Identification of Degradation Pathways of ... - ACS Publications

Subscriber access provided by the Henry Madden Library | California State University, Fresno

Environmental Processes

Identification of degradation pathways of chlorohydrocarbons in saturated low permeability sediments using compound-specific isotope analysis Philipp Wanner, Beth L. Parker, Steven W. Chapman, Glaucia da Penha Lima, Adam Gilmore, Elizabeth Erin Mack, and Ramon Aravena Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.8b01173 • Publication Date (Web): 05 Jun 2018 Downloaded from http://pubs.acs.org on June 5, 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 33

Environmental Science & Technology

1

Identification of degradation pathways of chlorohydrocarbons in

2

saturated low permeability sediments using compound-specific isotope

3

analysis

4 5

Philipp Wanner1*, Beth L. Parker1, Steven W. Chapman1, Glaucia Lima1,2, Adam

6

Gilmore1,3, E. Erin Mack4, Ramon Aravena1,5

7 8

1

9

of Guelph, 50 Stone Road East, Guelph, Ontario, Canada N1G 2W1

G360 Institute for Groundwater Research, College of Engineering and Physical Sciences, University

10

2

11

Canada, M5S 1A4

12

3

Regional Municipality of Halton, 1151 Bronte Road, Oakville, Ontario, Canada L6M 3L1

13

4

DuPont, 974 Centre Road, Wilmington, Delaware 19805, USA

14

5

Department of Earth and Environmental Sciences, University of Waterloo, 200 University Avenue

15

West, Waterloo, Ontario, Canada N2L 3GI

16

*Corresponding author ([email protected])

Department of Civil Engineering, University of Toronto, 35 St. George Street, Toronto, Ontario,

17 18

Abstract

19

This study aims to investigate whether compound-specific carbon isotope analysis

20

(CSIA) can be used to differentiate degradation pathways of chlorohydrocarbons in

21

saturated low permeability sediments. For that purpose a site was selected, where a

22

complex mixture of chlorohydrocarbons contaminated an aquifer–aquitard system.

23

Almost 50 years after contaminant releases, high resolution, concentration, CSIA and

ACS Paragon Plus Environment

1


Page 2 of 33

24

microbial profiles were determined. The CSIA profiles showed that in the aquitard cis-

25

dichloroethene (cDCE), first considered as a degradation product of trichloroethene

26

(TCE), is produced by dichloroelimination of 1,1,2,2-tetrachloroethane (TeCA). In

27

contrast, TeCA degrades to TCE via dehydrohalogenation in the aquifer indicating that

28

the aquifer-aquitard interface separates two different degradation pathways for TeCA.

29

Moreover, the CSIA profiles showed that chloroform (CF) is degraded to

30

dichloromethane (DCM) via hydrogenolysis in the aquitard and to a minor degree

31

produced by degradation of carbon tetrachloride (CT). Several microorganisms capable

32

of degrading chlorohydrocarbons were detected in the aquitard suggesting that aquitard

33

degradation is microbially mediated. Furthermore, numerical simulations reproduced the

34

aquitard concentration and CSIA profiles well, which allowed determining degradation

35

rates for each transformation pathway. This improves the prediction of contaminant fate

36

in the aquitard and potential magnitude of impacts on the adjacent aquifer due to back-

37

diffusion.

38 39 40 41 42 43 44

Graphical Abstract/TOC Art.

45 46


2

Page 3 of 33

47


Introduction

48

Due to improper handling, disposal and accidental spills chlorohydrocarbons are

49

common subsurface contaminants.1 After the release of chlorohydrocarbons at the surface

50

as dense non-aqueous phase liquid (DNAPL), rapid vertical migration into the subsurface

51

occurs due to their high density, low interfacial tension and viscosity.1,2 During their

52

migration in the saturated zone, DNAPLs accumulate on top of low permeability

53

sediments and are slowly dissolved forming a contaminant plume downgradient of the

54

source. With time accumulated chlorohydrocarbons diffuse into the underlying low

55

permeability sediments.3-5 After complete dissolution of the accumulated DNAPL or

56

remediation of the source zone, back-diffusion occurs from the low permeability

57

sediments forming a long-term contamination source.6-11 Recent studies revealed that the

58

more reducing conditions in low permeability sediments can trigger (bio)degradation of

59

chlorohydrocarbons.12-14 These studies assessed (bio)degradation processes in saturated

60

low permeability sediments at sites, where the contamination was restricted to a few

61

chlorohydrocarbons, and where the degradation pathways were well-known. However, at

62

many

63

chlorohydrocarbons. At such sites, degradation products can originate from different

64

parent compounds and some of the compounds can degrade to different daughter

65

products

66

tetrachloroethane (TeCA) can degrade to trichloroethene (TCE) via dehydrohalogenation

67

or to a mixture of cis-dichloroethene (cDCE) and trans-dichloroethene (tDCE) by

68

dichloroelimination or it can break down to 1,1,2-trichloroethane (1,1,2-TCA) by

69

hydrogenolysis.15-17 The multiple possible degradation pathways complicate the

industrial

sites

depending

on

the

the

contamination

(bio)chemical

involves

conditions.


complex

For

mixtures

instance

of

1,1,2,2-

3


Page 4 of 33

70

quantification of reactive processes in low permeability sediments and hinder the

71

examination of whether the reactions occur biotically or abiotically. Moreover, it hampers

72

the evaluation of which compounds are released from the aquitard to the adjacent aquifer

73

due to back-diffusion after source depletion, excavation or isolation.

74

Compound-specific isotope analysis (CSIA) has been developed as an effective

75

tool to track reactive processes affecting organic compounds in aquifer systems.18-21 The

76

method makes use of the preferential cleavage of bonds between light compared to heavy

77

isotopes leading to a progressive enrichment of heavy isotopes in the parent compared to

78

the daughter compound. As opposed to aquifers, it is unclear to what extent CSIA can be

79

used in saturated low permeability sediments. A previous study has quantified for the first

80

time chlorohydrocarbon degradation in saturated low permeability sediments at a site,

81

where a controlled release field experiment was conducted.14 In that case, the

82

contamination occurred only by few chlorohydrocarbons, the degradation pathways were

83

known, and the initial conditions of the contamination were well defined (composition,

84

spill time, volume) due to the artificial contamination. This facilitated the quantification

85

of reactive processes in low permeability sediments using CSIA. However, it remains

86

unclear whether CSIA can also be applied in low permeability sediments in real-world

87

scenarios, at accidental spill sites, where the contamination occurs by more complex

88

mixtures of chlorohydrocarbons and where multiple degradation pathways are possible.

89

At such sites CSIA might be especially helpful to identify the occurring degradation

90

pathways and to quantify their transformation rates.

91

This study aims to investigate for the first time whether CSIA can be used to a)

92

differentiate degradation pathways of chlorohydrocarbons in saturated low permeability


4

Page 5 of 33


93

sediments b) quantify the transformation rate of each degradation pathway and c) assess

94

whether the degradation process (biotic vs. abiotic) of each transformation pathway can

95

be identified by complementing CSIA with microbial analysis. For that purpose, a site

96

was selected, where an aquifer-aquitard system was contaminated by a complex mixture

97

of chlorinated ethanes, ethenes and methanes resulting in formation of a plume in the

98

aquifer and contamination of the underlying aquitard by diffusion below the source area

99

and plume. Several high-resolution concentrations, CSIA and microbial profiles were

100

determined nearly 50 years after the contamination occurred using continuous cores

101

retrieved from the aquifer–aquitard system at different distances downgradient of the

102

contamination source. Concentration and CSIA profiles were simulated using numerical

103

modeling to quantify degradation rates for each degradation pathway and to estimate

104

diffusive fluxes across the aquifer-aquitard interface. The present study provides new

105

insight into degradation pathways of chlorohydrocarbons in saturated low permeability

106

sediments, which is important for evaluating future conditions and what compounds will

107

be re-released to the adjacent aquifer due to back-diffusion.

108 109

Materials and Methods

110

Study site

111

The selected site is a former manufacturing facility located near the city of

112

Florence, South Carolina, USA and was used for the production of various synthetic

113

materials from organic compounds. The site consists of an aquifer overlying a thin

114

(thickness: 0.8 – 1.2 m) Cretaceous age aquitard (Fig. 1A). The aquifer consists of a thin

115

(20 cm) sandy clay layer, overlying a 2 m thick clayey sand unit and a 40 cm thick clean


5


Page 6 of 33

116

sand layer (Fig. 1A). The mean hydraulic conductivity, determined by re-packed falling

117

head permeameter tests22, is highest in the clean sand unit (4.8 x 10-5 m/s) followed by

118

the clayey sand (5.5 x 10-6 m/s) and the sandy clay unit (5.6 x 10-9 m/s).23 The water table

119

fluctuates seasonally between about 2.5 and 4.5 meters below ground surface23, whereby

120

the groundwater shows slightly reducing conditions (Oxidation-Reduction Potential: 150

121

– 300 mV) and a low pH (4.02 – 5.79).23 The groundwater flows from southwest towards

122

northeast, but exhibits seasonal shifts of more than 30 degrees (Fig. 1B). The underlying

123

thin aquitard consists of a continuous dark grey to black stiff clay layer comprised of

124

montmorillonite (75%), illite (15%) and kaolinite (10%).24 The aquitard shows a low

125

vertical hydraulic conductivity (7.0 x 10-11 m/s) determined from flexible wall

126

permeameter tests25 and a high mean organic matter content of 1.67% (n=24; SD:

127

0.31%).23

128

The aquifer–aquitard system at the site was contaminated between 1961 and 1972

129

by a complex mixture of chlorinated ethanes, ethenes and methanes including TeCA,

130

chloroform (CF), carbon tetrachloride (CT), and a minor amount of tetrachloroethene

131

(PCE). Over that period approximately 200 liters of the organic contaminants were

132

disposed as DNAPLs into a gravel-filled drain located in the aquifer indicated as the

133

source zone (Fig. 1A and 1B).

134 135 136 137 138


6

Page 7 of 33


139 140

A

141 142 143 144 145

B 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160

Figure 1. A) Geological cross-section of the aquifer-aquitard system along groundwater flow direction including ZVI-clay treated source zone (red), multilevel wells (rectangles with horizontal strokes; CMT-5, CMT-8, CMT-10), and coring locations (rectangles with diagonal strokes; cores C-10 and C-9) and B) map of the site showing ZVI-treated source zone (red shaded), multilevel wells (blue sun crosses), coring locations (red sun crosses) and seasonal variations of groundwater flow direction (blue arrows).

161

Dissolution of the DNAPL resulted in the formation of a contaminant plume in the

162

aquifer downgradient of the source zone and with increasing time, the contaminants were

163

transported by diffusion into the underlying aquitard. In December 2007, an in-situ

164

remedial action was undertaken by mixing the source zone with zero valent iron and


7


Page 8 of 33

165

bentonite (ZVI-bentonite) to lower the hydraulic conductivity in the source zone

166

(encapsulation) and to enhance chlorohydrocarbon degradation. After remediation, the

167

extent of the chlorohydrocarbon contamination in the aquifer was monitored by 23

168

multilevel wells containing a 3-port continuous multichannel tubing (CMT)26 resulting in

169

69 ports total. The CMT wells were installed in June 2008 and March 2009 along two

170

transects perpendicular to groundwater flow direction and along one longsect capturing

171

the supposed centre line of the chlorohydrocarbon plume (Fig. 1B).

172

The disposed chlorohydrocarbons at the site (TeCA, PCE, CT, CF) can be

173

transformed by different degradation pathways depending on (bio)chemical conditions

174

(Fig. 2). Under anaerobic conditions biotic sequential hydrogenolysis of TeCA can occur,

175

whereby a chlorine atom is replaced by a hydrogen atom during each step resulting in the

176

formation of 1,1,2-TCA, chloroethane and ethane15,27 (Fig. 2). Alternatively, TeCA can

177

also be transformed biotically and abiotically by dichloroelimination, whereby two

178

chlorine atoms are removed corresponding to a two-electron transfer reaction.

179

Dichloroelimination of TeCA results in the production of tDCE and cDCE at a ratio

180

between 1.5 and 4.515-17,27-29 (Fig. 2). Finally, TeCA potentially also degrades to TCE by

181

dehydrohalogenation, which is catalyzed by OH- and is therefore, an exclusively abiotic

182

process and independent of the redox conditions15,16,27,29 (Fig. 2). Additionally, TCE can

183

also be produced by abiotic and biotic hydrogenolysis of PCE30-32 (Fig 2), which is

184

however, of minor importance as only a small quantity of PCE was disposed in the source

185

zone. Further degradation of TCE can occur by biotic and abiotic sequential

186

hydrogenolysis resulting in the production of cDCE, VC and ethane30-47 (Fig. 2).

187

Transformation of TCE to chloroacetylenes via β-elemination mediated by zero valent


8

Page 9 of 33


188

iron in the source zone, can likely be excluded as previous studies demonstrated that this

189

process is of minor importance for ZVI-bentonite mixed source zones.48-51 For CT and

190

CT hydrogenolysis can take place resulting in the formation of CF and DCM revealing

191

that CF can be a parent as well as a daughter compound at the site (Fig. 2).

192

Hydrogenolysis of CT can be mediated abiotically by reduced iron bearing minerals52,53

193

and biotically by cometabolic reduction under anoxic conditions.53,54 Degradation of CF

194

by hydrogenolysis typically occurs biotically55 except for high pH (∼12) conditions under

195

which abiotic hydrogenolysis of CF also can also take place.56

196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212

Figure 2. Possible degradation pathways of disposed chlorohydrocarbons (TeCA, PCE, CT, CF) in the source zone at the site. For TeCA hydrogenolysis, dichloroelimination and/or dehydrohalogenation can occur, while PCE, CT and CF are potentially degraded via sequential hydrogenolysis. Note that TCE can be produced by dehydrohalogenation of TeCA and hydrogenolysis of PCE and cDCE by dichloroelimination of TeCA as well as by hydrogenolysis of TCE.

213


9


Page 10 of 33

214

Groundwater sampling, core retrieval, subsampling and extraction procedures for

215

chlorohydrocarbon concentration, CSIA, microbial and organic carbon content

216

analysis

217

A detailed description of groundwater sampling, core retrieval, subsampling and

218

extraction procedures for chlorohydrocarbon concentration, CSIA, organic carbon

219

content and microbial analysis is provided in sections 1.1. – 1.3. of the Supporting

220

Information (SI). Briefly, for the present study groundwater was sampled along the plume

221

centerline from multilevel wells CMT-5, CMT-8 and CMT-10 using Geopump peristaltic

222

pumps with dedicated 0.64 cm OD diameter Teflon tubing in March 2009 at about 48

223

years after the contamination occurred (Fig. 1). At the same time, two cores (C-10 and C-

224

9) were collected from the aquifer-aquitard system adjacent to the CMT-5 and CMT-8

225

multilevel wells located at 3.5 and 21.0 m distance from the source zone (Fig. 1A). The

226

cores were retrieved using a Geoprobe 7720DT direct-push rig and the Envirocore dual

227

tube sampling system as described by Einarson et al.57 The cores spanned a vertical

228

interval of about 3.5 m from the aquifer across the interface into the aquitard (Fig. 1B).

229

After retrieval, cores were photographed, logged and subsampled with a narrow spacing

230

(2.5-5 cm) to obtain high-resolution profiles for chlorohydrocarbon concentration, CSIA,

231

organic carbon content and microorganisms. Chlorohydrocarbons were extracted from

232

the subsamples as described by Parker et al.3 and White et al.58 for concentration analysis

233

and CSIA, respectively. Subsamples for organic carbon content were dried at 40°C for 24

234

hours and subsequently pulverized. The inorganic carbon was then removed from the

235

pulverized subsamples by using a HCl solution (50% v/v) followed by another drying

236

step prior analysis. The subsamples for microbial analysis were stored in duplicate or


10

Page 11 of 33


237

triplicate in autoclaved micro centrifuge tubes (2 mL) in a -20°C freezer until DNA was

238

extracted.

239 240

Concentration, CSIA, organic carbon content and microbial analysis

241

Detailed descriptions of analytical methods are available in sections 1.4. – 1.7. of

242

the SI. In brief, chlorohydrocarbon concentrations in the groundwater and in the extracts

243

of core subsamples were measured using a gas chromatograph coupled to a mass

244

spectrometer (GC-MS). CSIA was conducted in the extracts of the core subsamples by a

245

gas chromatograph coupled to an isotope mass spectrometer (GC-IRMS). The organic

246

carbon content in the soil subsamples was analyzed using a 300 Beckmann infrared

247

analyzer59. The aim of the microbial analysis was to gain information about the presence

248

of dehalogenating bacteria and to complement the CSIA results to gain more insight into

249

the degradation processes (biotic vs. abiotic). The analysis of microorganisms in the soil

250

subsamples was conducted by extracting the DNA using the PowerSoil DNA extraction

251

kit from MoBio Laboratories (Carlsbad, California) according to the manufacturers

252

protocol60 followed by nested polymerase chain reaction (PCR) amplification. PCR was

253

applied to DNA templates targeting the 16S rDNA gene using 8f/1541r universal primers

254

for universal Bacteria61, Dhc-730f/1350r for Dehalococcoides spp.62, Dhb-477f/647r for

255

Dehalobacter sp.63 and 341f-GC/534r for denaturing gradient gel electrophoresis

256

(DGGE).64,65 The DGGE method was applied, as it is a commonly used and well-

257

accepted method to provide fingerprints of phylotypes and to monitor spatial and

258

temporal variability of bacterial communities.66-71 In this study DGGE was used to

259

determine fingerprints of a wide spectrum of dehalogenating bacterial communities such


11


Page 12 of 33

260

as Sulfurospirillum, Acetobacterium and Delftia genera. Dehalococcoides spp. and

261

Dehalobacter sp. were analyzed to an aquitard depth of 1.12 m (core C-10) and 1.14 m

262

(core C-9), respectively, while DGGE was conducted to a depth of 0.33 m (core C-10)

263

and 0.37 m (core C-9), respectively.

264 265

Numerical modelling

266

The migration of chlorohydrocarbons in the aquitard was simulated using a 1D

267

reactive transport model that was previously developed by Wanner et al.14. The aims of

268

the simulation were to a) assess which degradation pathways lead to the observed

269

aquitard concentration and CSIA profiles, b) quantify degradation rates of each

270

transformation pathway in the aquitard and c) estimate diffusive fluxes across the aquifer-

271

aquitard interface. The modelling approach is described in detail in section 3 of the SI.

272

Briefly, it was assumed that the transport in the aquitard occurs by diffusion and that

273

sorption occurs linearly and non-competitive on organic matter. The simulations were

274

run for 48 years, which corresponds to the time period between the initial contamination

275

releases and the retrieval of cores from the aquifer-aquitard system at the site. In the

276

boundary conditions the temporal chlorohydrocarbon concentration evolution at the

277

bottom of the aquifer was considered as it determines the aquifer-aquitard concentration

278

gradient and hence, whether diffusion into or out of the aquitard occurs. To simulate

279

reactive processes in the aquitard, a first order degradation rate law was used justified by

280

the low chlorohydrocarbon concentrations (0-304 µg/g). It was supposed that TeCA is

281

degraded via dichloroelimination generating cDCE and tDCE at a ratio of 3:1 (Tab. S1,

282

SI) corresponding to the average ratio detected by previous studies.15-17 Furthermore,


12

Page 13 of 33


283

sequential hydrogenolysis of TCE, CT and CF in the aquitard was considered in the 1D

284

numerical model (Tab. S1, SI). Carbon isotope fractionation during dichloroelimination

285

of TeCA and sequential hydrogenolysis of TCE, CT and CF was simulated according to

286

Van Breukelen et al.72, whereby constant δ13C values were used in the boundary

287

conditions (Tab. S3, SI). In addition to reactive processes, isotope fractionation due to

288

physical processes such as diffusion and sorption were also included in the model

289

according to Wanner et al.73 and Wanner et al.74, respectively (Tab. S2, SI). Model

290

calibration was conducted against measured concentration and CSIA profiles using the

291

concentration during the presence and after source depletion, the degradation rate and the

292

degradation-induced isotope enrichment factor as fitting parameter (see section 3.4. of SI

293

for details). To quantify the goodness of the fit between measured and modelled

294

concentration and isotope ratio profiles, the root mean squared error was used (RMSE;

295

eq. 12 in SI).

296 297

Results and Discussion

298

Chlorohydrocarbon concentration data

299

Seven chlorohydrocarbons (TeCA, TCE, cDCE, tDCE, CT, CF, and DCM) were

300

detected in the aquifer–aquitard system at the site (Fig. 3). Close to the source zone (core

301

C-10), elevated TeCA, TCE, CF and CT concentrations were detected in the aquifer. The

302

peak concentrations (TeCA: 2735 µg/g; CF: 1172 µg/g: CT: 662 µg/g; TCE: 250 µg/g)

303

were measured within 15 cm distance from the aquifer–aquitard interface (Figs. 3A and

304

3B). High TeCA, CF, CT and TCE concentrations were also detected close to the

305

aquifer–aquitard interface in the adjacent CMT-5 multilevel well being consistent with


13


Page 14 of 33

306

the concentration analysis from the core subsamples from the aquifer (Figs. S1A – B, SI).

307

The detected high TCE concentration in the aquifer close to the source zone can only be

308

explained by reactive processes, since TCE was not part of the disposed chlorinated

309

hydrocarbon mixture. TCE is likely generated by dehydrohalogenation of TeCA, as TCE

310

production by hydrogenolysis of PCE should be minimal as only minor amounts of PCE

311

were co-disposed in the source zone. At 21.0 m distance from the source zone, in core C-

312

9, maximum concentrations of TeCA and CF were two orders of magnitude lower

313

compared to the location adjacent to the source zone (Figs. 3C and 3D). An even stronger

314

concentration

decrease

was

observed

for

CT


and

TCE

compared

to

14

Page 15 of 33


Aquifer – Aquitard Interface 315 316 317 318 319 320 321 322

Aquifer – Aquitard Interface

323 324 325 326 327 328 329 330 331

Figure 3. Measured concentration profiles in the aquifer-aquitard system and modelled concentration profiles in the aquitard of TeCA (green), TCE (blue), cDCE (orange), tDCE (grey), CT (wine red), CF (blue) and DCM (turquoise) in core C-10 (A and B) and C-9 (C and D). The squares represent the measuring points, while coloured continuous lines indicate the modeled concentration profiles in the aquitard. The goodness of the fit between measured and modelled concentration profiles was quantified with the root-mean-square Error (RMSE; eq. 12, Supporting Information).


15


Page 16 of 33

332

TeCA, CF and CT, showing concentrations close to the detection limit (Figs. 3C and 3D).

333

A stronger decrease of CT and TCE concentrations compared to TeCA, CF and CT along

334

the groundwater flow path was also observed in the sampled multilevel wells (CMT-5,

335

CMT-8, CMT-10: Fig. S1, SI). In multilevel CMT-10, which is located farthest away

336

from the source zone (45 meters), TCE concentration even decreased below detection

337

limits. The generally lower chlorohydrocarbon concentration in the aquifer with

338

increasing distance from source zone can be attributed to physical processes such as

339

dispersive effects and/or to reactive processes along the groundwater path. However,

340

based on the concentration data it is challenging to determine the relative contribution of

341

reactive and physical processes to the decrease of chlorohydrocarbon concentration along

342

the flow path.

343

In contrast to the aquifer, a wider spectrum of chlorohydrocarbons was detected in

344

the aquitard (TeCA, TCE, cDCE, tDCE, CT, CF, and DCM) (Fig. 3). The aquitard

345

penetration depth of the chlorohydrocarbons was greater close to the source zone (core C-

346

10: 1.27 m) compared to the more downgradient location (core C-9: 0.50 m) (Fig. 3). The

347

chlorohydrocarbon peak concentrations occurred within 10 – 35 cm distance below the

348

aquifer-aquitard interface at both coring locations. In core C-10, close to the source zone,

349

the highest peak concentration was detected for CF (305 µg/g) followed by TeCA (265

350

µg/g) DCM (90 µg/g), TCE (39 µg/g), cDCE (12 µg/g), tDCE (4.1 µg/g) and CT (0.2

351

µg/g) (Figs. 3A and 3B). In core C-9, the peak concentrations were one order of

352

magnitude lower for TeCA (33 µg/g) and CF (25 µg/g), lower by a factor 2.5 for DCM

353

(35 µg/g) and similar for cDCE (10.5 µg/g), tDCE (3.8 µg/g) and CT (0.1 µg/g)

354

compared to the peak concentrations observed in core C-10 (Figs. 3C and 3D).


16

Page 17 of 33


355

The shape of the aquitard concentration profiles with the peak occurring below

356

the aquifer–aquitard interface and decreasing concentrations both with depth and toward

357

the interface is characteristic for a diffusion/back-diffusion scenario as observed in

358

previous studies.8 The concentration profiles suggest that chlorohydrocarbons initially

359

diffused into the aquitard during the presence of the plume in the aquifer. After reduction

360

of the plume strength, back-diffusion occurred toward the aquifer due to the reversal of

361

the concentration gradient. However, the presence of multiple daughter compounds at

362

high concentration (TCE, cDCE, tDCE, DCM) in the aquitard beside the disposed parent

363

compounds (TeCA, CF, CT) indicates that significant reactive processes occur in the

364

aquitard.73,74 The more reactive behaviour of the aquitard compared to previously

365

investigated low permeability sediments73 and can be likely explained by its up to 25

366

times higher organic matter content (Fig. S2A, SI) enhancing reductive contaminant

367

degradation. The presence of multiple degradation products in the aquitard is different

368

than what is observed in the aquifer, where TCE was the only daughter compound found

369

at high concentration. Hence, reactive processes seem to occur to a stronger degree in the

370

aquitard compared to the aquifer. However, several of the detected daughter compounds

371

in the aquitard can originate from different precursors (Fig. 2), making it difficult to

372

identify degradation pathways based on the concentration profiles. To gain more insight

373

into possible degradation pathways, high-resolution CSIA profiles were determined and

374

are discussed in the following paragraph.

375 376

Compound-specific carbon isotope ratio profiles in the aquifer-aquitard system


17


Page 18 of 33

377

In core C-10 close to the source zone, TeCA shows the most enriched δ13C value

378

in the aquifer (-24.2‰), compared to cDCE (-32.3‰), CT (-41.1‰), TCE (-42.9‰) and

379

CF (-45.2‰). The depleted δ13C value of TCE reinforces the hypothesis derived from the

380

concentration data that TCE is produced by dehydrohalogenation of TeCA as

381

hydrogenolysis of PCE would produce a more enriched δ13C TCE value (∼ -29.3‰).30

382

This can be further substantiated by estimating the δ13C value of TCE considering TCE

383

production by TeCA dehydrohalogenation only30 (eq. 1, SI). The calculated δ13C TCE

384

value (-45.7‰) is close to the measured value (-42.9‰) confirming the hypothesis that in

385

the aquifer TCE is produced by dehydrohalogenation of TeCA and not by hydrogenolysis

386

of PCE. Only little changes of δ13C values (

Identification of Degradation Pathways of ... - ACS Publications

Recommend Documents