Biological Limitations of Dechlorination of cis-Dichloroethene during

Subscriber access provided by READING UNIV

Article

Biological limitations of dechlorination of cisdichloroethene during transport in porous media Itza Mendoza-Sanchez, Robin L Autenrieth, Thomas J McDonald, and Jeffrey A. Cunningham Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b04426 • Publication Date (Web): 13 Dec 2017 Downloaded from http://pubs.acs.org on December 14, 2017

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

Environmental Science & Technology 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 29

Environmental Science & Technology

Biological limitations of dechlorination of cis-dichloroethene during transport in porous media Itza Mendoza-Sancheza,b,*, Robin L. Autenriethb, Thomas J. McDonalda, and Jeffrey Cunninghamc a

Texas A&M University, School of Public Health, Department of Environmental and

Occupational Health, College Station, Texas 77843, USA b

Texas A&M University, Department of Civil Engineering, College Station, Texas 77843, USA

c

University of South Florida, Department of Civil and Environmental Engineering, Tampa,

Florida 33620, USA

█

AUTHOR INFORMATION

Corresponding Author *Address: Texas A&M University, School of Public Health, Department of Environmental and Occupational Health, 1266 TAMU College Station, TX 77843-1266; e-mail: [email protected]; telephone: +1-979-436-9329; fax: +1-979-436-9590.

word count in text (excluding References and Figure Captions):

4427

equivalent word count from 1 table, 2 small figures, 3 large figures:

2700

overall word count:

7127

1 ACS Paragon Plus Environment


1

ABSTRACT: We applied a mathematical model to data from experimental column studies to

2

understand the dynamics of successful and unsuccessful reductive dechlorination of chlorinated

3

ethenes in groundwater under different flow conditions. In laboratory column experiments

4

(reported previously), it was observed that complete dechlorination of cis-dichloroethene to ethene

5

was sustained at high flow velocity (0.51 m/d), but that dechlorination failed at medium or low flow

6

velocity (0.080 or 0.036 m/d). The mathematical model applied here accounts for transport of

7

chlorinated ethenes in flowing groundwater, mass transfer of chlorinated ethenes between mobile

8

groundwater and stationary biofilms, and diffusion and biodegradation within the biofilms. Monod

9

kinetics with competitive inhibition are used to describe biodegradation. Nearly all parameters

10

needed to solve the model are estimated independently from batch and non-reactive transport

11

experiments. Comparing the model predictions to the experimental results permits the evaluation of

12

three hypothesized biological limitations: insufficient supply of electron donor, decay of

13

dechlorinators’ biomass, and reduction in bacterial metabolism rates. Any of these three limitations

14

are able to adequately describe observed experimental data, but insufficient supply of electron

15

donor is the most plausible explanation for failure of dechlorination. Therefore, an important

16

conclusion of this investigation is that insufficient hydrogen production occurs if groundwater flow

17

is too slow to provide adequate flux of electron donor. Model simulations were in good agreement

18

with experimental results for both successful and unsuccessful dechlorination, suggesting the model

19

is a valid tool for describing transport and reductive dechlorination. An implication of our findings

20

is that in engineered or natural bioremediation of chloroethene-contaminated groundwater, not only

21

must the proper dechlorinating organisms be present, but also proper groundwater flow conditions

22

must be maintained or else dechlorination may fail.


Page 2 of 29

Page 3 of 29


23

INTRODUCTION

24

Bioremediation has extensively been studied for transforming perchloroethene (PCE),

25

trichloroethene (TCE), and dichloroethene (DCE) to non-toxic ethene in chloroethene-

26

contaminated groundwater. Under anaerobic conditions, such as are present in many

27

groundwater environments, biodegradation of PCE and TCE occurs through sequential reductive

28

dehalogenation in which the parent compounds are reduced to cis-dichloroethene (cis-DCE),

29

followed by vinyl chloride (VC), and finally ethene. The success of bioremediation in a

30

chloroethene-contaminated aquifer relies on the achievement of complete dechlorination to

31

ethene, because VC is the most toxic of the family of chlorinated ethenes.

32

Although complete or near-complete bioremediation of chloroethene-impacted

33

groundwater has been achieved in many instances, it has also been observed that complete

34

dechlorination to ethene is often not achieved.1–5 This sometimes leads to the accumulation of

35

partial dechlorination products, particularly cis-DCE and VC.3,6,7 Failure of complete

36

dechlorination can be caused by several factors. For instance, although a number of bacteria

37

have been found capable of partially reducing PCE or TCE, only some isolates of the species

38

Dehalococcoides mccartyi are able to fully reduce PCE and TCE to ethene.8–10 Therefore, partial

39

or incomplete dehalogenation in groundwater is sometimes attributed to the absence of D.

40

mccartyi in the aquifer.11,12 However, other factors can also lead to incomplete dechlorination.

41

For instance, it has also been found that dechlorination rates may slow down if there is

42

insufficient electron donor (hydrogen) or nutrients in the system.13–16 At any given contaminated

43

site, it is vital to select and design a remediation approach that will overcome any critical factors

44

that might limit the effectiveness of biodegradation. However, to ensure this, it is first necessary

45

to have an appropriate and sophisticated understanding of the different biological factors that

46

could possibly limit dechlorination.



47

Our research group previously reported17 the results of laboratory column experiments in

48

which near-complete dechlorination of cis-DCE was observed at high flow velocity (0.51 m/d),

49

but not at medium (0.080 m/d) or low (0.036 m/d) flow velocity. To explain this observation, we

50

hypothesized that flow rate controls the flux of electron donor through the column. Because the

51

columns were operated at fixed influent concentrations of yeast extract (the electron donor), the

52

flux of electron donor into the columns was directly proportional to the velocity. Hence, the

53

failure of low- and medium-velocity columns to sustain dechlorination could have been because

54

the flux of electron donor was inadequate for that process. However, that hypothesis has not yet

55

been tested, and therefore the cause of the observed velocity-dependent dechlorination is still

56

unknown.

57

If an appropriate mathematical model is available, the column data of Mendoza-Sanchez

58

et al.17 present an opportunity to quantitatively test different hypotheses for why the success of

59

dechlorination in the columns was velocity-dependent. Our research group previously developed

60

mathematical models and algorithms18,19 for simulating groundwater flow with dechlorination,

61

and these models may be appropriate for describing the laboratory column data. However, until

62

now, the model has not yet been applied to the column data, or indeed to any experimental data

63

set, and thus it is not yet known if the model is able to achieve its intended purpose of describing

64

transport with reductive dechlorination. Application of the mathematical model to the

65

experimental data might therefore address two research needs simultaneously:

66

verification/validation of the model, and elucidation of the critical mechanisms at work in the

67

laboratory experiments.

68

Therefore, the objectives of this paper are: (1) to apply the mathematical model of

69

Mendoza-Sanchez and Cunningham18 to the experimental data of Mendoza-Sanchez et al.17;

70

(2) to estimate all model parameters a priori via literature values or results of independent

71

experiments; (3) to validate or verify the capabilities of the model via comparison of the model 4 ACS Paragon Plus Environment

Page 4 of 29

Page 5 of 29


72

predictions to the experimental results; (4) to identify, via the comparison of model predictions

73

to experimental results, what biological mechanisms are capable of explaining the observed

74

effect of flow velocity on dechlorination; and (5) to assess which of the candidate mechanisms

75

are most plausible in the context of the experiments under consideration. Through these

76

objectives, we aim to achieve an overall goal of improving our understanding of what factors can

77

limit complete reductive dechlorination in chloroethene-contaminated groundwater.

78 79 80

MATERIALS AND METHODS Column and Batch Experiments. Details of the column experiments have been

81

presented previously17 and a schematic representation of the experimental set-up is shown in

82

Figure 1. Briefly, 60-cm glass columns were filled with homogeneous glass beads and

83

inoculated with the commercially available KB-1® culture (SiREM, Guelph, Ontario) which is

84

known to be capable of complete dechlorination of PCE to ethene.20 Synthetic anaerobic

85

groundwater containing a known and constant concentration of cis-DCE (30 µM, or 2.9 mg/L)

86

was pumped through the columns, as shown in Figure 1. The anaerobic medium contained yeast

87

extract to act as electron donor to support production of hydrogen for dechlorination. Three

88

different pore velocities, termed slow (0.036 m/d), medium (0.080 m/d), and fast (0.51 m/d),

89

were tested in duplicate. These velocities correspond to residence times of 16.7 d, 7.5 d, and 1.2

90

d, respectively, for one pore volume of fluid. The duration of the experiments were 41 d, 45 d,

91

and 47 d, respectively, corresponding to 2.5, 6, and 40 pore volumes treated for the slow-,

92

medium-, and fast-flow columns. Concentrations of cis-DCE, VC, and ethene were monitored

93

over time and space by sampling at 5 different sampling ports, spaced every 10 cm along the



94

column (see Figure 1), and analyzed by gas chromatography with flame ionization detection

95

(GC/FID).

96

In addition to these column experiments, we also conducted three non-reactive abiotic

97

control experiments with columns that were not inoculated with KB-1®, and two reactive

98

controls with columns inoculated with KB-1® but fed with solution that did not contain any cis-

99

DCE. These control experiments verified that disappearance of cis-DCE and production of VC

100

and ethene are biologically mediated.

101

Prior to conducting the column experiments, batch experiments were conducted inside an

102

anaerobic chamber (10% H2, 10% CO2, 80% N2). Batch microcosms, each containing 60 mL of

103

anaerobic medium, were inoculated with KB-1® culture and spiked with cis-DCE to obtain an

104

initial concentration of 30 µM. Abiotic controls were also conducted with batch bottles that

105

contained cis-DCE but no KB-1®, and bioreactive controls were conducted with batch bottles

106

that contained KB-1® but no cis-DCE. Periodically, samples were taken from the batch reactors

107

and analyzed via GC/FID for concentrations of cis-DCE, VC, and ethene. These batch

108

experiments were used to verify complete biologically mediated dechlorination of cis-DCE, and

109

to estimate biodegradation kinetic rate parameters. Results from the batch experiments are

110

presented in the Supporting Information (SI) and have not been presented previously.

111

Mathematical Model for Transport and Reaction. In a previous paper18 we

112

presented a mathematical model that accounts for three important processes: transport of

113

dissolved cis-DCE in the mobile fluid by advection and dispersion, mass transfer of chemical

114

species between the mobile fluid and stationary biofilms, and diffusion and non-linear Monod

115

kinetics (for both bacterial growth and biodegradation) within the biofilms. The Monod-kinetic

116

model, adopted here, was presented by Cupples et al.21 and describes reductive dehalogenation


Page 6 of 29

Page 7 of 29


117

from cis-DCE to ethene under substrate-limiting conditions and competitive inhibition between

118

cis-DCE and VC. The equations describing these processes are coupled, non-linear, partial

119

differential equations, for which efficient solution algorithms were also developed.18,19 A non-

120

dimensional version of the system of equations, with definitions of all variables, is presented in

121

the SI.

122

Estimation of Model Parameters. An objective of this work is to identify biological

123

mechanisms capable of explaining the observed effect of flow velocity on dechlorination by

124

comparing model predictions to experimental results. Application of the mathematical model

125

requires estimation of a number of parameters (e.g., flow velocity, dispersion coefficients,

126

Monod kinetic parameters). In this work, all model parameters for the column experiments

127

were estimated independently from values in literature or from separate batch and non-reactive

128

transport experiments. Therefore, estimation of parameters via “fitting” the model to

129

experimental column results was not required. In the mathematical model, there are 32 required

130

parameters for conducting the simulations. Of this total, 6 parameters were directly measured, 2

131

were estimated from batch experiments, 4 were estimated from non-reactive transport

132

experiments, 10 were estimated from the results of Cupples et al.,21 and 10 were estimated from

133

other papers in the literature. Details are provided in the following paragraphs and in the SI.

134

Biological Monod-kinetic parameters were defined by three means: (a) estimated or

135

directly measured from the experimental set-up; (b) obtained from values reported by Cupples et

136

al.;21 and/or (c) obtained from values that best reproduced separate batch reactive experimental

137

results. The last method was used to estimate only two parameters: an initial hydrogen

138

concentration, and a hydrogen production rate (i.e., rate at which H2 is produced from the

139

electron donor). To the best of our knowledge, experimental or reference values are not



140

available for these two parameters in the literature. To verify that the set of estimated Monod-

141

kinetic parameters from Cupples et al.21 are appropriate for our experimental setting, we applied

142

our mathematical biodegradation model to the batch experiments and verified that model

143

simulations matched experimental results.

144

Following estimation of the biological kinetic parameters, hydrodynamic dispersion

145

coefficients for the transport model were obtained by fitting advective-dispersive simulations to

146

experimental results of non-reactive column controls. Finally, other model parameters such as

147

mass-transfer coefficients, molecular diffusion coefficients, biofilm diffusion coefficients, and

148

biofilm thickness were estimated from values or correlations available in the literature.

149

Additional details are provided in the SI.

150

Application of the Model to Experimental Data. As discussed elsewhere,17

151

dechlorination of cis-DCE to ethene in the column experiments was successful under fast flow

152

conditions, but not under medium or slow flow conditions. To explain this behavior, we now

153

apply the mathematical model to all three sets of column data. For the fast-flow experiments, the

154

model is applied with all model parameters estimated independently as described above, i.e.,

155

there are no adjustable or “fitting” parameters employed. We refer to these model simulations as

156

“baseline” simulations.

157

For the medium- and slow-flow experiments, some modification to the model is

158

necessary to explain the failure of the dechlorination. As will be seen subsequently,

159

dechlorination began to fail after about 10 d (0.6 pore volumes) for the slow-flow columns, and

160

after about 15 d (2 pore volumes) for the medium-flow columns. A possible explanation for this

161

observation is that the electron donor originally present in the columns was either consumed or

162

flushed out of the columns within 10–15 d, and thereafter, the hydrogen production rate


Page 8 of 29

Page 9 of 29


163

decreased because hydrogen-producing bacteria lacked sufficient electron donor. Therefore, for

164

the medium- and slow-flow columns, we modify the mathematical model to reduce the hydrogen

165

production by a factor of 10 after some critical time (2 pore volumes for medium-flow columns

166

and 0.6 pore volumes for slow-flow columns). In these “hydrogen-limited” simulations, no other

167

model parameters are changed; the only change is the reduction of hydrogen production after the

168

observed critical time, presumably due to insufficient supply of electron donor.

169

Then, to explore if other possible factors could adequately describe the failure of

170

dechlorination, we also run two other sets of simulations in the medium- and slow-flow columns.

171

In “decay-limited” simulations, we increase the bacterial death rate coefficient by a factor of 10

172

after the observed critical time. In “metabolism-limited” simulations, we decrease the cis-DCE

173

biodegradation rate coefficients (Monod kinetic parameters) by a factor of 10 after the observed

174

critical time. Table 1 summarizes the scenarios considered and the associated model parameter

175

modifications for the unsuccessful dechlorination experiments. Comparison of all model

176

simulations to experimental data enables an evaluation of whether any of these possible

177

explanations can be rejected as invalid.

178 179

RESULTS AND DISCUSSION

180

The simulated and experimental breakthrough curves of cis-DCE, VC, and ethene for three

181

sampling ports (ports 2, 3, and 4, as indicated in Figure 1) of the fast-, medium-, and slow-flow

182

column experiments are depicted in Figures 2, 3, and 4, respectively. Data points represent

183

arithmetic averages of concentrations measured in two duplicate columns, and error bars indicate

184

the two actual concentrations measured in the individual columns. As seen from Figure 1, port 2

185

is the closest to the column inlet of these three ports, and therefore data at port 2 (bottom-row



186

panels in the Figures 2, 3 and 4) represent earlier travel time than the data from port 3 (middle-

187

row panels) or port 4 (top-row panels).

188

Figure 2, 3, and 4 show that dechlorination of cis-DCE to ethene was successful under

189

fast-flow conditions, but not under medium- or slow-flow conditions. To translate these results

190

to in situ conditions, the slow (3.6 cm/d) and medium (8.0 cm/d) flow rates are both in the range

191

of natural gradient conditions, and the fast flow rate (51 cm/d) might be observed under forced-

192

gradient conditions, e.g., if pumping is occurring during an active site remediation.

193

Modeling Successful Dechlorination. As seen in Figure 2, cis-DCE was efficiently

194

converted to ethene in the fast-flow column experiments. Although cis-DCE initially broke

195

through at close to its injected concentration of 30 µM (Figure 2, left-column panels), it was

196

subsequently biodegraded, accompanied by a transitory rise in the concentration of the reactive

197

intermediate VC (Figure 2, middle-column panels), and eventually stoichiometric formation of

198

ethene (Figure 2, right-column panels). To simulate experimental observations of the fast-flow

199

columns, we have applied the mathematical model as described above. All model parameters

200

have been estimated independently, not “fit” to the experimental data, and parameter values are

201

given in Table 1. As seen from Figure 2, the model successfully describes all important behavior

202

of cis-DCE, VC, and ethene, suggesting that the mathematical model developed previously18 is

203

properly accounting for the most important physical and biological processes occurring in the

204

system. In Figure 2, we have labeled the model simulation as the “baseline simulation” because

205

it is the model simulation obtained with all model parameters estimated a priori and with no

206

modifications made to the model.

207

Modeling Unsuccessful Dechlorination. As described above, successful

208

dechlorination was not sustained in the slow- and medium-flow column experiments, so we


Page 10 of 29

Page 11 of 29


209

modified the “baseline” model to account for different factors that could possibly be limiting

210

dechlorination (Table 1).

211

Figure 3 shows a comparison of the “baseline” and “hydrogen-limited” model

212

simulations to the experimental data for the slow-flow columns. The hypothesis of the

213

hydrogen-limited model simulation is that production of hydrogen decreased substantially after

214

the initial electron donor was consumed or flushed out of the column. Examination of Figure 3

215

shows that both models successfully describe the observations initially: concentrations of cis-

216

DCE and VC are low, and observed concentrations of ethene are high, indicating successful

217

complete dechlorination at early times. However, following this initial successful period, the

218

observed cis-DCE concentration rises towards its influent concentration, with some

219

transformation to VC but essentially no formation of ethene. The baseline model does not

220

describe this failure of dechlorination, which begins after about 0.6 pore volumes. However, the

221

modified (hydrogen-limited) model is able to capture all important features of the experimental

222

results. This suggests that the proposed hypothesis, i.e., decreased hydrogen production after

223

some critical time (in this case 0.6 pore volumes), is a viable hypothesis to explain unsuccessful

224

dechlorination in the slow-flow columns.

225

However, the success of the hydrogen-limited model does not rule out the possibility that

226

some other mechanism could possibly be responsible for the failure of sustained dechlorination

227

in the slow- and medium-flow columns. In particular, we have offered two alternate hypotheses,

228

the “decay-limited” and “metabolism-limited” scenarios, as described above and summarized in

229

Table 1. Figure 4 shows a comparison of the hydrogen-limited, decay-limited, and metabolism-

230

limited model simulations to the experimental data for the medium-flow columns. Examination

231

of Figure 4 shows that all three modified models are able to qualitatively describe the observed



232

behavior of the system. As with the slow-flow column (Figure 3), the dechlorination was

233

initially successful, with cis-DCE converted to ethene, but after some time, dechlorination failed,

234

and cis-DCE broke through the column at close to its influent concentration, with partial

235

conversion to VC, and only minimal conversion to ethene. The hydrogen-limited simulation and

236

the decay-limited simulation give nearly identical results, so that neither hypothesis can be

237

considered “better” based on agreement with the experimental data. The metabolism-limited

238

simulation gives somewhat different results, especially at late times, but still cannot be ruled out

239

based on agreement (or lack thereof) between simulation predictions and experimental data.

240

It is also worth noting that none of the three modified simulations have been “optimized”

241

to fit the experimental data. In each of the three modified simulations, one of the relevant rate

242

parameters was increased or decreased by a factor of 10 after some critical time, as summarized

243

in Table 1. It is possible that altering the rate by a different factor (other than 10) might lead to

244

improved agreement between model simulation and experimental results. However, throughout

245

this analysis, we have estimated all model parameters independently of the experimental results.

246

Therefore, since none of the three models have been “optimized,” we do not believe it is possible

247

to conclude that any model is better or worse than the others based on agreement with the

248

experimental data. The salient point is that all three models are able to reasonably describe the

249

observed experimental results, so to conclude which of the hypothesized rate limitations is most

250

reasonable, additional analysis is required, as described subsequently.

251

Assessment of the Mathematical Model. The mathematical model for transport and

252

biodegradation previously presented by Mendoza-Sanchez and Cunningham18 had not heretofore

253

been tested via comparison to experimental data; thus, one of the objectives of this paper was to

254

assess the model. The “baseline” model simulations of Figure 2 were performed with all model


Page 12 of 29

Page 13 of 29


255

parameters estimated independently of the column experiments, i.e., the model was not “fit” to

256

the data. Despite this, agreement between the model and the data (Figure 2) is excellent, i.e., the

257

model is able to describe all key behaviors of the cis-DCE, VC, and ethene. We conclude that

258

the proposed model is “valid” for describing one-dimensional transport and reaction of

259

chlorinated ethenes under the proper conditions. It may therefore be useful for such applications

260

as predicting the fractional conversion of cis-DCE to VC and ethene at a contaminated site,

261

predicting breakthrough curves or concentration profiles at a contaminated site, or determining

262

how much travel time is required to achieve a specified remediation goal. The model could also

263

be useful for identifying particular causes of biological limitations in other laboratory or field

264

experiments where dechlorination was observed not to proceed to completion.

265

However, we here note five important areas where the model could potentially be

266

improved in the future. (1) The model accounts for cis-DCE, VC, and ethene, but not the

267

commonly encountered perchloroethene or trichloroethene; the model could be extended to

268

include these parent compounds. (2) The model could be extended to account for different D.

269

mccartyi subpopulations, which has been recommended for environments under electron donor

270

limiting conditions.22 (3) The model is a “biofilm” model, i.e., it accounts for mass transfer

271

between the mobile fluid and stationary biofilms, and it assumes that the biotransformation

272

reactions occur within the biofilms. Cunningham and Mendoza-Sanchez have previously

273

considered23 the conditions under which such a “biofilm” model can be replaced by a simpler

274

model that treats biotransformation reactions as occurring in the aqueous (mobile) phase rather

275

than within stationary biofilms. Modifying the model in this fashion would presumably enable

276

the model to run much faster, and under certain conditions23 would not result in any loss in

277

accuracy. (4) The model is currently capable only of describing one-dimensional transport, such



278

as transport within laboratory columns or sufficiently homogeneous field sites. Most field sites

279

would require the model to be modified to include heterogeneous and multi-dimensional flow

280

fields. (5) As discussed previously, the “baseline” model is appropriate only when sustained

281

biodegradation is successful, and when dechlorination fails, the baseline parameters of the model

282

must be modified. Although the model results with the modified parameters were observed to be

283

successful (Figures 3 and 4), the model does not have a built-in method for a priori estimation of

284

when dechlorination will succeed or fail, nor how to modify the model parameters properly for

285

conditions of unsuccessful dechlorination. Developing these methods would likely require

286

additional experimental work as well as modifications to the model.

287

In particular, modeling efforts could be improved by measuring concentrations of

288

hydrogen gas and dechlorinating bacteria over space and time. This would help in a variety of

289

ways. First, it would provide additional measurements to which model predictions could be

290

compared, thereby further testing or verifying if the model is properly representing actual

291

conditions. Second, quantification of the reactive biomass in the system would be useful to

292

distinguish among different possible biological limitations, i.e., hydrogen, metabolism, or decay

293

limitation. Third, the model currently uses a very rough estimate for the hydrogen production

294

rate, and refinement of that estimate would be useful for ensuring the accuracy of the model.

295

Assessment of Proposed Mechanisms for Unsuccessful Dechlorination. As seen

296

from Figure 4, experimental data for unsuccessful dechlorination can be adequately described

297

with any of three proposed rate-limited model simulations (Table 1). However, we conclude that

298

the most plausible hypothesis is that hydrogen production is limited by insufficient supply of

299

electron donor, as originally proposed by Mendoza-Sanchez et al.17 This conclusion is based on

300

two factors.


Page 14 of 29

Page 15 of 29

301


First, in addition to concentration histories such as shown in Figures 2–4, we can examine

302

concentration profiles, i.e., graphs of the measured (or predicted) concentrations as functions of

303

position within the columns. Figure 5 shows simulated and experimental concentration profiles

304

for cis-DCE, VC, and ethene in the slow-flow columns after 1.4 pore volumes and after 2.2 pore

305

volumes. Agreement between the hydrogen-limited simulation and the experimental data is

306

excellent for all three chemicals. As discussed above, none of the modified model simulations

307

have been optimized, and therefore we can not truly compare model performance among the

308

three candidate models; nevertheless, the excellent performance of the hydrogen-limited

309

simulation lends confidence that the model is properly capturing the chemistry and biology of the

310

experiments.

311

Second, the hydrogen-limited simulation appears to have a firmer mechanistic foundation

312

than the other two proposed rate-limited scenarios. It seems plausible that, under low-flow

313

conditions, the original supply of electron donor may have been depleted within a certain period

314

of time, after which hydrogen production decreased due to insufficient supply of electron donor.

315

In contrast, it seems unlikely that the bacterial population(s) would begin to die at a faster rate

316

after a certain period of time, or that the bacterial population’s metabolic processes would slow

317

after a certain period of time. In the absence of a clear rationale for how or why these could

318

occur, we adopt as our working hypothesis that insufficient groundwater flow can limit the flux

319

of electron donor and thereby limit the production of hydrogen needed for sustained

320

dechlorination.

321

Implications for Groundwater Remediation. Based on the preceding analysis, we

322

conclude that reductive dechlorination fails if the supply of electron donor is insufficient to

323

sustain production of hydrogen. Notably, however, in the column experiments of Mendoza-



324

Sanchez et al.,17 the provided concentration of electron donor (yeast extract) was identical in

325

fast-, medium-, and slow-flow columns. The supply of electron donor at a contaminated field site

326

is not controlled only by, say, the concentration in a solution that is injected or amended at the

327

site; it is also controlled by the velocity of the groundwater, which is likely to be distributed

328

heterogeneously at the site. This implies that dechlorination activity may also be distributed

329

heterogeneously, even if the necessary organisms are present throughout the site. Clean-up of a

330

contaminated site must therefore include consideration of the effect of heterogeneous velocity

331

distribution on the flux of delivered chemicals.

332 333

█

334

Supporting Information

335

Mathematical model description; details on the estimation of transport and biodegradation

336

kinetic parameters; definition of non-dimensional numbers used in the mathematical model;

337

definition of parameters in the non-dimensional numbers; values of Monod kinetic parameters

338

for the baseline case that best describe microbial dechlorination behavior; values of transport

339

and kinetic parameters for the baseline simulation; values of non-dimensional numbers in the

340

three different simulations; simulated and experimental biodegradation results for different

341

cases of initial hydrogen concentration; simulated and experimental biodegradation results for

342

different cases of hydrogen production rates; simulated and experimental breakthrough curves

343

of cis-DCE for non-reactive control columns compared to advective-dispersive simulations.

344

This material is available free of charge via the Internet (PDF).

ASSOCIATED CONTENT

345


Page 16 of 29

Page 17 of 29


346

█

347

Corresponding Author

348

*I. Mendoza-Sanchez. Address: Texas A&M University, School of Public Health, Department of

349

Environmental and Occupational Health, 1266 TAMU, College Station, TX 77843-1266. E-mail:

350

[email protected]. Tel.: +1-979-436-9329. Fax: +1-979-436-9590.

351

Notes

352

The authors declare no competing financial interest.

AUTHOR INFORMATION

353 354

█

355

Financial support for this research was provided to Itza Mendoza-Sanchez from the following

356

sources: Consejo Nacional de Ciencia y Tecnología (CONACYT), México; and the United

357

States Geological Survey (USGS) through the Texas Water Resources Institute (TWRI). Any

358

opinions, findings, conclusions, or recommendations expressed in this paper are those of the

359

authors and do not necessarily reflect the views of CONACYT, USGS, or TWRI.

ACKNOWLEDGEMENTS



360

█

361

(1)

REFERENCES Hendrickson, E. R.; Payne, J. A.; Young, R. M.; Starr, M. G.; Perry, M. P.; Fahnestock,

362

S.; Ellis, D. E.; Ebersole, R. C. Molecular analysis of Dehalococcoides 16s ribosomal

363

DNA from chloroethene-contaminated sites throughout North America and Europe. Appl.

364

Environ. Microbiol. 2002, 68 (2), 485–495.

365

(2)

Liang, Y.; Cook, L. J.; Mattes, T. E. Temporal abundance and activity trends of vinyl

366

chloride (VC) degrading bacteria in a dilute VC plume at Naval Air Station Oceana.

367

Environ. Sci. Pollut. Res. 2017, 24, 13760–13774.

368

(3)

Atashgahi, S.; Lu, Y.; Zheng, Y.; Saccenti, E.; Suarez-Diez, M.; Ramiro-Garcia, J.;

369

Eisenmann, H.; Elsner, M.; Stams, A. J. M.; Springael, D.; et al. Geochemical and

370

microbial community determinants of reductive dechlorination at a site biostimulated with

371

glycerol. Environ. Microbiol. 2017, 19 (3), 968–981.

372

(4)

Adetutu, E. M.; Gundry, T. D.; Patil, S. S.; Golneshin, A.; Adigun, J.; Bhaskarla, V.;

373

Aleer, S.; Shahsavari, E.; Ross, E.; Ball, A. S. Exploiting the intrinsic microbial

374

degradative potential for field-based in situ dechlorination of trichloroethene contaminated

375

groundwater. J. Hazard. Mater. 2015, 300, 48–57.

376

(5)

Mundle, S. O. C.; Johnson, T.; Lacrampe-Couloume, G.; Perez-de-Mora, A.; Duhamel,

377

M.; Edwards, E. A.; Mcmaster, M. L.; Cox, E.; Revesz, K.; Lollar, B. S. Monitoring

378

biodegradation of ethene and bioremediation of chlorinated ethenes at a contaminated site

379

using compound-specific isotope analysis (CSIA). Environ. Sci. Technol 2012, 46, 1731–

380

1738.


Page 18 of 29

Page 19 of 29

381


(6)

Mattes, T. E.; Jin, Y. O.; Livermore, J.; Pearl, M.; Liu, X. Abundance and activity of vinyl

382

chloride (VC)-oxidizing bacteria in a dilute groundwater VC plume biostimulated with

383

oxygen and ethene. Environ. Biotechnol. 2015, 99, 9267–9276.

384

(7)

Zhang, S.; Hou, Z.; Du, X.; Li, D.; Lu, X. Assessment of biostimulation and

385

bioaugmentation for removing chlorinated volatile organic compounds from groundwater

386

at a former manufacture plant. Biodegradation 2016, 27, 223–236.

387

(8)

Maymó-Gatell, X.; Chien, Y.; Gossett, J. M.; Zinder, S. H. Isolation of a bacterium that

388

reductively dechlorinates tetrachloroethene to ethene. Science. 1997, 276 (5318), 1568–

389

1571.

390

(9)

He, J.; Ritalahti, K. M.; Aiello, M. R.; Löffler, F. E. Complete detoxification of vinyl

391

chloride by an anaerobic enrichment culture and identification of the reductively

392

dechlorinating population as a Dehalococcoides species. Appl. Environ. Microbiol. 2003,

393

69 (2), 996–1003.

394

(10)

Löffler, F. E.; Yan, J.; Ritalahti, K. M.; Adrian, L.; Edwards, E.A., Konstantinidis, K.;

395

Müller, J.A.; Fullerton, H.; Zinder, S.H.; Spormann,A.M. Dehalococcoides mccartyi gen.

396

nov., sp. nov., obligately organohalide-respiring anaerobic bacteria relevant to halogen

397

cycling and bioremediation, belong to a novel bacterial class, Dehalococcoidia classis

398

nov., order Dehalococcoidales ord. nov. and family Dehalococcoidaceae fam. nov.,

399

within the phylum Chloroflexi. Int. J. Syst. Evol. Microbiol. 2013, 63, 625-635.

400

(11)

Schaefer, C. E.; Condee, C. W.; Vainberg, S.; Steffan, R. J. Bioaugmentation for

401

chlorinated ethenes using Dehalococcoides sp.: Comparison between batch and column

402

experiments. Chemosphere 2009, 75 (2), 141–148.

403

(12)

Jugder, B.-E.; Ertan, H.; Bohl, S.; Lee, M.; Marquis, C. P.; Manefield, M. Organohalide



404

respiring bacteria and reductive dehalogenases: key tools in organohalide bioremediation.

405

Front. Microbiol. 2016, 7, 1–12.

406

(13) Azizian, M. F.; Behrens, S.; Sabalowsky, A.; Dolan, M. E.; Spormann, A. M.; Semprini,

407

L. Continuous-flow column study of reductive dehalogenation of PCE upon

408

bioaugmentation with the evanite enrichment culture. J. Contam. Hydrol. 2008, 100 (1–2),

409

11–21.

410

(14)

Berggren, D. R. V; Marshall, I. P. G.; Azizian, M. F.; Spormann, A. M.; Semprini, L.

411

Effects of sulfate reduction on the bacterial community and kinetic parameters of a

412

dechlorinating culture under chemostat growth conditions. Environ. Sci. Technol. 2013, 47

413

(4), 1879–1886.

414

(15)

Azizian, M. F.; Marshall, I. P. G.; Behrens, S.; Spormann, A. M.; Semprini, L.

415

Comparison of lactate, formate, and propionate as hydrogen donors for the reductive

416

dehalogenation of trichloroethene in a continuous-flow column. J. Contam. Hydrol. 2010,

417

113 (1–4), 77–92.

418

(16)

Miura, T.; Yamazoe, A.; Masako, I.; Ohji, S.; Hosoyama, A.; Takahata, Y.; Fujita, N. The

419

impact of injections of different nutrients on the bacterial community and its

420

dechlorination activity in chloroethene-contaminated groundwater. Microbes Env. 2015,

421

30 (2), 164–171.

422

(17)

Mendoza-Sanchez, I.; Autenrieth, R.; McDonald, T.; Cunningham, J. Effect of pore

423

velocity on biodegradation of cis-dichloroethene (DCE) in column experiments.

424

Biodegradation 2010, 21 (3), 365–377.

425 426

(18)

Mendoza-Sanchez, I.; Cunningham, J. Efficient algorithms for modeling the transport and biodegradation of chlorinated ethenes in groundwater. Transp. Porous Media 2012, 92


Page 20 of 29

Page 21 of 29


427 428

(1), 165–185. (19) Mendoza-Sanchez, I.; Cunningham, J. Efficient algorithm for modeling transport in

429

porous media with mass exchange between mobile fluid and reactive stationary media.

430

Transp. Porous Media 2007, 68 (3), 285–300.

431

(20)

cultures dominated by Dehalococcoides. Fems Microbiol. Ecol. 2006, 58, 538–549.

432 433

Duhamel, M.; Edwards, E. A. Microbial composition of chlorinated ethene-degrading

(21)

Cupples, A. M.; Spormann, A. M.; McCarty, P. L. Vinyl chloride and cis-dichloroethene

434

dechlorination kinetics and microorganism growth under substrate limiting conditions.

435

Environ. Sci. Technol. 2004, 51, 1635–1642.

436

(22)

Mayer-Blackwell, K.; Azizian, M.F.; Green, J.K.; Spormann, A.M.; Semprini, L. Survival

437

of vinyl chloride respiring Dehalococcoides mccartyi under long-term electron donor

438

limitation. Environ. Sci. Technol. 2017, 38 (4), 1102–1107.

439 440

(23)

Cunningham, J. A.; Mendoza-Sanchez, I. Equivalence of two models for biodegradation during contaminant transport in groundwater. Water Resour. Res. 2006, 42 (2), W02416.

441 442



443

Table 1. Summary of modeling scenarios applied to experimental data Scenario baseline simulation

Description of scenario all model parameters estimated from literature values, batch experiments, and non-reactive control experiments

hydrogen-limited simulation

rate of hydrogen production from electron donor reduced by a factor of 10 as compared to baseline simulation

decay-limited simulation

bacterial decay rate coefficient increased by a factor of 10 as compared to baseline simulation

metabolism-limited simulation

maximum bacterial growth rate coefficient and maximum specific substrate utilization rate coefficient both decreased by a factor of 10 as compared to baseline simulation

444 445 446


Page 22 of 29

Page 23 of 29

447


█

FIGURE CAPTIONS

448 449

Figure 1. Schematic representation of the experimental set-up (after Mendoza-Sanchez et

450

al.17). The arrow inside the column indicates the direction of the flow.

451 452

Figure 2. Simulated and experimental breakthrough curves of cis-DCE, VC, and ethene for

453

three sampling ports (port 2, 3, and 4) of the fast-flow column experiments.

454 455


456

three sampling ports (port 2, 3, and 4) of the slow-flow column experiments.

457 458


459

three sampling ports (port 2, 3, and 4) of the medium-flow column experiments.

460 461

Figure 5. Simulated and experimental concentration profiles for cis-DCE, VC, and ethene in

462

the slow-flow columns after 1.4 and 2.2 pore volumes.

463 464 465



466 467 468 469 470 471 472

Figure 1. Schematic representation of the experimental set-up (after Mendoza-Sanchez et al.17). The arrow inside the column indicates the direction of the flow.


Page 24 of 29

Page 25 of 29


473 474 475 476


477

three sampling ports (port 2, 3, and 4) of the fast-flow column experiments.

478



479 480 481 482


483

three sampling ports (port 2, 3, and 4) of the slow-flow column experiments.

484


Page 26 of 29

Page 27 of 29


485

486 487 488 489


490

three sampling ports (port 2, 3, and 4) of the medium-flow column experiments.

491



492 493 494 495

Figure 5. Simulated and experimental concentration profiles for cis-DCE, VC, and ethene in

496

the slow-flow columns after 1.41 and 2.2 pore volumes.

497


Page 28 of 29

Page 29 of 29


TOC art 84x47mm (300 x 300 DPI)

ACS Paragon Plus Environment

Biological Limitations of Dechlorination of cis-Dichloroethene during

Recommend Documents