as DO

76

and nitrate) will oxidize reduced metals and radionuclides subsequently increasing mobility of

77

uranium. 23, 24

78

The influx of electron acceptors into reduced environments is recognized to stimulate

79

microbial activity. Previous column studies indicate that upon exposure to an oxidant, re-

80

oxidation of bioreduced U(IV) occurs 21, 25, 26 along with changes to microbial population

81

structures. 27, 28 With increased microbial metabolic activity, virus production has also been

82

observed to increase 29 which could further contribute to carbon flux. While the role of viruses

83

in subsurface systems is poorly understood, viruses have been described to play a significant role

84

in carbon cycling in marine and freshwater pelagic environments through the lysis of host cells

85

during the process of lytic reproduction and the subsequent release of available carbon and

86

nutrients. 30-32 Viruses have been recognized via direct counts, 33-35 metagenomic data, 36, 37

87

transcripts of viral proteins, 38 and electron microscopy 39 in shallow alluvial aquifers including

88

U contaminated environments such as the U.S. Department of Energy (DOE) Rifle field site.

89

Groundwater from organic rich regions of this aquifer also contained high viral loads, likely due

90

to the greater microbial activity expected in organic-rich sediments. 35 Given the abundance of

91

viruses in shallow alluvial aquifers, processes in the subsurface similar to those observed in

92

surface waters can further drive biogeochemical cycling in subsurface systems and subsequently

93

influence metal/radionuclide mobility. While viruses have been demonstrated to be abundant in

94

groundwater and subsurface environments, 29, 40, 41 the biogeochemical role they play in situ in

95

subsurface sedimentary environments remains poorly characterized.



96

In an effort to investigate the effect of naturally occurring oxidant (DO) influx on the

97

stability of U in a reduced aquifer we injected oxygen-saturated groundwater into a previously

98

bioreduced experimental plot (U.S. Department of Energy (DOE) Rifle field site). The shallow

99

unconfined alluvial aquifer contains U-bearing sediments that contribute to groundwater U

100

concentrations that exceed the maximum contaminant level (MCL).42 Biostimulation of metal

101

reducers by acetate injection has been used as a U remediation strategy in the Rifle aquifer and

102

resulted in the successful removal of dissolved U from groundwater via the biogenic

103

precipitation of reduced U(IV) minerals.8-11 A byproduct of the in situ acetate injection was the

104

accumulation of biomass and the production of artificially reduced sediments, creating a

105

bioreduced zone in the aquifer.43, 44 In addition to U(IV)-bearing minerals, the bioreduced zone

106

in this aquifer contains not only reduced soluble chemical species but also reduced minerals such

107

as mackinawite (FeS) and framboidal pyrite (FeS2).5, 8 These artificially bioreduced sediments

108

share many similar characteristics as natural buried organic-rich sediment lenses such as

109

concentration of mineral U(IV) and an abundance of iron sulfide minerals. 4, 42 Thus, this

110

bioreduced zone may be indicative of biogeochemical behavior in the wider Upper Colorado

111

River Basin region where naturally reduced sediments are believed to play a major role in the

112

fate and transport of groundwater U. 42 The final acetate injection was completed a year prior to

113

the experiment described here. In this experiment groundwater collected from an upstream well

114

was sparged with air and injected back into the aquifer. The experiment lasted 123 days from

115

August 18, 2012 to December 19, 2012. During this period, the injection was repeated twice,

116

followed by an injection at an increased rate to increase the total amount of DO injected into the

117

aquifer per unit time. Each injection was separated by pauses. Groundwater geochemical changes


Page 6 of 38

Page 7 of 38


118

were monitored over time concurrent with cell and virus abundance as an assessment of

119

microbial activity.

120 121 122

MATERIALS AND METHODS Field site. The in situ field experiment was conducted on the DOE Rifle field site located

123

480 m east of Rifle, Colorado (USA). The site hosts a shallow, unconfined alluvial aquifer

124

situated beneath a floodplain formed by a meander of the Colorado River. This aquifer is

125

composed of Holocene-age alluvium consisting of sandy gravel and gravelly sand interspersed

126

with silts and clays deposited by the river and overlying the Paleogene Wasatch Formation 9, 45,

127

which serves as a local aquitard. The bottom 3-4.6 m of the alluvial sediments are saturated, but

128

groundwater level fluctuates and can increase by as much as 1.5-1.8 m during periods of high

129

runoff. Hydraulic conductivity is estimated to be 2-10 m/day with an average alluvium porosity

130

estimated to be 15-35%. 46 The major source of groundwater in the aquifer is subsurface flow

131

and recharge from the north, flowing southwest towards the Colorado River (Fig. 1) with

132

localized spatial and temporal variations during high runoff. Additional minor contributions to

133

groundwater flow potentially come from infiltration from an on-site ditch, recharge from

134

precipitation, or hyporheic inflow of water from the Colorado River. Infiltration from the low

135

permeability Wasatch Formation is deemed insignificant. 47 The groundwater is characterized as

136

slightly reducing with DO concentrations typically less than 0.2 mg/L. 9, 46 Further details of the

137

Rifle site geology, hydrogeology and history are presented elsewhere. 8, 46

138 139

Aquifer Conditions Prior to Oxygenated Groundwater Injection. Biostimulation of the

140

indigenous metal/radionuclide reducing microbial community with acetate created a bioreduced



141

zone consisting of immobilized U within this aquifer. Generation of the bioreduced zone was

142

accomplished through acetate injection over two successive field seasons (2010-2011; 2 and 1

143

year prior respectively) into an experimental gallery (Plot C) as described elsewhere. 48, 49 Plot C

144

is oriented 190 degrees azimuthal from north and generally oriented in the direction of

145

groundwater flow at the time of the experiments (Fig. 1). There, acetate stimulation of

146

indigenous microorganisms led to the generation of a bioreduced zone characterized by elevated

147

reduced Fe concentrations and immobilized U. 49, 50

148

Prior to the beginning of the experiment, neutral pH (7.2-7.5) groundwater was suboxic

149

in all wells (DO concentration 0.05-0.11 mg/L; ORP -132 mV – -196 mV) (Table 1) consistent

150

with prior reports of aquifer conditions.

151

elevated aqueous (filtered through 0.45 µm PTFE membrane) ferrous iron (Fe(II)) and aqueous

152

sulfide concentrations (3.4-5.1 mg/L and 9.59-63.26 µg/L, respectively), further supporting

153

suboxic conditions within the aquifer (Table 1). This was in contrast to the upgradient well

154

CU01 (unreduced region of aquifer), in which low Fe(II) and sulfide concentrations (0.27 mg/L

155

and 1.92 µg/L, respectively) were observed (Table 1). Groundwater sulfate concentrations were

156

similar amongst all wells (Table 1). Groundwater nitrate concentrations were not monitored over

157

the course of the experiment but were measured below the detection limit in ca. 80% of

158

background wells in Plot C with the highest concentration measured as 0.37 mg/L. Groundwater

159

U concentrations varied between 66 µg/L to 201 µg/L, while Mn varied between 1.02 mg/L to

160

2.11 mg/L (Table 1).

9, 46, 49, 50

Wells within the bioreduced zone contained

161 162

Oxygenated Groundwater Injection. Oxygenated groundwater injectate was prepared by

163

sparging groundwater with air until saturation with atmospheric O2 and amended with a


Page 8 of 38

Page 9 of 38


164

conservative tracer, deuterium (as D2O; tank concentration δD=+240‰). The source of

165

groundwater originated from an unamended well (CU01) and was pumped directly into a storage

166

tank (18,000 L). Oxygenated groundwater was injected into the aquifer during three different

167

periods separated by pauses between each injection period. Each injection and pause period is

168

denoted by a numbered phase. In order to account for migration time, the phases in downgradient

169

wells are delineated by tracer concentrations at each monitoring well. As such, the corresponding

170

dates of each phase vary between wells and are indicated on each figure. During Phase 1 of the

171

experiment (days 0-23), oxygenated groundwater was injected into 6 injection wells (CG01-

172

CG06) at a rate of 36 mL/min per well in order to achieve a final DO concentration of 2-2.5

173

mg/L. The injectate was circulated between adjacent wells using a peristaltic pump. 8 During

174

Phase 2 (days 23-51), mineralization in the injectate line resulted in a decrease in the delivery of

175

oxygenated groundwater as indicated by tracer concentrations (Fig. S1). Later during Phase 3

176

(days 51-80), injection lines were checked periodically and cleaned to maintain flow rates. The

177

injectate migrated through the aquifer, passing through downgradient wells. During Phase 4

178

(days 75-95), injection halted again, with a brief injection from day 79-85. On day 85, the brief

179

injection was halted for tank re-filling and equipment maintenance. During Phase 5 (days 95-

180

125), the injection rate was increased by 2.5-fold (90 mL/min per well) in order to increase the

181

aquifer DO concentration to ca. 5-6 mg/L. Because O2 delivery was achieved by groundwater

182

injection, subsurface flow rates may have changed by 10 mL/hr during the slow injection phase

183

to 15 mL/hr during the fast injection phase. However, water table levels did not rise in

184

comparison to unamended controls (Fig. S9A). Due to a temperature differential, the injectate

185

did slightly increase groundwater temperature no more than +1°C (before day 70) and decrease

186

no more than -2°C (after day 70) (Fig. S8).



187

Several precipitation (rain and snow) events occurred over the course of the experiment

188

with a maximum of 13 mm of precipitation recorded (Fig. S9B). Days 5, 37, 55, and 84 were the

189

only events associated with any considerable rise in the water table (Fig. S9A).

190

Actual groundwater DO concentrations were measured along with ORP in purged wells

191

using multi-parameter sondes (detection limit 0.04 mg/L, YSI, OH). The amount of O2 delivered

192

was inferred from measured concentrations of the deuterium tracer quantified using a Liquid

193

Water Isotope Analyzer (Los Gatos Research). The concentration of injected O2 inferred for

194

each well (that is the DO concentration if the injected O2 were not consumed) was calculated by

195

the ratio of D2O measured in the well to the original D2O concentration in the tank multiplied by

196

the saturation concentration of DO at the water temperature: =

197

× ,

The rate of O2 delivery into each well was determined by the following equation: =

Δ − ∆ × Δ !

198

Where is the volume of the well, and ∆ is the decrease in DO concentration due to

199

flushing. ∆ is calculated by the equation: ∆ = " # $ %& − " #

200

where ! is the first order rate of loss of the tracer determined by measuring the rate of loss of the

201

tracer during phases 2 and 4 when injectate delivery has stopped. Aqueous U concentrations in

202

downgradient wells were corrected for the soluble U delivered by injection. Injected

203

groundwater accounted for no more than 10% of sampled groundwater at its greatest extent

204

during the slow injection phase and less than 18% during the fast injection phase.


Page 10 of 38

Page 11 of 38

205


Over the course of the experiment, groundwater samples for cation, anion, pH, and Fe(II)

206

analyses were collected from certain wells. Effects of the O2 delivery were studied in the four

207

wells (CD18, CD01, CD02, CD03) closest in proximity to where the DO influx meets the

208

bioreduced zone. These were compared to two controls: an upgradient control that had never

209

received acetate amendments (CU01) and a control that was previously biostimulated with

210

acetate but in which no O2 is delivered (CD11) (Fig. 1). Other parameters (DO, ORP,

211

temperature) were measured in situ within the wells. Samples for cell and viral enumeration

212

over the course of the experiment were also collected.

213

214

Geochemical analyses. Samples for anion, dissolved inorganic carbon (DIC), and DOC analyses

215

were filtered (0.45 µm PTFE) and stored at 4°C in no-headspace HDPE (anion) and glass vials

216

(DIC/DOC) until analysis. 51, 52 Anions were measured using ion chromatography (ICS-2100,

217

Dionex, CA) equipped with AS18 analytical columns. 53 Cations (U, Fe, Mn) were quantified

218

using inductively coupled plasma mass spectrometry (Elan DRCII ICP-MS, Perkin Elmer, Inc.)

219

following filtration (0.45 µm PTFE) and acidification (0.2 mL 12N HNO3 per 20mL sample).

220

Total cation groundwater concentrations will include submicron colloids less than 0.45 µm.

221

Ferrous iron concentrations were measured on filtered samples (0.45 µm PTFE) immediately in

222

the field upon sampling using the 1,10-Phenanthroline colorimetric method (Hach Company). 54

223

Sulfide was measured spectrophotometrically immediately upon sampling using the methylene

224

blue method (Hach Company). 55 Measurements for DIC/DOC were made on a Total Organic

225

Carbon Analyzer (TOC-VCSH; Shimadzu, Corp.). DOC was obtained as the difference between

226

total dissolved carbon and DIC. Spearman’s correlation analysis was conducted to test the

227

relationship between O2 consumed within the aquifer and change in abundance of U and other



228

aqueous cations. All statistical computation was conducted using GraphPad Prism (v5.02,

229

GraphPad Software, Inc.). Significance level for the regressions was chosen at a P-value of

230

Uranium Retention in a Bioreduced Region of an Alluvial Aquifer

Recommend Documents