Linking Spectral Induced Polarization (SIP) and Subsurface Microbial

Subscriber access provided by Oregon State University

Article

Linking spectral induced polarization (SIP) and subsurface microbial processes: Results from sand column incubation experiments Adrian Mellage, Christina M Smeaton, Alex Furman, Estella Atekwana, Fereidoun Rezanezhad, and Philippe Van Cappellen Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b04420 • Publication Date (Web): 16 Jan 2018 Downloaded from http://pubs.acs.org on January 16, 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 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 34

Environmental Science & Technology

1

Linking spectral induced polarization (SIP) and

2

subsurface microbial processes: Results from sand

3

column incubation experiments

4

Adrian Mellage†*, Christina M. Smeaton†, Alex Furman§, Estella A. Atekwana‡┴, Fereidoun

5

Rezanezhad† & Philippe Van Cappellen†

6

†

7

200 University Ave W, ON, Canada N2L 3G1

8

§

9

Israel

University of Waterloo, Water Institute and Department of Earth & Environmental Sciences,

Technion – Israel Institute of Technology, Civil and Environmental Engineering, Haifa 32000,

10

‡

11

Stillwater, OK 74078, USA

12

┴

13

Environment, Newark, DE 19716, USA

14

*

Oklahoma State University, Boone Pickens School of Geology, 105 Noble Research Center,

University of Delaware, Department of Geological Sciences, College of Earth, Ocean, and

Corresponding author. Phone: +1 519 888 4567 ext. 38757, E-mail: [email protected]

15

For submission to: Environmental Science & Technology

16

KEYWORDS: Spectral induced polarization (SIP), chargeability, relaxation time, DNRA,

17

dissimilatory iron reduction, Shewanella oneidensis, reactive transport, microbial growth.

ACS Paragon Plus Environment

1


Page 2 of 34

18

ABSTRACT

19

Geophysical techniques, such as Spectral Induced Polarization (SIP), offer potentially powerful

20

approaches for in-situ monitoring of subsurface biogeochemistry. The successful implementation

21

of these techniques as monitoring tools for reactive transport phenomena, however, requires the

22

de-convolution of multiple contributions to measured signals. Here, we present SIP spectra and

23

complementary biogeochemical data obtained in saturated columns packed with alternating

24

layers of ferrihydrite-coated and pure quartz sand, and inoculated with Shewanella oneidensis

25

supplemented with lactate and nitrate. A biomass-explicit diffusion-reaction model is fitted to the

26

experimental biogeochemical data. Overall, the results highlight that (1) the temporal response of

27

the measured imaginary conductivity peaks parallels the microbial growth and decay dynamics

28

in the columns, and (2) SIP is sensitive to changes in microbial abundance and cell surface

29

charging properties, even at relatively low cell densities (=) 6?@ + >6?@

(5)

569 where :;)9 [h-1] is the corresponding maximum specific growth rate of the bacteria, .=) [mol L1

] is the lactate concentration, .6?@ [mol L-1] the concentration of nitrate (x = 1) or nitrite (x = 2),

56 >=) [mol L-1] is the half-saturation constant for lactate uptake (assumed the same for both

DNRA steps), >6?@ [mol L-1] is the half-saturation constant for nitrate (x = 1) or nitrite (x = 2),

and A [cells LPM-1] is the cell abundance or biomass.


10

Page 11 of 34


209

Microbial growth using ferrihydrite (FH) as the terminal electron acceptor is energetically less

210

favorable than DNRA. Dissimilatory FH reduction is therefore assumed to be kinetically

211

inhibited by the presence of the higher energy yielding electron acceptors nitrate and nitrite: BE 4BC7D5 = :;)9 ∙

212 213

IJK >6? FB0GGG ∙ H .=) @ ∙ ∙ ∙A BE F IJK ∙ H + > .=) + >=) . + > B0GGG B0GGG 6?@ 6?@

(6)

BE where :;)9 [h-1] is the maximum specific growth rate supported by dissimilatory FH reduction,

FB0GGG [mol g-1] is the concentration of FH (multiplied by a conversion factor, H

M

= (" ∙ L O, N

where

214

BE ρb is the bulk density), and >=) and >B0GGG [mol L-1] are the half-saturation constants for the

215

electron donor and acceptor, respectively. The fourth term on the RHS is a generalized inhibition

216

IJK term,47 where >6? [mol L-1] is the inhibition constant due to the presence of NO3- or NO2-. @

217

When both nitrogen species are present, the expression for 4BC7D5 [cells LPM-1 h-1] contains two

218

inhibition terms.

219 220

With the above reaction rate expressions, the DRM yields the following system of equations governing the changes in aqueous and solid-bound chemical concentrations, plus the biomass: -.=) - .=) 1 1 1 1 = 0 − Q45678M ∙ − 45678 ∙ − 4BC7D5 ∙ S ∙ Q S -/ -2 RM R RE T -.8 - .8 1 1 1 1 = 0 + Q45678M ∙ + 45678 ∙ + 4BC7D5 ∙ S ∙ Q S -/ -2 RM R RE T -.6?E - .6?E 1 1 = 0 − 245678M ∙ ∙ Q S -/ -2 RM T

-.6? - .6? 1 1 1 = 0 + Q245678M ∙ − 0.6745678 ∙ S ∙ Q S -/ -2 RM R T -.6CY - .6CY 1 1 = 0 + 0.6745678 ∙ ∙ Q S -/ -2 R T

]M -.B0 - .B0 1−T 1 1 = 0 ∙ + ( > − 0.84BC7D5 ∙ ∙ Q S Z1 \ [ -/ -2 T RE T

(7) (8) (9) (10) (11) (12)


11


.BC 1 = − 10.44BC7D5 ∙ / RE

Page 12 of 34

(13)

._8` 1 = − 3.24BC7D5 ∙ / RE

(14)

A = 45678M + 45678 + 4BC7D5 − 45Db8c /

(15)

221

In Equations (7-12), rates describing concentration changes of aqueous species (in units of mol

222

L-1 h-1) are obtained by dividing the microbial growth rates by their corresponding growth yields,

223

Yi [cells molLac-1]. The Yi values are continuously updated during a simulation based on the

224

temporally changing Gibbs energies of the catabolic reactions.48 The reader is referred to the

225

Supporting Information (Figure S1) for a detailed explanation of the dynamic growth yield

226

calculations. The rate of change of the biomass concentration, Equation (15), is expressed in

227

units of cells LPM-1 h-1, while changes of FH and magnetite concentrations over time are given in

228

mol LPM-1 h-1. Sorption of Fe2+ produced by the reductive dissolution of FH is included assuming

229

a linear isotherm, which yields the retardation term in squared brackets on the RHS of equation

230

(12). Transport and reaction were solved iteratively at each time step (∆t = 60 s) in MATLAB®.

231

The Crank-Nicolson method was implemented to solve the diffusive transport of dissolved

232

species (∆x = 1 mm), with zero-flux boundary conditions at the top and bottom of the domain.

233

The solutions for the ordinary differential equations describing the reactive terms were

234

approximated using the fifth-order Runge-Kutta49 method.

235


12

Page 13 of 34

236


RESULTS AND DISCUSSION

237

Geochemistry and DRM. The aqueous geochemical data along with fitted DRM trends for

238

the three layers in the column reactors (i.e., the top and bottom FS layers and the middle QS

239

layer) are presented in Figure 2. The overall agreement between the model simulations and the

240

experimental results support our conceptual understanding of the reactive system. During the

241

initial 56 hours of the experiment, DNRA dominates: NO3- concentrations decrease rapidly

242

within the first 32 hours of the experiment, resulting in the production of NO2- which is

243

concurrently consumed as the electron acceptor in the second DNRA step (Figure 2d-f). The

244

simulated peak concentration of NO2- in all three layers coincides with the complete depletion of

245

NO3-, beyond which, according to the DRM simulations, NO2- is no longer produced. Continued

246

utilization of NO2- then causes its disappearance by t = 56 hours.

247

Upon the complete consumption of NO2-, the detection of aqueous Fe2+ followed by its

248

generally increasing concentration provides evidence for iron reduction in both FS layers for the

249

remaining 181 h of the experiment (Figure 2g-i). In the two FS layers, FH reduction is

250

accompanied by the further decrease in the lactate concentration and increase in the acetate

251

concentration. By contrast, the concentrations of lactate and acetate in the FH-free QS layer

252

remain nearly constant after cessation of DNRA. The slight increase in aqueous Fe2+ in the QS

253

layer is due to diffusion from the FS layers.

254

Given that the geochemical data were obtained from six successive sacrificial columns, the

255

consistency of the temporal trends in Figure 2 imply that the experimental design yielded

256

reproducible reactive geomicrobial systems. Notwithstanding the differences in the measured

257

aqueous concentrations of NO3-, NO2- and Fe2+ between the top and bottom FS layers, lactate

258

consumption was not significantly different between the two layers (regression p-value = 0.017),


13


Page 14 of 34

259

thus suggesting comparable microbial growth kinetics. The slower conversion of lactate to

260

acetate after t = 56 hours, indicates, as expected, that S. oneidensis grows faster when using

261

DNRA as its energy-yielding reaction, relative to dissimilatory FH reduction. The model fit to

262

the data yields maximum specific growth rates with NO3- and NO2- as terminal electron donors

263

that are 5-6 times higher than with FH. A complete list of the values of the reaction parameters

264

applied in the simulations shown in Figure 2 is given in Table 2 (see Supporting Information for

265

the transport parameter values, Table S1).

266

Biomass growth. The DRM predicts the changes in microbial biomass under the experimental

267

conditions (see Equation 15). The calculations assume a uniform initial cell density equal to that

268

of the saturating pore solution (1.26×107 cells mLPM-1). During the DNRA phase, relatively rapid

269

growth occurs in both the FS and QS layers reaching a maximum cell density of 6.6×107 cells

270

mLPM-1. The best fit to the geochemical results was obtained by using a slightly lower maximum

271

specific growth rate (µmax) for the second step in the overall DNRA reaction (see Table 2),

272

possibly reflecting a toxic effect of NO2- on S. oneidensis.39

273

Upon complete consumption of NO2-, the predicted microbial growth rate slows down in the

274

FS layers, with the biomass increasing up to 8.1×107 cells mLPM-1 by the end of the experiment.

275

In the absence of an electron acceptor, cellular decay dominates in the QS layer, with the

276

biomass decreasing to a final concentration of 3.9×107 cells mLPM-1. The DRM predicted cell

277

densities fall within the ranges estimated from the ATP concentrations measured in the FS and

278

QS layers over the course of the experiment, 1.26 – 8.1×107 and 0.3 – 7.1×107 cells mLPM-1,

279

respectively (based on an empirical relationship between ATP and cell density, see Supporting

280

Information).


14

Page 15 of 34


281

SIP responses. The time-dependent (spectral) imaginary conductivities (σ'') measured in the

282

top FS and middle QS layers at selected time-points during the experiment are shown in Figure

283

3. A frequency peak for σ'' appeared approximately 12 hours after starting the experiment.

284

Presumably, this is the time required for equilibration of electrochemical exchange processes

285

affecting the chemistry of the EDL of the cells and sand grains in the freshly packed columns. In

286

the top FS layer, σ'' showed a distinct peak developing between 5 and 10 Hz and increasing in

287

magnitude from 2.3×10-4 to a maximum of 5.2×10-4 S m-1 (Figure 3a). The peak in the FS layer

288

first shifted to lower frequencies with increasing time, followed by a shift to higher frequencies

289

between 124 and 237 hours. A similar peak developed in the QS layer (Figure 3b), but at a

290

slightly higher frequency, between 10 and 40 Hz, and only reached a maximum value of 2.7×10-4

291

S m-1. The measured maximum phase shifts in the top FS and QS layers were 3.6 and 1.4 mrad,

292

respectively (for phase shift and impedance data see Supporting Information, S4 and S5). The σ''

293

spectra prior to 72 h in the QS layer did not exhibit a peak and measurable polarization was only

294

detected between 0.01 and 0.1 Hz (Figure 3b). We ascribe this to poor contact at the electrode-

295

sand interface, likely due to air entrapment during the packing procedure. Shortly before the 72 h

296

measurement, the potential electrodes in the QS layer were tightened, which improved contact

297

and yielded the σ'' spectra shown in Figure 3b. The absence of a frequency peak before 72 h in

298

the QS layer is thus believed to reflect a technical issue rather than variations in the EDL.

299

The σ'' spectral response in the bottom FS layer did not exhibit a distinct peak. Sedimentation

300

of poorly bound FH (from the iron coating procedure) during packing resulted in an

301

accumulation of fine FH particles at the bottom of the columns, inferred from the lower

302

impedance magnitudes measured in the bottom layer (see Supporting Information, section S5).

303

We speculate that the elevated pore water conductivity in the bottom FS layer as well as


15


Page 16 of 34

304

adsorption of poorly bound FH to the cells’ outer membrane dampened the electrochemical

305

polarization of the bacteria. The SIP data from the bottom FS layer is therefore omitted from the

306

discussion here (for example data and further details, see Supporting Information, section S5).

307

The production of dissolved ionic species from both DNRA and reductive dissolution of the

308

FH coating could potentially affect the EDL polarization of both the sand grains and bacteria.

309

The production of about 2 mM NH4+ by DNRA likely had little impact on the electrochemical

310

polarization of the Stern layer because of the already high porewater concentration of NH4+ (18

311

mM), and the fact that monovalent cations remain hydrated and tend to form mobile outer sphere

312

complexes at solid surfaces.24 In contrast, Fe2+ more strongly adsorbs, forming inner sphere

313

complexes at the cell membrane and sand grain surfaces, resulting in a reduction of ion mobility

314

in the EDL and a potential dampening of polarization.24,

315

production on the measured temporal dynamics of σ'' is not considered to be very strong, as

316

evidenced by the relaxation time dynamics discussed in the next section.

50

Nonetheless, the effect of Fe2+

317

The different spectral locations of the σ'' peaks in the FS and QS layers likely reflect the

318

presence or absence of the FH coating, due to the possible contribution of membrane polarization

319

to the measured spectra. Additionally, the shift in peak frequency observed in the FS layer (but

320

not observed in the QS layer, Figure 3b and c) may be due to changes in EDL properties of the

321

iron coating, for example, as a result of cell attachment and, following the end of DNRA,

322

sorption of Fe2+ and the formation of a reduced precipitate, tentatively identified as magnetite.

323

Because these effects are absent from the QS layer, the corresponding changes in the σ'' spectra

324

should be more uniquely related to biomass changes.8 The spectra recorded in the QS layer fall

325

in the same spectral range as those previously reported for cell suspensions and cell-sand

326

mixtures of Zymomonas mobilis, whose SIP responses were attributed to the polarization of the


16

Page 17 of 34


327

bacterial cells.3 Thus, overall, the measured SIP responses appear to be sensitive to the

328

abundance and surface electrical properties of the cells, even at the relative low cell densities

329

used in the experiment.51

330

Linking SIP and biogeochemistry. The dependence of the SIP signals on changes in the

331

microbial biomass are supported by comparing the temporal trends in the peak values of σ'' and

332

the cell densities predicted by the DRM (Figure 4). For the FS layer, the σ'' peak values capture

333

the sharp decrease in net biomass growth that accompanies the transition from DNRA to

334

dissimilatory iron reduction. In the QS layer, however, the trend of the σ'' peak value lags behind

335

the predicted cell density curve. We attribute the initial lag in electrode response before 72 h to

336

poor contact as discussed in the previous section. The lag between 72 and 118 h, possibly stems

337

from low point of zero charge of pure quartz (pHpzc = 3.0)52 creating a less favorable

338

environment for attachment of the negatively charged cells, than the FH coated sand (pHpzc = 8.0

339

for pure FH),53 thus hindering the build-up of stationary chargeable surfaces. This hypothesis is

340

consistent with Abdel Aal et al.20 who measured higher levels of bacterial adsorption and

341

increasing σ'' as a function of the increasing fraction of iron-oxide coated sand. Nonetheless, in

342

contrast to the FS layer, the peak σ'' value in the QS layer decreases during the second half of the

343

experiment, consistent with the predicted biomass decay following the exhaustion of all external

344

terminal electron acceptors.

345

The maximum σ'' peak value reached in the QS layer is about half that in the FS layer (Figure

346

4). A higher chargeability in the FS layer is expected because of the additional polarization

347

associated with the presence of the FH coating.5 Higher pore water conductivity in the middle QS

348

layer (see Supporting Information, section S5) likley also contributed to reduce σ'' in the QS

349

layer.54 The results therefore highlight that the relationship between SIP and microbial


17


Page 18 of 34

350

abundance is non-unique and depends on the nature of the matrix. This obviously complicates

351

the interpretation of SIP signals in natural porous media.

352

The fitted relaxation times (τ) extracted from the Cole-Cole model fits44 (refer to Supporting

353

Information, Figure S3) show systematic changes over time (Figure 5c and d). Normalized

354

chargeability (mn) values were also extracted and are presented in the Supporting Information for

355

the top FS layer. Trends in mn are analogous to those measured for peak σ'' and are therefore not

356

further discussed. In both the FS and QS layers, the highest τ values occurred towards the end of

357

the DNRA phase. Subsequently, τ values in the QS layer dropped abruptly, while they decreased

358

more slowly in the FS layer. Given that τ is related to the length-scale over which ionic back-

359

diffusion takes place after current is applied (Equation 3), several processes could explain the

360

observed τ trends. If the observed SIP spectra mainly result from the polarization of the

361

microbial cells,3 then the temporal changes of τ could be due to changes in cell size, where cells

362

become smaller during periods of energetic stress55 as they consume internal energy reserves.56,

363

57

364

layers. The most favorable catabolic reaction (i.e., DNRA) would result in cells growing larger,

365

while the cells would shrink slowly under the less favorable iron reducing conditions and more

366

rapidly in the absence of external electron acceptors.58

In other words, the τ trends could reflect the variable energetic conditions in the FS and QS

367

Without any direct experimental evidence to confirm changes in cell size we acknowledge that

368

other factors may contribute to variations in Cole-Cole parameters including, for example,

369

surface complexation of divalent cations (in particular Fe2+, which would have resulted in an

370

increase in τ, the opposite of what is observed and therefore not likely a major driver of the

371

temporal trends of τ),24,

372

chargeability22 and other EDL properties of the cells.59

50

mineral transformations,4,

5

plus changes in the morphology,56,

57


18

Page 19 of 34


373

The range in relaxation times obtained here (0.012 – 0.1068 s) corresponds to effective

374

polarization diameters between 1.04 and 3.1 µm, using Equation (3) together with the model of

375

Revil et al.22 for the polarization of a typical bacterial cell (which yields Ds = 11.2 µm2 s-1).

376

Estimates of diffusion coefficients of ion mobility in the Stern layer remain uncertain.60 Weller et

377

al.60 highlight the large variations in surface diffusion coefficients among different sandstones

378

and postulate that d, the length scale defined by τ in Equation 3, is not only a function of particle

379

size, but also particle surface characteristics and the mineral composition of a given sample. For

380

bacterial cells, cell surface chemistry along with cell size controls Ds. Both of these properties

381

are temporally dynamic and we acknowledge that the value for Ds is likely quite variable.

382

Nevertheless, the available Ds estimate for bacterial cells yields dimensions that fall within the

383

range of typical cell diameters or lengths in the case of rod-shaped bacteria such as S.

384

oneidensis,61 but are orders of magnitude smaller than the smallest sand grains (150 µm), ruling

385

out a major control associated with electrochemical polarization at the sand-grain scale. The

386

magnitude of the relaxation time, together with the evidence discussed earlier, thus strongly

387

suggest that polarization of the cells largely controls the observed SIP responses, even in the

388

presence of the iron coating in the FS layer.

389

In their study of microbially mediated iron sulfide mineral transformations, Slater et al.5

390

derived τ values between 0.2 and 0.4 s, that is, values at the upper end of our range. They further

391

calculated effective polarization diameters that were much larger than the FeS encrusted cells

392

observed in their study. Their calculation, however, was based on a surface ionic diffusion

393

coefficient for sands (Ds = 3000 µm2 s-1).62 If, instead, the Ds value for bacterial cells proposed

394

by Revil et al.22 is used (Ds = 11.2 µm2 s-1), then the effective polarization diameter is on the

395

order of 4 – 6 µm, which agrees well with the size of the FeS encrusted cells. Thus, we speculate


19


Page 20 of 34

396

that in the study of Slater et al.5 SIP signals were also controlled by polarization processes acting

397

at the cellular scale, rather than at the grain or pore space scales.

398

In summary, the SIP signatures obtained in the sand columns record changes in the density,

399

size and surface charging properties of the bacterial cells, even at relative low cell densities (≤

400

108 cells mL-1), such as those found in subsurface environments, and in the presence of

401

polarizable mineral surfaces. While our results are encouraging, the acquisition of SIP data that

402

can reliably detect variations in phase shifts in the range reported here is not yet feasible at the

403

field scale. However, our results provide new insights into the mechanisms controlling measured

404

polarization responses. Hence, they help posit SIP as a non-invasive method for monitoring

405

microbial activity in bench-scale reactive transport studies, and they highlight the need to

406

improve the sensitivity of SIP measurements as a necessary condition for its use in field

407

applications.


20

Page 21 of 34

408


FIGURES

409

Figure 1. Schematic diagram and photograph of the packed column reactor equipped with non-

410

polarizable Ag-AgCl electrodes. The TOP and BOT layers consist of ferrihydrite (FH) coated

411

quartz sand and the MID layer of pure quartz sand. C and P followed by a number correspond to

412

(+) or (-) current and potential electrodes, respectively. The magnification on the right shows the

413

color of the FH coating at the start of the experiment and after 237 h.

414

Figure 2. Measured pore-water geochemistry from the sacrificially sampled column reactors and

415

corresponding diffusion-reaction model (DRM) fits (Equations 7 to 15). Data points are

416

represented by individual markers and model fits are plotted as solid (lactate, nitrate and ferrous

417

iron) and dashed lines (acetate and nitrite).

418

Figure 3. Imaginary conductivity (σ'') response between 0.01 and 1000 Hz at selected time

419

points: (a) increase in σ'' and stabilization within the upper iron coated sand (FS) layer; (b)

420

increase in σ'' in the quartz sand (QS) layer during the first 162 hours, and (c) decrease in σ'' in

421

the QS layer beyond 162 hours.

422

Figure 4. Comparison of the temporal trends of the diffusion-reaction model (DRM) simulated

423

abundance of Shewanella oneidensis and the measured peak SIP-imaginary conductivity in the

424

upper iron coated sand (FS) layer (a) and quartz sand (QS) layer (b). The right-hand axis scale

425

varies between the plots due the difference in the magnitude of the peak σ'' values between the

426

layers. The scale was adjusted to highlight the similarities between the cell density and σ'' trends.

427

Figure 5. Temporal changes in diffusion-reaction model (DRM) predicted reaction rates (a and

428

b) and SIP-derived relaxation times (c and d); DNRA1 and 2 are the two consecutive steps in

429

dissimilatory nitrate reduction to ammonium; FHRED is the rate of ferrihydrite (FH) reduction


21


Page 22 of 34

430

and DECAY is the first-order cellular decay rate. The plots on the left are for the upper iron

431

coated sand (FS) layer, those on the right for the QS layer. The omission of relaxation time data

432

before 70 hours in the quartz sand (QS) layer is due to the late appearance of a spectral peak in

433

the SIP responses, related to electrode contact issues discussed in the text, which yields Cole-

434

Cole fits that are dependent on the initial conditions.


22

Page 23 of 34


435

TABLES

436

Table 1. Microbially mediated reactions; stoichiometries are normalized to 1 mol of lactate. Dissimilatory nitrate reduction to ammonium (DNRA)

∆Gr [kJ molLac-1]a

Y [cells mol-1] a,b

1

C3H5O3- + 2NO3- → C2H3O2- + 2NO2- + HCO3- + H+

-320

5.72×1013

2

C3H5O3- + 0.67NO2- + 0.67H2O + 0.33H+ → C2H3O2- + 0.67NH4+ + HCO3-

-301

5.44×1013

∆Gr [kJ molLac-1] a

Y [cells mol-1] a

-252

4.76×1013

C3H5O3- + 0.5NO3- + 0.5H2O → C2H3O2- + 0.5NH4+ + 2HCO3Ferrihydrite reduction to magnetite (FHRED) 3

C3H5O3- + 10.4Fe(OH)3(s) + 0.6H+ → C2H3O2- + 3.2Fe3O4(s) + 0.8Fe2+ + 16.4H2O + HCO3-

a

Gibbs free energies and microbial growth yields [cells molLac-1], under the initial experimental conditions.48

Yield coefficients converted from [C-molbiomass molLac-1] using an average cell mass of 3.5·1012 [cells g-1] for Shewanella sp.63-65 assuming an idealized biomass chemical composition of C5H9O2.5N.

b

437 438

Table 2. Reaction parameters of the diffusive reactive transport model (DRM) where ED and EA

439

are the electron donor and electron acceptor, respectively. Parameter

Units

Value

Parameter

Maximum specific growth rates 5M :;)9 5 :;)9 BE :;)9

Value

EA Monod and inhibition constants

[h-1]

2.5×10-2

a

[h-1]

2.0×10-2

a

[h-1]

4.0×10-3

a

>6?e >6?f

>B0GGG IJK >6? @

ED Monod constants

a

Units

[mol L-1]

2.0×10-6

66

[mol L-1]

4.1×10-6

66

[mol L-1]

2.0×10-3

67

[mol L-1]

1.6×10-5

47

[h-1]

2.2×10-3

a

[L kg-1]

0.4

a

Decay and sorption

56 >=)

[mol L-1]

1.0×10-3

BE >=)

[mol L-1]

5.2×10-4

66, 67

68

!5Db8c >[

fitted parameter estimated from literature ranges

66-68, 47

440


23


Page 24 of 34

441

AUTHOR INFORMATION

442

Corresponding Author

443

Adrian Mellage ([email protected]); Phone: +1 519-888-4567 ext. 38757

444

ACKNOWLEDGEMENTS

445

Funding for this work was provided by the Canada Excellence Research Chair (CERC) program

446

and the Waterloo-Technion University Cooperation Programme. We would also like to

447

acknowledge the TERRE-CREATE program’s support for a research exchange, enabling

448

collaboration with Technion – Israel Institute of Technology. We thank three anonymous

449

reviewers and Dr. Lee Slater for their constructive comments which greatly improved this work.

450 451

ASSOCIATED CONTENT

452

Supporting Information. Additional material includes Supporting Information, with a more

453

detailed description of experimental preparation methods as well as addenda regarding specific

454

model calculations. The MATLAB code used in this study is available from the authors upon

455

request. This information is available free of charge via the Internet at http://pubs.acs.org.

456 457


24

Page 25 of 34


458

REFERENCES

459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502

1. Hao, N.; Moysey, S. M.; Powell, B. A.; Ntarlagiannis, D., Evaluation of Surface Sorption Processes Using Spectral Induced Polarization and a 22Na Tracer. Environ. Sci. Technol. 2015, 49, (16), 9866-9873. 2. Davis, C. A.; Atekwana, E.; Atekwana, E.; Slater, L. D.; Rossbach, S.; Mormile, M. R., Microbial growth and biofilm formation in geologic media is detected with complex conductivity measurements. Geophys. Res. Lett. 2006, 33, (18), L18403. 3. Zhang, C.; Revil, A.; Fujita, Y.; Munakata-Marr, J.; Redden, G., Quadrature conductivity: A quantitative indicator of bacterial abundance in porous media. Geophysics 2014, 79, (6), D363-D375. 4. Personna, Y. R.; Ntarlagiannis, D.; Slater, L.; Yee, N.; O'Brien, M.; Hubbard, S., Spectral induced polarization and electrodic potential monitoring of microbially mediated iron sulfide transformations. J. Geophys. Res. 2008, 113, G02020. 5. Slater, L.; Ntarlagiannis, D.; Personna, Y. R.; Hubbard, S., Pore‐scale spectral induced polarization signatures associated with FeS biomineral transformations. Geophys. Res. Lett. 2007, 34, L21404. 6. Ntarlagiannis, D.; Williams, K. H.; Slater, L.; Hubbard, S., Low‐frequency electrical response to microbial induced sulfide precipitation. J. Geophys. Res. 2005, 110, G02009. 7. Placencia-Gómez, E.; Slater, L.; Ntarlagiannis, D.; Binley, A., Laboratory SIP signatures associated with oxidation of disseminated metal sulfides. J. Contam. Hydrol. 2013, 148, 25-38. 8. Placencia-Gómez, E.; Parviainen, A.; Slater, L.; Leveinen, J., Spectral induced polarization (SIP) response of mine tailings. J. Contam. Hydrol. 2015, 173, 8-24. 9. Abdel Aal, G. Z.; Atekwana, E. A., Spectral induced polarization (SIP) response of biodegraded oil in porous media. Geophys. J. Int. 2014, 196, (2), 804-817. 10. Schwartz, N.; Huisman, J.; Furman, A., The effect of NAPL on the electrical properties of unsaturated porous media. Geophys. J. Int. 2012, 188, (3), 1007-1011. 11. Schwartz, N.; Shalem, T.; Furman, A., The effect of organic acid on the spectral-induced polarization response of soil. Geophys. J. Int. 2014, 197, (1), 269-276. 12. Williams, K. H.; Kemna, A.; Wilkins, M. J.; Druhan, J.; Arntzen, E.; N’Guessan, A. L.; Long, P. E.; Hubbard, S. S.; Banfield, J. F., Geophysical monitoring of coupled microbial and geochemical processes during stimulated subsurface bioremediation. Environ. Sci. Technol. 2009, 43, (17), 6717-6723. 13. Revil, A.; Florsch, N., Determination of permeability from spectral induced polarization in granular media. Geophys. J. Int. 2010, 181, (3), 1480-1498. 14. Marshall, D. J.; Madden, T. R., Induced polarization, a study of its causes. Geophysics 1959, 24, (4), 790-816. 15. Atekwana, E. A.; Slater, L. D., Biogeophysics: A new frontier in earth science research. Rev. Geophys. 2009, 47, (4), RG4004. 16. Atekwana, E. A.; Atekwana, E. A., Geophysical signatures of microbial activity at hydrocarbon contaminated sites: A review. Surv. Geophys. 2010, 31, (2), 247-283. 17. Prodan, C.; Mayo, F.; Claycomb, J.; Miller Jr, J.; Benedik, M., Low-frequency, low-field dielectric spectroscopy of living cell suspensions. J. Appl. Phys. 2004, 95, (7), 3754-3756. 18. Abdel Aal, G. Z.; Atekwana, E. A.; Slater, L. D.; Atekwana, E. A., Effects of microbial processes on electrolytic and interfacial electrical properties of unconsolidated sediments. Geophys. Res. Lett. 2004, 31, (12), L12505.


25


503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548

Page 26 of 34

19. Ntarlagiannis, D.; Yee, N.; Slater, L., On the low‐frequency electrical polarization of bacterial cells in sands. Geophys. Res. Lett. 2005, 32, L24402. 20. Abdel Aal, G.; Atekwana, E.; Radzikowski, S.; Rossbach, S., Effect of bacterial adsorption on low frequency electrical properties of clean quartz sands and iron‐oxide coated sands. Geophys. Res. Lett. 2009, 36, (4), L04403. 21. Placencia-Gómez, E.; Slater, L. D., Electrochemical spectral induced polarization modeling of artificial sulfide-sand mixtures. Geophysics 2014, 79, (6), EN91-EN106. 22. Revil, A.; Atekwana, E.; Zhang, C.; Jardani, A.; Smith, S., A new model for the spectral induced polarization signature of bacterial growth in porous media. Water Resour. Res. 2012, 48, (9), W09545. 23. Leroy, P.; Revil, A., A mechanistic model for the spectral induced polarization of clay materials. J. Geophys. Res. 2009, 114, B10202. 24. Vaudelet, P.; Revil, A.; Schmutz, M.; Franceschi, M.; Bégassat, P., Induced polarization signatures of cations exhibiting differential sorption behaviors in saturated sands. Water Resour. Res. 2011, 47, (2). 25. Pelton, W.; Ward, S.; Hallof, P.; Sill, W.; Nelson, P. H., Mineral discrimination and removal of inductive coupling with multifrequency IP. Geophysics 1978, 43, (3), 588-609. 26. Cole, K. S.; Cole, R. H., Dispersion and absorption in dielectrics I. Alternating current characteristics. J. Chem. Phys. 1941, 9, (4), 341-351. 27. Nordsiek, S.; Weller, A., A new approach to fitting induced-polarization spectra. Geophysics 2008, 73, (6), F235-F245. 28. Ustra, A.; Slater, L.; Ntarlagiannis, D.; Elis, V., Spectral induced polarization (SIP) signatures of clayey soils containing toluene. Near Surf. Geophys. 2012, 10, (6), 503-515. 29. Zisser, N.; Kemna, A.; Nover, G., Relationship between low-frequency electrical properties and hydraulic permeability of low-permeability sandstones. Geophysics 2010, 75, (3), E131-E141. 30. Chen, J.; Kemna, A.; Hubbard, S. S., A comparison between Gauss-Newton and Markovchain Monte Carlo–based methods for inverting spectral induced-polarization data for Cole-Cole parameters. Geophysics 2008, 73, (6), F247-F259. 31. Slater, L., Near surface electrical characterization of hydraulic conductivity: From petrophysical properties to aquifer geometries—A review. Surv. Geophys. 2007, 28, (2-3), 169197. 32. Zhang, C.; Slater, L.; Redden, G.; Fujita, Y.; Johnson, T.; Fox, D., Spectral induced polarization signatures of hydroxide adsorption and mineral precipitation in porous media. Environ. Sci. Technol. 2012, 46, (8), 4357-4364. 33. Slater, L.; Day‐Lewis, F.; Ntarlagiannis, D.; O'Brien, M.; Yee, N., Geoelectrical measurement and modeling of biogeochemical breakthrough behavior during microbial activity. Geophys. Res. Lett. 2009, 36, (LI4402). 34. Müller, B.; Stierli, R.; Gächter, R., A low-tech, low-cost passive sampler for the longterm monitoring of phosphate loads in rivers and streams. J. Environ. Monit. 2008, 10, (7), 817820. 35. Myers, C.; Nealson, K. H., Bacterial manganese reduction and growth with manganese oxide as the sole electron acceptor. Science 1988, 240, (4857), 1319-1321. 36. Viollier, E.; Inglett, P.; Hunter, K.; Roychoudhury, A.; Van Cappellen, P., The ferrozine method revisited: Fe(II)/Fe(III) determination in natural waters. Appl. Geochem. 2000, 15, (6), 785-790.


26

Page 27 of 34

549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594


37. Schwertmann, U., Solubility and dissolution of iron oxides. In Iron nutrition and interactions in plants, Springer: 1991; pp 3-27. 38. Gao, H.; Yang, Z. K.; Barua, S.; Reed, S. B.; Romine, M. F.; Nealson, K. H.; Fredrickson, J. K.; Tiedje, J. M.; Zhou, J., Reduction of nitrate in Shewanella oneidensis depends on atypical NAP and NRF systems with NapB as a preferred electron transport protein from CymA to NapA. ISME J. 2009, 3, (8), 966-976. 39. Zhang, H.; Fu, H.; Wang, J.; Sun, L.; Jiang, Y.; Zhang, L.; Gao, H., Impacts of nitrate and nitrite on physiology of Shewanella oneidensis. PLoS ONE 2013, 8, (4), e62629. 40. Heidelberg, J. F.; Paulsen, I. T.; Nelson, K. E.; Gaidos, E. J.; Nelson, W. C.; Read, T. D.; Eisen, J. A.; Seshadri, R.; Ward, N.; Methe, B., Genome sequence of the dissimilatory metal ion–reducing bacterium Shewanella oneidensis. Nat. Biotechnol. 2002, 20, (11), 1118-1123. 41. Ringuet, S.; Sassano, L.; Johnson, Z. I., A suite of microplate reader-based colorimetric methods to quantify ammonium, nitrate, orthophosphate and silicate concentrations for aquatic nutrient monitoring. J. Environ. Monit. 2011, 13, (2), 370-376. 42. Hobbie, J. E.; Daley, R. J.; Jasper, S., Use of nuclepore filters for counting bacteria by fluorescence microscopy. Appl. Environ. Microbiol. 1977, 33, (5), 1225-1228. 43. Abdel Aal, G. Z.; Atekwana, E. A., Spectral induced polarization (SIP) response of biodegraded oil in porous media. Geophys. J. Int. 2014, 196, 805-817. 44. Weigand, M.; Kemna, A., Relationship between Cole–Cole model parameters and spectral decomposition parameters derived from SIP data. Geophys. J. Int. 2016, 205, (3), 14141419. 45. Bretschger, O.; Obraztsova, A.; Sturm, C. A.; Chang, I. S.; Gorby, Y. A.; Reed, S. B.; Culley, D. E.; Reardon, C. L.; Barua, S.; Romine, M. F., Current production and metal oxide reduction by Shewanella oneidensis MR-1 wild type and mutants. Appl. Environ. Microbiol. 2007, 73, (21), 7003-7012. 46. Coker, V.; Gault, A.; Pearce, C.; Van der Laan, G.; Telling, N.; Charnock, J.; Polya, D.; Lloyd, J., XAS and XMCD evidence for species-dependent partitioning of arsenic during microbial reduction of ferrihydrite to magnetite. Environ. Sci. Technol. 2006, 40, (24), 77457750. 47. Watson, I. A.; Oswald, S. E.; Mayer, K. U.; Wu, Y.; Banwart, S. A., Modeling kinetic processes controlling hydrogen and acetate concentrations in an aquifer-derived microcosm. Environ. Sci. Technol. 2003, 37, (17), 3910-3919. 48. Heijnen, J.; Van Dijken, J., In search of a thermodynamic description of biomass yields for the chemotrophic growth of microorganisms. Biotechnol. Bioeng. 1992, 39, (8), 833-858. 49. Butcher, J. C., Numerical methods for ordinary differential equations. John Wiley & Sons: 2016. 50. Vaudelet, P.; Revil, A.; Schmutz, M.; Franceschi, M.; Bégassat, P., Changes in induced polarization associated with the sorption of sodium, lead, and zinc on silica sands. J. Colloid Interface Sci. 2011, 360, (2), 739-752. 51. Griebler, C.; Mindl, B.; Slezak, D.; Geiger-Kaiser, M., Distribution patterns of attached and suspended bacteria in pristine and contaminated shallow aquifers studied with an in situ sediment exposure microcosm. Aquat. Microb. Ecol. 2002, 28, (2), 117-129. 52. Sverjensky, D. A., Zero-point-of-charge prediction from crystal chemistry and solvation theory. Geochim. Cosmochim. Acta 1994, 58, (14), 3123-3129. 53. Schwertmann, U.; Fechter, H., The point of zero charge of natural and synthetic ferrihydrites and its relation to adsorbed silicate. Clay Miner. 1982, 17, (4), 471-476.


27


595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636

Page 28 of 34

54. Slater, L.; Glaser, D., Controls on induced polarization in sandy unconsolidated sediments and application to aquifer characterization. Geophysics 2003, 68, (5), 1547-1558. 55. Mellage, A.; Eckert, D.; Grösbacher, M.; Inan, A. Z.; Cirpka, O. A.; Griebler, C., Dynamics of suspended and attached aerobic toluene degraders in small-scale fow-through sediment systems under growth and starvation conditions. Environ. Sci. Technol. 2015, 49, (12), 7161-7169. 56. Lennon, J. T.; Jones, S. E., Microbial seed banks: the ecological and evolutionary implications of dormancy. Nat. Rev. Microbiol. 2011, 9, (2), 119-130. 57. Lever, M. A.; Rogers, K. L.; Lloyd, K. G.; Overmann, J.; Schink, B.; Thauer, R. K.; Hoehler, T. M.; Jørgensen, B. B., Life under extreme energy limitation: a synthesis of laboratory-and field-based investigations. FEMS Microbiol. Rev. 2015, 39, (5), 688-728. 58. Riedel, T. E.; Berelson, W. M.; Nealson, K. H.; Finkel, S. E., Oxygen consumption rates of bacteria under nutrient-limited conditions. Appl. Environ. Microbiol. 2013, 79, (16), 49214931. 59. Soni, K. A.; Balasubramanian, A. K.; Beskok, A.; Pillai, S. D., Zeta potential of selected bacteria in drinking water when dead, starved, or exposed to minimal and rich culture media. Curr. Microbiol. 2008, 56, (1), 93-97. 60. Weller, A.; Zhang, Z.; Slater, L.; Kruschwitz, S.; Halisch, M., Induced polarization and pore radius—A discussion. Geophysics 2016, 81, (5), D519-D526. 61. Lee, C. K.; Kim, A. J.; Santos, G. S.; Lai, P. Y.; Lee, S. Y.; Qiao, D. F.; Anda, J. D.; Young, T. D.; Chen, Y.; Rowe, A. R., Evolution of cell size homeostasis and growth rate diversity during initial surface colonization of Shewanella oneidensis. ACS Nano 2016, 10, 9183−9192. 62. Tarasov, A.; Titov, K., Relaxation time distribution from time domain induced polarization measurements. Geophys. J. Int. 2007, 170, (1), 31-43. 63. Yoon, S.; Sanford, R. A.; Löffler, F. E., Shewanella spp. use acetate as an electron donor for denitrification but not ferric iron or fumarate reduction. Appl. Environ. Microbiol. 2013, 79, (8), 2818-2822. 64. Saltikov, C. W.; Cifuentes, A.; Venkateswaran, K.; Newman, D. K., The ars detoxification system is advantageous but not required for As (V) respiration by the genetically tractable Shewanella species strain ANA-3. Appl. Environ. Microbiol. 2003, 69, (5), 2800-2809. 65. Bonneville, S.; Behrends, T.; Van Cappellen, P.; Hyacinthe, C.; Röling, W. F., Reduction of Fe(III) colloids by Shewanella putrefaciens: a kinetic model. Geochim. Cosmochim. Acta 2006, 70, (23), 5842-5854. 66. Yan, S.; Liu, Y.; Liu, C.; Shi, L.; Shang, J.; Shan, H.; Zachara, J.; Fredrickson, J.; Kennedy, D.; Resch, C. T., Nitrate bioreduction in redox-variable low permeability sediments. Sci. Total Environ. 2016, 539, 185-195. 67. Roden, E. E., Microbiological controls on geochemical kinetics 1: fundamentals and case study on microbial Fe(III) oxide reduction. In Kinetics of Water-Rock Interaction, Springer: 2008; pp 335-415. 68. Liu, C.; Kota, S.; Zachara, J. M.; Fredrickson, J. K.; Brinkman, C. K., Kinetic analysis of the bacterial reduction of goethite. Environ. Sci. Technol. 2001, 35, (12), 2482-2490.

637


28

Page 29 of 34 Environmental Science & Technology

t=0h FH reduction


t=237h

FS (TOP) Org. acids [mM]

20

QS Environmental Science & Technology (a) (b)

FS (BOT) Page 30 of 34 (c)

15 Lac Ac

10 5

DNRA sp. [uM]

0 2000 (d)

1600 1200

NO -3

800

NO -2

(f)

(h)

(i)

400 0 1200 Fe2+

1000 Fe 2+ [uM]

(e)

Fe

(g)

2+

800 600 400 200 0 0

50

100 150 Time [hours]

200

ACS Paragon Plus Environment 0 50 100 150 200

Time [hours]

0

50

100 150 Time [hours]

200

Imaginary Conductivity σ′′ [S/m]

Page 31 of 34

6

×10 -4 1h 3Technology Environmental Science & (b) 13h 25h 30h 51h 72h 90h 115h 140h 162h 172h 210h 237h

×10 -4 (a)

4

2

2 1 0

×10 -4 3 (c) 2 1

0 0.01

0.1

Paragon Plus 1 10 ACS100 1000Environment 0 Frequency [Hz] 0.01 0.1

1 10 Frequency [Hz]

100

×10 7 (a)

4.25

sim-Bio SIP-σ′′

3 1.75

0 Biomass [Cells/mL-PM]

×10 -4 5.5

FS (TOP)

σ ′′ [S/m]

9 8 7 6 5 4 3 2

Page 32 of 34

9 8 7 6 5 4 3 2

50

×10 7 (b)

100 QS

150

200 ×10 -4 4.5 3.75 3 2.25

0

50

1.5 sim-Bio 0.75 SIP-σ′′ 0 100 150 200 Time [h]


σ ′′ [S/m]

Biomass [Cells/mL-PM]



Rate [Cells/mL-PM/h]

Page 33 of 34

10

×10 4

FS (TOP/BOT)

10 (a)

7.5

5

2.5

2.5

DNRA1 DNRA2 FHRED DECAY

(b)

0 0

30

60

90 120 150 180 210 240

0.12 Cole-Cole τ [s]

QS

7.5

5

0

×10 4

0

30

60

90 120 150 180 210 240

0.12 (c)

0.09

(d) 0.09

Top

Mid

0.06

0.06

0.03

0.03

0

0 0

30

60

90 120 150 180 210 240 Time [h]

0

30


60

90 120 150 180 210 240 Time [h]

Polarization of bacteria: SIP

DOC Page 34 of 34 Environmental Science & Technology 𝝈′′ / 𝒎 𝒏

charge storage

E(ω)

TEA Energy

σ'' ∝ Growth

ACSionParagon Plus Environment back-diffusion 𝝉=

𝒅2 [𝑠] 8𝐷𝑠

τ∝d t

Linking Spectral Induced Polarization (SIP) and Subsurface Microbial

Recommend Documents