Effect of Initial Speciation of Copper- and Silver-Based Nanoparticles

Subscriber access provided by UNIVERSITY OF ADELAIDE LIBRARIES

Article

Effect of Initial Speciation of Copper- and Silver-based Nanoparticles on their Long-term Fate and Phytoavailability in Freshwater Wetland Mesocosms John P Stegemeier, Astrid Avellan, and Gregory V. Lowry Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b02972 • Publication Date (Web): 10 Oct 2017 Downloaded from http://pubs.acs.org on October 11, 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

1

Effect of Initial Speciation of Copper- and Silver-based

2

Nanoparticles on their Long-term Fate and

3

Phytoavailability in Freshwater Wetland Mesocosms

4

John P. Stegemeier, 1,2 Astrid Avellan,1,2 Gregory V. Lowry 1,2*

5 6 7 8

1 Center for the Environmental Implications of NanoTechnology (CEINT)

9

2 Civil & Environmental Engineering, Carnegie Mellon University, Pittsburgh, PA 15213, United States

10 11

*corresponding author: Gregory V. Lowry, [email protected], +1 (412) 268-2948

12

Keywords: Heavy metals, phytoavailability, nanoparticle uptake, mesocosm, nanoparticle fate,

13

environmental nanotechnology

14

0

ACS Paragon Plus Environment


15

TOC Art

16

Rapid

Ag0 NPs

Rapid

Ag2S Location:

Slow

Egeria Sediments

CuO NPs CuS NPs Cu(NO3)2

Cu-Organic Matter Rapid

(Cycling)

Cu-Thiol

17 18

1


Page 2 of 29

Page 3 of 29

19


Abstract

20

Ag0 and CuO engineered nanomaterials (ENMs), or their sulfidized forms are introduced

21

into freshwater wetlands through wastewater effluent and agricultural runoff. Knowledge about

22

the rates of transformations of these ENMs in realistic environments, and the impact of the form

23

of the incoming ENM (i.e. sulfidized or pristine) on bioavailability and fate is limited. Here, five

24

freshwater wetland mesocosms were exposed to 3g of total metal as CuO, CuS, Ag0 or Ag2S

25

ENMs, or soluble CuNO3, added weekly for one month. Total metal and metal speciation was

26

measured in sediment and plant samples collected one, three, six and nine months after addition.

27

The form of the added ENM did not affect the metal distribution, and ENMs distributed similarly

28

to added ionic Cu or Ag. For the dosing condition used, ˜50% of the added Ag or Cu metal mass

29

was found in Egeria densa plant tissue, with the remainder primarily in the surficial sediment.

30

Ag0 and CuO ENMs transformed quickly in sediment, with no evidence of CuO and only ~4% of

31

silver present as Ag0 ENM one week after the last ENM addition. In contrast to sediment, Ag0

32

and CuO ENMs were persistent in E. densa tissues for up to 9 months and 6 months,

33

respectively. The persistence of ENMs in E. densa suggests that chronic exposure to both the

34

transformed and initially added ENMs is possible.

35

1. Introduction

36

Engineered nanomaterials (ENM) may be inadvertently released into fresh water

37

ecosystems from a variety of sources,1 e.g. in waste water treatment plant (WWTP) effluent2, 3 or

38

run-off from agricultural fields containing nano-enabled fertilizers and pesticides.4-6

39

Understanding the fate of these ENMs in freshwater ecosystems is essential for remediation

40

efforts, fate modeling and forecasting risk.7-9 2



41

Page 4 of 29

Silver and copper based nanoparticles are common ENMs in consumer products and

42

nano-enabled agricultural amendments (e.g. fungicides and pesticides).4,

6, 10

43

soluble and chemically stable ENM counterparts (TiO2, CeOx, SiO2, AlOx and Au), silver and

44

copper based ENMs can readily transform.11-13 The speciation of the transformed ENMs will

45

depend on the environmental conditions, e.g. redox state, pH, chemical composition, organic

46

carbon content.14-16 ENMs may enter wetlands in either a relatively pristine form, or after

47

significant transformation has occurred. For example, ENMs in biosolids applied as fertilizer will

48

likely contain metal sulfides or metal phosphates rather than the initial metallic or metal oxide

49

forms used in products.17-19 In contrast, agrochemicals such as CuO ENMs used as fungicides

50

may enter wetlands in their “as applied” form in agricultural run-off. Therefore, it is important to

51

understand the transformation rates and the fate of both “as applied” and transformed

52

nanomaterials in wetlands.

Unlike their less

53

The speciation or form of the ENM controls their solubility, reactivity, bioavailability,

54

and environmental fate.20-23 For example, sulfidation was shown to reduce their toxicity of Ag

55

ENMs towards several aquatic species24-26, as well as the cytotoxicity of CuO ENMs.27 However,

56

the sulfidation of CuO ENMs increased their toxicity to zebrafish due to the formation of very

57

small (~a few nm) CuS particles that are susceptible to oxidative dissolution.14, 28 Despite the low

58

solubility of silver sulfide, these ENMs have been found to be bioavailable to terrestrial20, 21 and

59

aquatic plants29. Although sulfidation is expected to be an important transformation for both

60

copper and silver nanoparticles,3,

61

transformations in a complex environment is needed to accurately determine their potential for

62

ecological risk.

30, 31

a better understanding of the time scale and types of

3


Page 5 of 29


63

Aquatic plants serve a particularly important role in ecosystems, influencing physical

64

(light penetration, temperature and hydrodynamics), chemical (dissolved oxygen, organic carbon

65

and nutrient levels) and physiological aspects of freshwater systems.32 Macrophytes are one of

66

the major species found in the water column of a wetland ecosystem. Often, these environments

67

are dominated by a single vascular plant species33. Nutrient absorption and metal uptake and

68

storage by aquatic plants are important for heavy metal removal from the water column34, so they

69

will likely also have an important effect on ENM fate and introduction into the food web.

70

Although some studies have investigated the interactions between selected ENMs and aquatic

71

plants,26,

72

natural waters40, 41 on their uptake into aquatic plants in natural freshwater systems is not well

73

characterized.21,

74

poorly understood.39 In-vivo transformations of internalized ENM are very poorly characterized,

75

even though they would impact the long-term fate of ENM, their effects on plant health, 42-45 and

76

determine the species of ENM that organisms feeding on plants would be exposed to.

29, 35-39

many knowledge gaps still exist. The impact of transformations of ENM in

26

The mechanisms driving ENM internalization by aquatic plants remains

77

Mesocosm experiments are a bridge between “clean” laboratory experiments and realistic

78

environmental exposures. They incorporate the complexity of aquatic organisms and microbial

79

activity while providing realistic environmental exposure conditions (sunlight, seasonal and daily

80

temperature cycling) and therefore provide insights into behaviors expected in natural systems 46.

81

Recently, we demonstrated that CuO and Ag ENMs affected sediment microorganisms in

82

freshwater wetland mesocosms differently than CuS and Ag2S ENMs in the first three months of

83

exposure, but by 300 days after addition there was little difference in the microbial community

84

structure between treatments47. This suggests that ENM transformation had occurred fairly

4



85

rapidly, i.e. in the first few months after addition or sooner, but the time scales for transformation

86

were not reported in that study.

87

The main goals of the present study are to (1) determine the effect of initial ENM

88

speciation on their persistence and the timescales of their transformations, and (2) identify the

89

impact of the initial speciation of the ENM on uptake of ENMs by the aquatic plant, E. densa by

90

determining the metal uptake and speciation in the plants over time. Five mesocosms were

91

amended with Ag, Ag2S, CuO or CuS ENMs to assess if initial speciation affects fate, behavior,

92

and biouptake by E. densa over nine months. Cu ENM speciation and fate was also compared to

93

added ionic Cu species.

94

2. Materials and Methods

95

2.1 Synthesis of ENMs

96

Gum arabic coated silver and silver sulfide nanoparticles were obtained from CEINT’s

97

ENM synthesis center, and added to the mesocosms located in the Duke Forest. The synthesis

98

and characterization of these particles are provided in Yin et al., 2011.48 Multiple batches were

99

combined, and the nanoparticles were purified and concentrated by dialysis (Optiflux F200NR

100

Fresenius Polysulfone Dialyzer, Fresenius Medical Care).

101

Copper oxide ENMs (1000 ppm). This

251

concentration is in the range reported for E. densa which can accumulate copper58. Since the

252

plant tissues collected and analyzed were from the youngest stems, the Cu apparently remained

253

bioavailable to E. densa for the duration of the experiment and/or was mobile in the plant tissue

254

throughout the nine month study. However, it cannot be ruled out that reduced growth rates due

255

to the added Cu may have resulted in sampling sections at later time points which were originally

256

exposed during the Cu addition.

257

A significant mass of the added Cu became associated with the E. densa plants. Using the

258

measured average Cu concentrations of ~3000mg/kg (dry weight) of the Cu-based nanoparticle

259

treatments, the average of the total dry mass of E. densa from 15 similar mesocosms after 9

260

months (1208 ±222 g dry weight/m3), and the volume of the mesocosm (0.4m3), we estimate that

261

1.4 ±0.2 g, or ~50 ±15% of the total copper added as ENMs was associated with plant tissue. The

262

number is higher for the Cu(NO3)2 amendment. Thus, it appears that E. densa accumulates

263

metals from the water column or sediment regardless of whether or not they are introduced as

264

dissolved or nanoparticulate species. These measurements of total Cu cannot distinguish if

265

uptake was as dissolved Cu or Cu ENMs, but X-ray analysis of plant tissues (discussed later in

266

the paper) suggests uptake of both dissolved metals and ENMs.

13


Page 15 of 29


267

Similar trends were found in the silver concentrations in the surficial sediments and dried

268

E. densa tissues exposed to Ag2S and Ag0 ENMs (Figures 1C and 1D). Similar to Cu, there is

269

about 50% of the added metals ending being associated to E. densa tissues after 9 months, with a

270

statistically significant (students t-test) gradual decline in the Ag concentration over time in the

271

surficial sediment. The silver concentration in the E. densa plant tissue also remained fairly

272

constant throughout the experiment for both exposures. At all of the time points sampled, the

273

Ag0 ENM exposed samples contained more Ag than the Ag2S ENM exposed plants, indicating

274

greater bioavailability. The elevated concentrations of Ag in new growth of the plants over the

275

course of the study indicate that the Ag remains bioavailable during the 9 month experiment. A

276

prior study in similar mesocosms also indicated a continued bioavailability of Ag to aquatic

277

plants despite partial transformation to Ag2S ENMs23.

278

3.4 Cu Speciation in Sediments

279

A summary of the Cu EXAFS fitting for the surficial sediments are shown in Table 1.

280

EXAFS fits are provided in Figure 2. Several important conclusions can be drawn from the Cu

281

speciation over time in the three different treatments, CuO, CuS, and Cu(NO3)2. First, the

282

sediment exposed to CuO or CuS ENMs did not show the presence of the initial ENM material

283

in any sample, indicating that they were transformed in the sediment on a timescale less than one

284

week as this was the time that had elapsed between the last dose and the first sediment samples

285

collected. This is consistent with the relatively rapid dissolution of CuO ENMs (t1/2~70h) in the

286

mesocosm water column under quiescent conditions,49 and the ability of these CuS ENMs to

287

undergo oxidative dissolution14. Second, the speciation of Cu in both of the ENM treatments

288

(CuO and CuS ENMs) is similar, comprised of a mixture of Cu-S-R bound (organic Cu-sulfides) 14



Page 16 of 29

289

and Cu-O-R bound (Cu bound to organic matter) copper, regardless of the initial form of Cu

290

ENM added. However, there was a greater tendency for formation of Cu-S-R species in the

291

ENM amended mesocosms compared to the Cu(NO3)2 amended mesocosm, which had Cu-O-R

292

representing the majority of Cu speciation. This suggests that added Cu ions are more readily

293

ligated by NOM constituents that control their fate, whereas the slower transformation of Cu

294

ENMs leads to slightly different fate processes. Third, the addition of either CuO or CuS ENMs

295

resulted in formation of low amounts of metallic copper found in some samples. No evidence of

296

metallic Cu was found in the mesocosm amended with Cu(NO3)2. The reason why metallic Cu

297

was formed in the ENM amended mesocosm and not the Cu(NO)3 amended mesocosm is

298

unclear, but it may result from a difference in the availability and distribution of the metal. The

299

CuO ENMs dissolve over days to a week49, releasing Cu in a localized reducing environment

300

with NOM, making reduction to metallic Cu possible.59 The CuS dissolution rate was not

301

measured in mesocosm water, but a similar argument can be made for these particles. In contrast,

302

added as a ionic species, Cu disperses more readily and is immediately bound by natural organic

303

matter, preventing localized free Cu concentrations that are high enough to be reduced to

304

metallic Cu.

305

Interestingly, the Cu in collected sediments for all three treatments was found to be 100% Cu-O-

306

R at the winter sampling point (3m). The predominance of Cu-O-R in surficial sediment during

307

winter is consistent with the relatively more oxidizing environment during this time period. The

308

predominance of the Cu-O-R is likely from the large influx of organic matter from the

309

senescence prior to winter. It is important to note that this speciation is for copper present in the

310

surficial sediment only, and some CuS is likely to present in the deeper sediments given the rapid

311

reduction in oxygen saturation during the first 1.5 centimeter of sediment at that sampling time 15


Page 17 of 29


312

(Figure S8). Although the sampling technique was designed to collect the top layer of sediment

313

and be consistent with sampling depth and amount of sediment collected, the amount of

314

heterogeneity this type of sampling introduced could not be accurately quantified.

315 316 317 318 319

Table 1. Linear combination fitting results of k3-weighted Cu EXAFS spectra (Figure 2) for surficial sediment and Egeria densa samples fit over k-range 2 - 8.5 Å. The percentages have ±15% uncertainties. Presented with the R-factor (Rf) and the reduced χ2 parameters to indicate the high quality of the fits Matrix Model compound

320 321

E. densa

Sediments Cu-O-R Cu-S-R (%) (%)

Cu0 (%)

Sum (%)

Rf

CuO Red Cu-O-R ENM χ2 (%) (%)

CuS (%)

Cu-S-R Sum (%) (%)

Rf

Red χ2

CuO ENM 1 month 3 month 6 month 9 month

39 103 41 44

41 33 39

23 24 -

103 103 99 83

0.08 0.14 0.19 0.15

0.41 0.93 1.43 0.26

53 44 53 -

44 44 44 83

-

-

97 88 97 83

0.01 0.01 0.01 0.01

0.04 0.02 0.04 0.04

CuS ENM 1 month 3 month 6 month 9 month

24 103 91

99 117 -

8 -

130 103 117 91

0.03 0.08 0.07 0.07

0.25 0.41 0.35 0.48

-

80 32 69 75

62 -

33 28

80 94 102 103

0.04 0.04 0.02 0.01

0.11 0.06 0.04 0.04

Cu(NO 3 ) 2 1 month 3 month 6 month 9 month

106 109 101 110

-

-

106 109 109 110

0.01 0.05 0.01 0.02

0.07 0.29 0.06 0.13

a

100 102 106 a

a

a

100 102 106 a

0.01 0.01 0.01 a

0.02 0.03 0.05 a

a. no E. densa remaining after 9 months.

16



322 323 324 325

Figure 2. Cu EXAFS spectra (black) and linear combination fits (red) for CuO exposed sediment (top left) and plant tissue (top right), CuS ENM exposed sediment (middle left) and plant tissues (middle right) and Cu(NO3)2 exposed sediment (bottom left) and plant tissues (bottom right).

326 327

17


Page 18 of 29

Page 19 of 29


328 329

3.5 Cu Speciation in Egeria densa

330

The copper speciation in the E. densa plant tissue exposed to CuO ENMs persisted as

331

CuO (tenorite) for up to six months (Table 1). The Cu speciation one month after the initial dose

332

was a mixture of the added CuO ENMs and Cu bound to organic matter (Cu-O-R), suggesting

333

that the CuO ENMs may be directly taken up into the plant tissue and protected from dissolution.

334

Dissolution of the CuO ENMs was expected based on their absence in sediment and their

335

relatively high dissolution rate in mesocosm water49. The persistence of CuO samples collected 6

336

months after adding CuO ENMs suggests a relatively slow in vivo transformation/dissolution of

337

those ENMs, but its absence could also be a result of sample heterogeneity. Regardless, the

338

persistence of the CuO ENM signal associated with E. densa tissues over 6 months indicates that

339

they can persist adhered to or inside plant tissues even though they are readily transformed in the

340

sediment compartment of the mesocosm.

341

In contrast to CuO ENMs, the copper speciation in the E. densa exposed to CuS ENMs

342

found limited evidence of the added CuS ENM phase. The majority of Cu was either Cu-O-R or

343

Cu-S-R. In only one sample was an inorganic CuS phase observed, and it better matched a

344

natural covellite model (CuS) than the CuS ENM model compound. This suggests that either the

345

CuS ENM are not associated with the plant tissues, or that any CuS ENM that have associated

346

with the plant tissue are transformed to become associated with O- and S- organic ligands. The

347

generally greater amount of S-associated Cu associated with plants exposed to CuS ENMs as

348

compared to either CuO ENMs or Cu(NO3)2could be a result of uptake of CuS ENMs (i.e. co-

18



349

exposure to Cu and S), or potentially because the plants natural defense, e.g. glutathione, is more

350

able to cope with the CuS ENM than either the CuO or Cu(NO3)2.

Page 20 of 29

351

The Cu(NO3)2 exposed plant tissue had a similar Cu speciation at all times, showing Cu

352

primarily bound to O-ligands. In contrast to added CuS ENMs, no organic Cu-S-R phase was

353

found by Cu EXAFS fitting for plants when Cu(NO3)2 was used as the copper source. This may

354

be a result of Cu toxicity overwhelming the plant defense response to heavy metals, lowering the

355

ability to form Cu-S-R phases.

356

3.6 Ag Speciation in Sediments

357

The Ag speciation for the surficial sediment and E. densa samples exposed to Ag0 and

358

Ag2S ENMs is summarized in Table 2 (XANES spectra and fit are provided in Figure S9). The

359

Ag speciation in the surficial sediment exposed to Ag0 ENMs one week after the final dose is a

360

mixture of Ag2S and Ag bound to thiol with a very small fraction (4%) remaining as metallic Ag,

361

indicating that Ag0 ENMs transform into silver sulfide in the surficial sediment in less than 1-4

362

weeks after addition. This transformation to silver sulfide is more rapid than previously reported,

363

with other studies showing up to 15% of added Ag ENMs remaining as metallic silver after 28

364

days in marine sediments60 while another study shows 18% remaining as metallic Ag after 18

365

months in a similar freshwater wetland.23 The short timescale for transformation observed here

366

may be attributed to the small size of the added Ag0 ENMs compared to these previous studies.

367

The Ag speciation in sediment remains fairly consistent over time, with the primary species

368

being Ag2S and a minor Ag-S-R (Ag-thiol) species present. The overall result of silver rapidly

369

transforming and stabilizing as silver sulfide is consistent with other studies using soils or

370

freshwater sediments under reducing conditions11,

61,

62

. However, there is a gradual

19


Page 21 of 29


371

transformation from Ag2S to more Ag-S-R like character in the surficial sediments, suggesting

372

that a slow natural transition of the Ag2S phase to a potentially more bioavailable organic Ag

373

phase (Ag-thiol) is occurring. This could explain why new growth of E. densa continued to

374

contain elevated levels of silver.

375

376

3.7 Ag Speciation in Egeria densa

377

The Ag speciation in the tissue of E. densa plants in mesocosms added with Ag0 ENMs

378

was a mixture of Ag0, Ag2S and Ag bound to thiol suggesting either a portion the Ag0 ENMs

379

were sorbed into/onto the plant tissue and persisting as ENMs, or that the Ag ENMs were

380

dissolving and undergoing ion-uptake followed by reductive precipitation of Ag0 ENM which

381

has been shown to occur in many other plants exposed to ionic Ag.63, 64 Even though a recent

382

study has shown the reduction of Ag2S to Ag0,65 we do not believe this is the main mechanism

383

explaining the persistence of Ag0 in our system, because no metallic Ag was detected in plants

384

collected from in the Ag2S ENM amended mesocosm. The Ag speciation in plant tissue remains

385

fairly consistent throughout the study suggesting the E. densa is protecting the Ag0 in the tissues

386

from becoming sulfidized on a long term basis or that there is a pseudo-steady state achieved

387

between the dissolved and nanoparticulate silver species within the plants.

388

The Ag speciation in the Egeria densa tissue exposed to Ag2S ENMs for all sample times

389

is quite similar to that for Ag0. However, there is a greater fraction of Ag present as Ag2S in

390

plants collected from mesocosms that were added with Ag2S ENMs, and there is no evidence of

391

elemental silver. Since many plant species are known to synthesize metallic silver ENMs after 20



Page 22 of 29

392

being exposed to silver ions, the absence of Ag0 in the samples exposed to Ag2S ENMs is

393

consistent with the lower ion release expected for Ag2S ENM compared to Ag0 ENMs.

394

395 396 397 398 399

Table 2. Linear combination fitting results of Ag XANES spectra for surficial sediment (left) and Egeria densa tissue (right) samples fit over -25 to +100 eV with respect to the Ag Kα edge. The percentages have ±15% uncertainties. All of the Rf and reduced χ2 (goodness-of-fit) parameters were in the 10-5 and 10-3 range, respectively. Matrix Model Compound

400

E. densa

Sediments Metallic Ag (%)

Ag2S

Ag-Thiol Sum (%)

(%)

(%)

Metallic Ag (%)

Ag2S

Ag-Thiol Sum (%)

(%)

(%)

Ag 0 ENM 1 month 3 month 6 month 9 month

4 -

97 101 82 72

19 28

101 101 101 100

17 11 23

47 89 70 62

37 30 15

101 100 100 100

Ag 2 S ENM 1 month 3 month 6 month 9 month

-

84 95 87 89

16 6 12 12

100 101 99 101

-

83 90 83 85

18 10 18 20

101 100 101 99

401 402

4. Implications

403

The findings of this case study have many important implications concerning the fate of

404

ENMs in natural wetlands and freshwater ecosystems. Even though the concentration of added

405

ENMs are relatively high (target of 100 mg/kg in the top layer of the sediment), this study give

406

us very important insights on Cu- and Ag- based ENM transformations in a stressed

407

environment. First, both CuO ENMs and Ag0 ENMs that find their way into the sediment 21


Page 23 of 29


408

compartment transform very rapidly after introduction (

Effect of Initial Speciation of Copper- and Silver-Based Nanoparticles

Recommend Documents