Transformation of Silver Nanoparticles in Sewage ... - ACS Publications

Subscriber access provided by UNIV OSNABRUECK

Article

Transformation of Silver Nanoparticles in Sewage Sludge during Incineration Christoph Meier, Andreas Voegelin, Ana Elena Pradas del Real, Géraldine Sarret, Christoph Rüdiger Mueller, and Ralf Kaegi Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.5b04804 • Publication Date (Web): 03 Feb 2016 Downloaded from http://pubs.acs.org on February 4, 2016

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 26

Environmental Science & Technology

Transformation of Silver Nanoparticles in Sewage Sludge during Incineration Christoph Meier,† Andreas Voegelin,‡ Ana Pradas del Real,¶ Geraldine Sarret,¶ Christoph R. Mueller,§ and Ralf Kaegi∗,‡ 1

†Zhaw, Zurich University of Applied Sciences, Winterthur, Switzerland ‡Eawag, Swiss Federal Institute of Aquatic Science and Technology, Duebendorf, Switzerland ¶ISTerre (Institut des Sciences de la Terre), Universite Grenoble Alpes and CNRS, France §ETH Zurich, Laboratory of Energy Science and Engineering, Zuerich, Switzerland E-mail: [email protected] Phone: +41(0) 58 765 5273. Fax: +41 (0)58 765 5802

2

Abstract

3

Silver nanoparticles (Ag-NP) discharged into the municipal sewer system are ac-

4

cumulated in the sewage sludge. Incineration and agricultural use are currently the

5

most important strategies for sewage sludge management. Thus, the behavior of

6

Ag-NP during sewage sludge incineration is essential for a comprehensive life cy-

7

cle analysis and a more complete understanding of the fate of Ag-NP in the (urban)

8

environment. We therefore spiked metallic Ag0 -NP to a pilot waste water treatment

9

plant and digested the sludge anaerobically. The sludge was then incinerated on a

10

bench-scale fluidized bed reactor in a series of experiments under variable conditions.

11

Complementary results from X-ray absorption spectroscopy (XAS) and electron mi-

12

croscopy - energy dispersive X-ray (EM-EDX) analysis revealed that Ag0 -NP trans-

13

formed into Ag2 S–NP during the wastewater treatment, in agreement with previous

1

ACS Paragon Plus Environment


14

studies. Based on a principal component analysis and subsequent target testing of the

15

XAS spectra, Ag0 was identified as a major Ag component in the ashes but Ag2 S was

16

clearly absent. The re-formation of Ag0 -NP was confirmed by EM-EDX. The fraction

17

of Ag0 of the total Ag in the ashes was quantified by linear combination fitting (LCF)

18

of XAS spectra and values as high as 0.8 were found in sewage sludge incinerated at

19

800 ◦C in a synthetic flue gas atmosphere. Low LCF totals (72 % to 94 %) indicated at

20

least one missing reference spectra. The presence of spherical Ag-NP with a diame-

21

ter < 50 nm extending into the sub-nm range was revealed from electron microscopy

22

analyses. Size effects resulting in increased percentages of surface Ag atoms and dis-

23

torted Ag lattices may result in experimental XAS spectra slightly different to the XAS

24

spectra of the pure reference compounds used for the LCF analyses and these differ-

25

ences likely explain the low totals.

26

Introduction

27

In this work the term silver nanoparticles (Ag-NP) refers to any silver containing nanopar-

28

ticles. Where relevant we explicitly use the terms Ag0 -NP or Ag2 S-NP to refer to metallic

29

or sulfidic Ag-NP.

30

Colloidal silver has a century-long record as biocide. 1 In recent years, Ag-NP have

31

been introduced in an increasing number of consumer products such as cosmetics and

32

textiles. 2 The biocidal efficacy of Ag greatly varies among different silver compounds

33

and strongly depends on solubility of the respective silver compounds. 3 However, in

34

addition to the toxicity caused be the Ag+ ions, a nanoparticle-specific toxicity is currently

35

discussed. 4–9 Results from mass flow analyses indicated that the majority of Ag-NP used

36

in consumer products will be released into urban wastewater systems, including sewers

37

and wastewater treatment plants (WWTP). 10,11 Consequently, the fate and transformation

38

of Ag-NP in urban wastewater systems is of key importance to assess the risks associated

39

with the increased use of Ag-NP. 4,12–17 2


Page 2 of 26

Page 3 of 26


40

Full and lab scale experiments revealed that the sulfidation of metallic Ag-NP to spar-

41

ingly soluble silver sulfide (Ag2 S) occurs within typical retention times in urban wastew-

42

ater systems, 18–20 which strongly mitigates the toxicity of released Ag-NP. 17,21 A high NP

43

removal efficiency (mostly > 95 %) during wastewater treatment has been reported, 14,22–25

44

which in turn results in an accumulation of nanoscale Ag2 S (transformed Ag-NP) in the

45

sewage sludge. 22

46

Currently, incineration and agricultural use are the main routes for sewage sludge dis-

47

posal. 26 In the United States, about 15 % of the total amount of sewage sludge is inciner-

48

ated. 27 In the EU, about one third of the sewage sludge is incinerated, but large variations

49

exist among the different member states. 28 In Switzerland, sewage sludge is exclusively

50

incinerated as the agricultural use of biosolids is prohibited since 2008, and the German

51

federal environment agency (Umweltbundesamt, UBA) recommends to ban the agricul-

52

tural use of sewage sludge in Germany within the next 10 -20 years. 28 Accordingly, it is

53

expected that the incineration of sewage sludge will become increasingly important and

54

may reach a share of almost 40 % in the old EU member states (EU-15) by 2020. 29

55

The co-combustion in municipal waste incineration facilities is feasible, but an increas-

56

ing pressure to recycle phosphorus (P) from sewage sludge ashes favors the incinera-

57

tion in mono-combustion facilities, were exclusively sewage sludge is incinerated. The

58

two major incinerator designs for sewage sludge are the multiple hearth and the flu-

59

idized bed reactors. Due to techno-economical advantages of fluidized-bed reactors over

60

other reactor types, fluidized-beds are most commonly used for sewage sludge mono-

61

combustion. 26,30 In Germany, for example, 19 out of 26 mono-combustion facilities are

62

fluidized bed reactors. 31

63

Despite the increasing use and related release of Ag-NP into wastewater systems and

64

the increasing amounts of sewage sludge incinerated in fluidized bed reactors, the behav-

65

ior of the Ag-NP during the incineration is only poorly understood.

66

Based on incineration experiments conducted in a muffle oven, Impellitteri et al. 12 3



67

68

concluded that 30 % to 50 % of the Ag2 S present in the sewage sludge was converted into Ag0 during the incineration. The presence of metallic Ag is consistent with the equi-

69

librium phase diagram of the Ag-S-O system at typical incineration conditions. Ag2 S is

70

unstable in an oxidizing atmosphere 32 and the formation of Ag2 O and AgO is not fa-

71

vored at elevated temperatures. 33 However, equilibrium phase diagrams of simplified

72

systems may be applicable to complex multi-component systems such as sewage sludge

73

ash only to a limited extend. The goal of the present study was, therefore, to investigate

74

the morphological and chemical changes of Ag-NP along their path through managed

75

waste facilities with a special focus on the incineration in a fluidized bed reactor.

76

For that purpose, Ag0 -NP were spiked to a pilot WWTP and the digested sludge was

77

incinerated in a bench-scale fluidized bed reactor operated at typical incineration condi-

78

tions. Transformed Ag-NP in the digested sewage sludge and in the sewage sludge ash

79

were investigated using scanning and scanning transmission electron microscopy (SEM

80

and STEM) and the speciation of Ag in selected samples was assessed by X-ray absorption

81

spectroscopy (XAS).

82

Materials and methods

83

The starting materials, the WWTP and the Ag-NP spiking protocol were the same as

84

described by Pradas del Real et al. 34 More detailed information can be found therein and

85

in the SI.

86

Characterization of the Ag-NP PVP coated Ag-NP (NanoAmor, Nanostructured and

87

Amorphous Materials, Inc. Housten, TX, USA) were used for spiking experiments of the

88

WWTP. Dynamic light scattering (DLS) measurements conducted on Ag0 -NP suspended

89

in doubly deionized (DDI) water indicated an average particle size of 61 +/- 10 nm. This

90

corresponded well with results from transmission electron microscopy (TEM) analysis of

91

the particles, which indicated a diameter of 47 +/- 7 nm for spherical (70 %) and maxi4


Page 4 of 26

Page 5 of 26


92

mum and minimum Feret diameters of 64 +/-3 nm x 26 +/-2 nm for elongated particles

93

(30 %). The smaller coherently scattering domain size of 23 +/- 3 nm derived from X-ray

94

diffraction analysis suggested that the Ag-NP were polycrystalline.

95

Preparation of sewage sludge spiked with Ag-NP

96

sludge was produced by spiking an Ag-NP suspension to a pilot WWTP (nitrification and

97

denitrification reactor, primary and secondary clarifier followed by anaerobic digestion).

98

Ag-NP were added continuously to the denitrification reactor. To simulate Ag-NP en-

99

tering the digester via the primary sludge, Ag-NP were spiked to primary sludge once

100

per day before primary and activated sludge were mixed in the thickener and subse-

101

quently delivered to the anaerobic digester (see Figure S1). To setup the spiking protocol,

102

the WWTP was modeled as a continuously stirred tank reactors (CSTR) as described by

103

Kaegi et al (2011) 22 but with an additional CSTR representing the anaerobic digester. The

104

Ag-NP spike protocol was designed to achieve an Ag concentration of 400 mg kg−1 total

105

suspended solid (TSS) in the digested sludge. This Ag concentration allowed us to clearly

106

identify Ag-NP in the sludge using microscopic methods. It is about one log unit above

107

Ag concentrations in sludge from field scale WWTP, although Ag concentrations up to

108

850 mg kg−1 TSS have been reported. 35 The detailed spiking protocol is described in the

109

SI. The background concentration of Ag in the sewage sludge, determined on samples

110

samples collected before the spiking procedure was 14 mg kg−1 TSS.

111

Bench-scale fluidized bed reactor and muffle oven The dried sludge was incinerated

112

in an electrically heated bench-scale fluidized bed reactor operated in batch mode (Fig-

113

ure S2). The reactor consisted of a 24 mm inner diameter quartz glass tube with an inte-

114

grated porous foam distributor plate, jacketed by a round tube furnace (Carbolite, UK).

115

The fluidizing gas was manually controlled by a rotameter (Aalborg, USA). In each run,

116

40 g of Al2 O3 (300 µm to 450 µm) were used as bed material. The incineration tempera-

117

ture was approached and held by an external control circuit. An N-type thermocouple 5

Ag-NP containing digested sewage



118

was inserted from the open top of the reactor tube and provided the feedback to the heat

119

load control. The tip of the thermocouple was placed 4 cm above the distributor plate

120

and slightly below the bed surface. When a steady bed temperature was reached, 1 g of

121

granulated sludge that had previously been ground and sieved to a fraction of 300 µm

122

to 450 µm was inserted into the reactor. Devolatilization was completed within no more

123

than 15 s. Ash particles were collected in an air filter cartridge trough a stainless steel tube

124

mounted above the open top of the reactor.

125

We used a synthetic gas mixture mimicking the incineration atmosphere in full scale

126

mono-combustion reactors. This synthetic gas mixture contained 4 % O2 , 12 % CO2 and

127

300 ppm SO2 in addition to N2 and was humidified to a dew point temperature of about

128

70 ◦C. This gas mixture is comparable to the flue gas compositions presented by Hart-

129

mann et al. 36 and Ogada et al. 37 and is thus considered as representative for mono-

130

combustion facilities. Both pressurized air as well as bottled nitrogen were used in addi-

131

tional experiments. The sand bed temperature in fluidized bed incinerators is typically

132

at 750 ◦C to 800 ◦C and the freeboard temperature (i.e., above the sand bed) is at about

133

850 ◦C, somewhat higher due to the heat released by the volatile burnout. 30 The reactor

134

temperature was set to 800 ◦C in most of the experiments, but additional runs were con-

135

ducted at 600 ◦C and 900 ◦C. The incineration time for solid fuel combustion comprising

136

drying, devolatilization and residual char combustion is dependent on the particle size

137

and the initial moisture content. Urciuolo et al. 38 reported incineration times of a few

138

minutes for cm-sized sludge particles. Accordingly, ash samples were extracted after 1

139

min, 10 min and 2 h.

140

For comparison, further experiments were conducted in a muffle oven (Carbolite, UK).

141

For that purpose, dried sludge samples were placed in a porcelain crucible and incin-

142

erated at 700 ◦C, 800 ◦C and 900 ◦C. The temperature was increased at a steady rate of

143

5 ◦C min−1 until the incineration temperature was reached and then kept constant for 2 h.

144

Samples were inserted 300 ◦C below the set point temperature (1 h before the final tem6


Page 6 of 26

Page 7 of 26


145

perature, considering the given temperature ramp) and directly removed from the hot

146

state after 2 h.

147

A list of all incineration experiments conducted is presented in Table 1. We further

148

refer to the ash samples by the number assigned in this table (e.g., Ash 5.1).

149

Electron microscopy

150

and a STEM. The SEM (NOVA NanoSEM230, FEI, USA) was operated at an acceleration

151

voltage of 20 keV and a backscattered electron (BSE) detector was used for image for-

152

mation. The presence of Ag was verified with an EDX (energy dispersive X-ray) system

153

(INCA 4.15, X MAX 80, Oxford, UK). For SEM analysis, samples were ground with a

154

mortar and pestle and sprinkled on a carbon pad.

Selected ash and sludge samples were investigated with a SEM

155

To investigate ash samples at higher resolutions, a STEM (HD2700Cs, Hitachi, Japan),

156

operated at an acceleration voltage of 200 keV was used. For image formation, a high-

157

angle annular dark-field (HAADF) and a secondary electron (SE) detector were used.

158

EDX signals were recorded with an EDX system (EDAX, USA) attached to the micro-

159

scope and the data was processed using Digital Micrograph (v 1.85, Gatan Inc, USA). For

160

STEM analyses, a few mg of ground sample material was sonicated in 5 ml isopropanol

161

for 15 min. After 2 h of settling, 0.1 ml of the supernatant was directly centrifuged onto

162

the TEM grid at 15 000 g for 1 h.

163

XAS data acquisition and analysis Sludge and ash samples were analyzed by XAS at

164

the Ag K-edge (25 514 eV) at the SuperXAS (X10DA) beamline at the Swiss Light Source

165

(SLS, Villigen, Switzerland). Samples were prepared as 7-mm pellets from about 50 mg

166

of dried sludge or ash material. Spectra were collected at room temperature in fluores-

167

cence mode using a five-element silicon drift detector (SDD). Typically, two scans per

168

sample were recorded. All data processing was performed using NumPy 39 and Larch. 40

169

Individual scans were rebinned onto a consistent energy grid and merged. Extraction of

170

the EXAFS signal relied on the autobk algorithm 41 implemented in Larch. The subse7



Page 8 of 26

−1

−1

171

quent data analysis was performed on k2 weighted EXAFS spectra from 2 Å

172

The number of relevant Ag spectral components in the ash samples was estimated by a

173

principle component analysis (PCA). Subsequently, target testing (TT) was used to iden-

174

tify Ag reference spectra that can be described by the spectral components that resulted

175

from the PCA analysis. These Ag reference spectra were then used to evaluate the sample

176

EXAFS spectra by linear combination fitting (LCF) using Larch. Further details on the

177

PCA-TT-LCF procedure are provided in the SI.

178

to 10 Å

.

The following reference spectra were considered for TT analysis: Ag-foil (Ag0 ) and

179

Ag2 SO4 (recorded at SLS, SuperXAS), Ag2 S, Ag-cysteine, Ag2 O, AgO and AgCl (recorded

180

at ESRF, DUBBLE) and Ag3 PO4 (distributed with the Demeter software package 42 ). All

181

spectra were recorded at room temperature.

182

Inductively Coupled Plasma - Optical Emission Spectrometry (ICP-OES)

183

50 mg to 60 mg of dried sludge or ash were digested in 1 mL HNO3 (65 %, ultrapure,

184

Merck KGaA, Germany) and 200 µL H2 O2 (35 %, Merck KGaA, Germany) and 0.5 mL HF

185

(48 %, suprapure, Carl Roth GmbH + Co. KG, Germany) using a microwave assisted acid

186

digestion system (UltraClave 3, MLS GmbH). Selected elements (Ti, Mn, Cu, Zn, Ag, Ba)

187

were measured with a ICP-OES (CirusCCD, SPECTRO Analytical Instruments GmbH,

188

D). Instrumental detection limit were 10 µg L−1 .

189

Results and discussion

190

Ag concentration in sewage sludge and sewage sludge ashes The average Ag con-

191

centration in the digested sludge was about 390 mg kg−1 TSS (Table S2) with a relative

192

standard deviation of 46 % and the total Ag concentrations in the ashes varied accord-

193

ingly (Table 1). This variations were less pronounced for other elements (Ti, Mn, Cu, Zn

194

and Ba, see Table S2) and are therefore likely caused by the spiking procedure.

8


Between

Page 9 of 26


SEM-BSE imaging revealed Ag-

195

Characterization of Ag and Ag-NP in the dried sludge

196

NP associated with the digested sewage sludge (Figure S3A, Ag-NP are visible as bright

197

spots). The Ag-NP primarily occurred as agglomerated structures of up to 1 µm in diam-

198

eter. We assume that Ag-NP homoaggregation already occurred in the stock suspension

199

used for Ag-NP dosing, as Ag-NP are expected to heteroaggregate with the biological

200

flocks after being introduced into the sewage sludge. SEM-EDX spectra obtained from ag-

201

glomerated Ag-NP in the sludge matrix revealed a close association of Ag and sulfur (S)

202

(Figure S3B). Although S was also present in the sludge matrix, S signal intensities were

203

substantially elevated in spectra recorded on Ag-NP, indicating the presence of an Ag-S

204

compound. These observations were confirmed by STEM-EDX analysis (Figure 2D and

205

F). Furthermore, the Ag K-edge EXAFS spectrum of the dried sludge closely matched the

206

Ag2 S reference spectrum (Figure S3C). This indicates that the Ag0 -NP became completely

207

sulfidized during waste water treatment and / or anaerobic digestion, in agreement with

208

recent studies on the transformation of Ag-NP in urban waste water systems. 14,22,43,44

209

Identification of Ag-NP in sewage sludge ashes by SEM and STEM

210

were first screened for the presence of Ag-NP using SEM. As in the sludge samples, the

211

Ag-NP appeared as bright spots in the BSE image, and their presence was confirmed by

212

EDX analysis (see next section). Four different types of Ag-NP were repeatedly observed

213

(Figure 1). Ag-NP of the first type were of comparable size and shape as those detected in

214

the dried sludge (Figure 1A, Ash 5.1). These Ag-NP were sensitive to the electron beam

215

and changed their morphology during the electron beam irradiation. The inset shows

216

the same sample location after 30 s of irradiation. Ag-NP of the second type had a well-

217

defined morphology and were not affected by the electron beam irradiation (Figure 1B,

218

Ash 7.1) The slaggy particle shapes suggest that temperatures were high enough to trigger

219

a phase transformation. As a third type, individual spherical Ag-NP of considerably

220

smaller sizes compared to the first two types were observed (Figure 1C, Ash 10). Ag-NP

9


The ash samples


221

of the fourth type were represented by ’sprinkles’ (Figure 1D, again Ash 7.1) with sizes

222

probably extending well below the size detection limit of the BSE imaging mode of the

223

SEM, which was in the order of a few tens of nm.

224

To investigate selected ash samples in more detail and and to get an indication about

225

the structural relation between the ash matrix and the Ag-NP, further studies were con-

226

duced on a STEM. The microscope was equipped with a SE and a HAADF detector

227

to probe the surface and the atomic weight of selected sites of interest. Three images

228

recorded from a 1 µm sized ash particle (Ash 5.1) are given in Figure 2A-C. Nanoscale

229

surface structures revealed in the SE image (Figure 2A), were not correlated to Ag-NP ob-

230

served in the respective HAADF image (Figure 2B), which represents an elemental con-

231

trast. Thus, Ag-NPs seem to be incorporated into the matrix of the ash particles. Images

232

recorded at higher resolutions (Figure 2C) further revealed that the Ag-NP ’sprinkles’ al-

233

ready identified by SEM analysis may well be composed of even smaller Ag-NP possibly

234

extending into the sub-nm size range.

235

Analysis of selected Ag-NP by EDX

236

(Figure 1E) revealed substantial signal intensities of Al, P, Si, and Ca which were related

237

to the ash matrix (background spectrum in Figure 1E). Elemental distribution maps (Fig-

238

ure 1F) showed elevated Ag signal intensities corresponding to bright areas in the BSE

239

image, but the corresponding S signal intensities remained at background levels. Fur-

240

thermore, S was not detected in STEM-EDX spectra of Ag-NP in ash samples (Figure 2E,

241

Ash 5.1) in strong contrast to spectra recorded on Ag-NP in the dried sludge (Figure 2D).

242

Thus, results from both SEM and STEM analyses suggest a decomposition of Ag2 S-NP

243

and the formation of new, most probably metallic Ag-NP, although the presence of Ag2 O-

244

NP cannot be excluded based on EDX analysis.

245

Identification of Ag species in the ashes by Ag K-edge EXAFS spectroscopy The Ag

246

K-edge EXAFS spectra of selected ash samples are shown in Figure 3. To assess the num-

The SEM-EDX spectra of Ag-NP in ash samples

10


Page 10 of 26

Page 11 of 26


247

ber of principal components required to describe the observed variability, a PCA based

248

on the spectra of all ash samples listed in Table 1 was performed. Based on the so-called

249

indicator function as well as a one-tailed F-test, 45 only two PCs were required to describe

250

the sample spectra (see Table S3 in the SI). This is supported by the good agreement be-

251

tween the sample EXAFS spectra and their reconstruction based on the first two compo-

252

nents confirming that two PCs accounted for the essential features of the EXAFS spectra

253

(Figure 3). Target testing (TT) was performed to identify the most suitable Ag reference

254

spectra to describe the experimental spectra by linear combination fit (LCF) analysis. The

255

likelihood of a reference spectrum to represent a spectral component in the dataset is ex-

256

pressed by the semi-empirical SPOIL function. 45 The SPOIL values for all tested reference

257

spectra are listed in Table S4. References with SPOIL < 1.5 are considered excellent and

258

references with SPOIL values of 1.5-3 as good for LCF analysis. Accordingly, the spectra

259

of Ag0 (SPOIL 0.8), Ag3 PO4 (SPOIL 1.6) and Ag2 SO4 (SPOIL 2.8) were selected as possi-

260

ble reference spectra. Reference spectra with SPOIL values > 6 are unacceptable for LCF

261

analysis, suggesting that Ag2 S (SPOIL 9.9) was not a significant component in the spectral

262

dataset. In line with their SPOIL values, the reference spectra of Ag0 , Ag3 PO4 and Ag2 SO4

263

were reasonably well reproduced by the first two principal components (Figure 3).

264

TT was also performed on calculated binary mixtures and we found that a mixture of

265

15 % Ag2 S and 85 % Ag0 resulted in the lowest SPOIL value (0.3). This could indicate that

266

a minor fraction of Ag2 S resisted transformation during incineration, but may also result

267

from the small size or poor crystallinity of metallic Ag-NP formed during incineration as

268

compared to bulk Ag metal. Likewise, a mixture of 33 % Ag2 SO4 and 67 % Ag3 PO4 also

269

resulted in a smaller SPOIL value (1.3) than obtained for the individual reference spectra.

270

Ag2 SO4 and Ag3 PO4 exhibit similar EXAFS spectra and represent Ag coordinated to O in

271

the first-shell. Their combination thus most probably represents an approximation to the

272

average Ag-O coordination occurring in the ash samples.

273

The weights of the first two PCs on the individual ash samples and on selected ref11



274

erence spectra are plotted in Figure 4. All experimental spectra plot along a straight line

275

bracketed by the Ag0 reference and the Ag2 SO4 -Ag3 PO4 mixture. In combination with

276

the very high SPOIL value of the Ag2 S reference spectrum, this indicates that the residual

277

fraction of Ag2 S in the ash samples was negligible.

278

Metallic Ag0 fraction in sewage sludge ashes Based on the statistical analysis of the EX-

279

AFS spectra we identified Ag0 as a major component in ash samples. Ag3 PO4 or Ag2 SO4

280

or a mixture of both was found as second component in the dataset. For LCF analysis,

281

we therefore used Ag0 and Ag3 PO4 as reference spectra. The resulting Ag0 fractions are

282

given in Table 1 and the complete LCF results including statistical parameters are given

283

in Table S5in the SI. The Ag0 fraction varied substantially with the incineration time and

284

atmosphere (Figure 5). The conditions matching full scale incineration most closely (e.g.,

285

fluidized bed reactor, synthetic flue gas, 800 ◦C, retention times between 1 and 10 min)

286

resulted in a metallic fraction of about 70 % to 80 %. The fast decomposition of Ag2 S,

287

yielding Ag-Ag coordinated silver, is also observed in the air and the N2 atmosphere and

288

therefore seems to be triggered by the thermal exposure, independent of the fluidizing

289

gas composition. In both the air and the N2 atmosphere, the metallic Ag fraction further

290

increased as the incineration time was extended to 2 h. In the synthetic flue gas atmo-

291

sphere, in contrast, an extended incineration time of 2 h resulted in a decreased metallic

292

Ag fraction (50 %). This decrease of Ag0 is accompanied with an increase of the Ag3 PO4

293

fraction which we interpret as a slow oxidation of the Ag. This slow oxidation may be

294

caused by a direct interaction between the Ag-NP and the synthetic flue gas or by an in-

295

direct interaction of the Ag-NP with the altered ash matrix (e.g., in-situ capture of SO2 by

296

calcium in the ash resulting in CaSO4 formation).

297

In fluidized bed experiments conducted with synthetic flue gas over 2 h, the Ag0 per-

298

centage consistently increased with temperature from 23 % (600 ◦C) over 48 % (800 ◦C) to

299

69 % (900 ◦C). Thus, the above mentioned slow oxidation of Ag is sensitive towards the

12


Page 12 of 26

Page 13 of 26

300

301

302

303


incineration temperature. In muffle oven experiments, lower Ag0 fractions (23 % to 49 %) were obtained compared to the fluidized bed experiments, regardless of the incineration temperature (700 ◦C to 900 ◦C). These lower Ag0 fractions are in agreement with results from comparable ex-

304

periments (Ag2 S spiked sewage sludge incinerated in a muffle oven over 4 h at 850 ◦C)

305

reported by Impellitteri et al. 12

306

The low totals of the LCF analyses (72 % to 94 %, see Table S5) suggested that an addi-

307

tional reference spectra is required to properly describe the experimental dataset.

308

Comparison between EM and XAS The absence of Ag2 S-NP in ash samples from 1 min

309

fluidized bed experiments revealed by EM analysis suggested a rapid decomposition of

310

Ag2 S-NP during incineration. These findings are consistent with results from the analysis

311

of Ag K-edge EXAFS spectra, which suggested that Ag2 S was not a significant species in

312

the ash samples. The EDX spectra of the selected particles showed a strong Ag signal but

313

only negligible signal intensities for S and P compared to the background spectra. Thus,

314

based on SEM- and TEM-EDX analyses, no evidence for the presence of a distinct phase

315

such as Ag3 PO4 or Ag2 SO4 in addition to Ag0 was found.

316

The reference spectra of Ag3 PO4 was included in the LCF fits to account for an Ag-

317

O coordination environment. The Ag3 PO4 reference could also represent Ag complexed

318

to O-containing ligands throughout the ash, i.e., in a form not readily detectable by EM,

319

explaining the apparent discrepancy between results from EM (no indication for Ag3 PO4 )

320

and XAS (inclusion of Ag3 PO4 in the LCF analyses).

321

In the SEM different types of Ag-NP have been observed, and one Ag-NP type was

322

very sensitive to electron beam irradiation. These beam sensitive particles may repre-

323

sent metallic Ag but structurally different than the reference spectra used for LCF analy-

324

ses. Results from TEM analysis further suggested that the newly formed Ag-NP (’sprin-

325

kles’) extended into the sub-nm range. At this length scale, the fraction of surface Ag

13



326

approaches unity, resulting in differences in Ag-Ag distances compared to Ag0 in bulk

327

phase and a concomitant increase in the fraction of O-coordinated Ag. 46 Both structural

328

and size effects resulting in differences between experimental and reference EXAFS spec-

329

tra may explain the low LCF totals.

330

Environmental Impact

331

important waste management facilities after their likely discharge into wastewater. The

332

In this study we documented the transformation of Ag0 -NP in

observed transformation of Ag0 -NP during the waste water / sludge treatment to to spar-

333

ingly soluble Ag2 S-NP is in agreement with results from previous studies on the fate of

334

Ag-NP in urban wastewater systems. 12–14,43,44 Ag2 S is less of an environmental concern

335

compared to Ag0 due to its limited solubility and its resistance towards oxidation un-

336

der environmentally relevant conditions. However, in this study, we revealed the rapid

337

formation of new Ag0 -NP from Ag2 S-NP during incineration in a fluidized bed reactor

338

under typical field scale incineration conditions. Therefore, the rapid transformation of

339

Ag2 S-NP in the sewage sludge into Ag0 -NP during the incineration should be considered

340

in life cycle assessments of Ag-NP, especially as the further processing and the final use of

341

342

343

Page 14 of 26

the sewage sludge ashes is still under discussion. Furthermore, the formation of Ag0 -NP with modified physico-chemical properties as indicated by their increased susceptibility to electron beam irradiation, and the structural associations between Ag0 -NP and the ash

344

matrix, possibly affecting the leaching behavior of the ash, needs to be addressed in future

345

studies.

346

Acknowledgement

347

We acknowledge the Electron Microscopy Centers at ETH Zurich (EMEZ, Zurich, Switzer-

348

land) and at Empa (Swiss Federal Institute for Materials Science and Technology, Dueben-

349

dorf, Switzerland) for providing access to the microscopes. The Swiss Light Source (SLS,

350

Villigen, Switzerland) is acknowledged for the allocation of beamtime. We thank Maarten 14


Page 15 of 26


351

Nachtegaal and Grigory Smolentsev (SLS) for support at the SuperXAS beamline (SLS)

352

as well as Sergey Nikitenko for support at the DUBBLE beamline (ESRF). We also thank

353

Maggy Lengke and Gordon Southam for sharing the Ag3 PO4 reference spectra provided

354

with Demeter. This projects was additionally suppoted by the French program LabEx Ser-

355

enade (11-LABX-0064), ISTerre, CNRS (PEPS project NANOPLANTE) and COST ES1205

356

(ENTER).

357

Supporting Information Available

358

Schematic layout of the pilot WWTP, characterization of the dried sewage sludge, de-

359

scription and schematics of the fluidized bed reactor, concentrations of selected elements

360

measured by ICP-OES, TEM-HAADF images, and details on the PCA-TT-LCF data treat-

361

ment data are provided in the SI. This material is available free of charge via the Internet

362

at http://pubs.acs.org/.

363

References

364

365

366

367

368

369

(1) Nowack, B.; Krug, H. F.; Height, M. 120 years of nanosilver history: implications for policy makers. Environmental Science & Technology 2011, 45, 1177 – 1183. (2) Woodrow Wilson International Center for Scholars, The Project on Emerging Nanotechnologies. Woodrow Wilson International Center for Scholars. (3) Ratte, H. T. Bioaccumulation and toxicity of silver compounds: a review. Environmental Toxicology and Chemistry 1999, 18, 89 – 108.

370

(4) Levard, C.; Hotze, E. M.; Lowry, G. V.; Brown, G. E. J. Environmental Transforma-

371

tions of Silver Nanoparticles: Impact on Stability and Toxicity. Environmental Science

372

& Technology 2012, 46, 6900–6914. 15



373

(5) Fabrega, J.; Luoma, S. N.; Tyler, C. R.; Galloway, T. S. Silver nanoparticles: behaviour

374

and effects in the aquatic environment. Environment International 2011, 37, 517 – 531.

375

(6) Marambio-Jones, C.; Hoek, E. A review of the antibacterial effects of silver nanoma-

376

terials and potential implications for human health and the environment. Journal of

377

Nanoparticle Research 2010, 12, 1531 – 1551.

378

(7) Xiu, Z.; Zhang, Q.; Puppala, H. L.; Colvin, V. L.; Alvarez, P. Negligible particle-

379

specific antibacterial activity of silver nanoparticles. Nano letters 2012, 12, 4271 – 4275.

380

(8) Yin, L.; Cheng, Y.; Espinasse, B.; Colman, B. P. More than the ions: the effects of silver

381

nanoparticles on Lolium multiflorum. Environmental Science & Technology 2011, 45,

382

2360 – 2367.

383

(9) González, A. G.; Mombo, S.; Leflaive, J.; Lamy, A.; Pokrovsky, O. S.; Rols, J.-L. Sil-

384

ver nanoparticles impact phototrophic biofilm communities to a considerably higher

385

degree than ionic silver. Environmental Science and Pollution Research 2014, 22, 8412–

386

8424.

387

(10) Sun, T. Y.; Gottschalk, F.; Hungerbühler, K.; Nowack, B. Comprehensive probabilistic

388

modelling of environmental emissions of engineered nanomaterials. Environmental

389

Pollution 2014, 185, 69 – 76.

390

391

(11) Keller, A. A.; McFerran, S.; Lazareva, A.; Suh, S. Global life cycle releases of engineered nanomaterials. Journal of Nanoparticle Research 2013, 15, 1692.

392

(12) Impellitteri, C. A.; Harmon, S.; Silva, R. G.; Miller, B. W.; Scheckel, K. G.; Luxton, T. P.;

393

Schupp, D.; Panguluri, S. Transformation of silver nanoparticles in fresh, aged, and

394

incinerated biosolids. Water research 2013, 47, 3878–3886.

395

(13) Lombi, E.; Donner, E.; Taheri, S.; Tavakkoli, E.; Jämting, Å. K.; McClure, S.; Naidu, R.;

396

Miller, B. W.; Scheckel, K. G.; Vasilev, K. Transformation of four silver/silver chlo16


Page 16 of 26

Page 17 of 26


397

ride nanoparticles during anaerobic treatment of wastewater and post-processing of

398

sewage sludge. Environmental Pollution 2013, 176, 193–197.

399

(14) Kaegi, R.; Voegelin, A.; Ort, C.; Sinnet, B.; Thalmann, B.; Krismer, J.; Hagendorfer, H.;

400

Elumelu, M.; Mueller, E. Fate and transformation of silver nanoparticles in urban

401

wastewater systems. Water research 2013, 47, 3866–3877.

402

(15) Kaegi, R.; Voegelin, A.; Sinnet, B.; Zuleeg, S.; Siegrist, H.; Burkhardt, M. Transfor-

403

mation of AgCl nanoparticles in a sewer system — A field study. Science of The Total

404

Environment 2015, 535, 20 – 27.

405

406

(16) Liu, J.; Pennell, K. G.; Hurt, R. H. Kinetics and mechanisms of nanosilver oxysulfidation. Environmental Science & Technology 2011, 45, 7345 – 7353.

407

(17) Levard, C.; Hotze, E. M.; Colman, B. P.; Dale, A. L.; Truong, L.; Yang, X. Y.; Bone, A. J.;

408

Brown, G. E., Jr.; Tanguay, R. L.; Di Giulio, R. T.; Bernhardt, E. S.; Meyer, J. N.; Wies-

409

ner, M. R.; Lowry, G. V. Sulfidation of Silver Nanoparticles: Natural Antidote to Their

410

Toxicity. Environmental Science & Technology 2013, 47, 13440–13448.

411

(18) Kim, B.; Park, C.-S.; Murayama, M.; Hochella, M. F. J. Discovery and Characteriza-

412

tion of Silver Sulfide Nanoparticles in Final Sewage Sludge Products. Environmental

413

Science & Technology 2010, 44, 7509–7514.

414

(19) Thalmann, B.; Voegelin, A.; Sinnet, B.; Morgenroth, E.; Kaegi, R. Sulfidation Kinetics

415

of Silver Nanoparticles Reacted with Metal Sulfides. Environmental Science & Technol-

416

ogy 2014, 48, 4885–4892.

417

(20) Brunetti, G.; Donner, E.; Laera, G.; Sekine, R.; Scheckel, K. G.; Khaksar, M.;

418

Vasilev, K.; De Mastro, G.; Lombi, E. Fate of zinc and silver engineered nanoparticles

419

in sewerage networks. Water research 2015, 77, 72–84.

17



420

(21) Reinsch, B. C.; Levard, C.; Li, Z.; Ma, R.; Wise, A. Sulfidation of silver nanoparticles

421

decreases Escherichia coli growth inhibition. Environmental Science & Technology 2012,

422

46, 6992–7000.

423

(22) Kaegi, R.; Voegelin, A.; Sinnet, B.; Zuleeg, S.; Hagendorfer, H.; Burkhardt, M.;

424

Siegrist, H. Behavior of metallic silver nanoparticles in a pilot wastewater treatment

425

plant. Environmental Science & Technology 2011, 45, 3902–3908.

426

(23) Kiser, M. A.; Ryu, H.; Jang, H.; Hristovski, K.; Westerhoff, P. Biosorption of nanopar-

427

ticles to heterotrophic wastewater biomass. Water research 2010, 44, 4105 – 4114.

428

(24) Limbach, L. K.; Bereiter, R.; Müller, E.; Krebs, R.; Gälli, R.; Stark, W. J. Removal of

429

Oxide Nanoparticles in a Model Wastewater Treatment Plant: Influence of Agglom-

430

eration and Surfactants on Clearing Efficiency. Environmental Science & Technology

431

2008, 42, 5828–5833.

432

(25) Westerhoff, P.; Song, G.; Hristovski, K.; Kiser, M. A. Occurrence and removal of tita-

433

nium at full scale wastewater treatment plants: implications for TiO 2 nanomaterials

434

. Journal of Environmental Monitoring 2011, 13, 1195–1203.

435

(26) Fytili, D.; Zabaniotou, A. Utilization of sewage sludge in EU application of old and

436

new methods—A review. Renewable and Sustainable Energy Reviews 2008, 12, 116–140.

437

(27) Beech, N.; Crawford, K.; Goldstein, N.; Kester, G. A national biosolids regulation, qual-

438

ity, end use and disposal survey: Final report; North East Biosolids and Residuals Asso-

439

ciation, 2007.

440

441

442

(28) Wiechmann, B.; Dienemann, C.; Kabbe, C.; brandt, S. Klärschlammentsorgung in der Bundesrepublik Deutschland. OPUS-IDN/13541 (29) Kelessidis, A.; Stasinakis, A. S. Comparative study of the methods used for treatment

18


Page 18 of 26

Page 19 of 26


443

and final disposal of sewage sludge in European countries. Waste Management 2012,

444

32, 1186 – 1195.

445

446

447

448

(30) Werther, J.; Ogada, T. Sewage sludge combustion. Progress in Energy and Combustion Science 1999, 25, 55 – 116. (31) Krüger, O.; Adam, C. Recovery potential of German sewage sludge ash. Waste Management 2015, 45, 400–406.

449

(32) Zivkovic, D.; Sokic, M.; Zivkovic, Z.; Manasijevic, D.; Balanovic, L.; Strbac, N.; Coso-

450

vic, V.; Boyanov, B. Thermal study and mechanism of Ag2S oxidation in air. Journal

451

of Thermal Analysis and Calorimetry 2013, 111, 1173–1176.

452

453

(33) Karakaya, I.; Thompson, W. T. The Ag-O (silver-oxygen) system. Journal of Phase Equilibria 1992, 13, 137–142.

454

(34) Pradas del Real, A. E.; Castillo-Michel, H.; Kaegi, R.; Sinnet, B.; Magnin, V.; Find-

455

ling, N.; Villanova, J.; Carriere, M.; Santaella, C.; Fernandez-Martinez, A.; Levard, C.;

456

Sarret, G. Fate of Ag-NPs in Sewage Sludge after Application on Agricultural Soils.

457

Environmental Science & Technology 2016, acs.est.5b04550.

458

(35) USEPA, Targeted National Sewage Sludge Survey, Appendix A; 2009.

459

(36) Hartman, M.; Svoboda, K.; Pohorelý, M. Combustion of dried sewage sludge in a

460

461

462

fluidized-bed reactor. Industrial & engineering . . . 2005, (37) Ogada, T.; Werther, J. Combustion characteristics of wet sludge in a fluidized bed Release and combustion of the volatiles. Fuel 1996, 75, 617–626.

463

(38) Urciuolo, M.; Solimene, R.; Chirone, R.; Salatino, P. Fluidized bed combustion and

464

fragmentation of wet sewage sludge. Experimental Thermal and Fluid Science 2012, 43,

465

97–104.

19



466

(39) van der Walt, S.; Colbert, S. C.; Varoquaux, G. The NumPy Array: A Structure for

467

Efficient Numerical Computation. Computing in Science & Engineering 2011, 13, 22–

468

30.

469

470

(40) Newville, M. Larch: An Analysis Package for XAFS and Related Spectroscopies. Journal of Physics: Conference Series 2013, 430, 012007.

471

(41) Newville, M.; L¯ıvin¸š, P.; Yacoby, Y.; Rehr, J. J.; Stern, E. A. Near-edge x-ray-absorption

472

fine structure of Pb: A comparison of theory and experiment. Physical Review B 1993,

473

47, 14126.

474

475

(42) Ravel, B. Demeter:

X-ray Absorption Spectroscopy Using Feff and Ifeffit.

http://bruceravel.github.io/demeter/ accessed September 10th 2015.

476

(43) Ma, R.; Levard, C.; Judy, J. D.; Unrine, J. M.; Durenkamp, M.; Martin, B.; Jefferson, B.;

477

Lowry, G. V. Fate of Zinc Oxide and Silver Nanoparticles in a Pilot Wastewater Treat-

478

ment Plant and in Processed Biosolids. Environmental Science & Technology 2014, 48,

479

104–112.

480

(44) Doolette, C. L.; McLaughlin, M. J.; Kirby, J. K. Transformation of PVP coated silver

481

nanoparticles in a simulated wastewater treatment process and the effect on micro-

482

bial communities. Chemistry Central Journal 2013, 7, doi:10.1186/1752–153X–7–46.

483

(45) Malinowski, E. R. Factor Analysis in Chemistry; 2002.

484

(46) Srabionyan, V. V.; Bugaev, A. L.; Pryadchenko, V. V.; Makhiboroda, A. V.;

485

Rusakova, E. B.; Avakyan, L. A.; Schneider, R.; Dubiel, M.; Bugaev, L. A. EXAFS

486

study of changes in atomic structure of silver nanoparticles in soda-lime glass caused

487

by annealing. Journal of Non-Crystalline Solids 2013, 382, 24–31.

20


Page 20 of 26

Page 21 of 26


Table 1: Sample labels and experimental conditions together with the Ag0 fraction derived from LCF analyses and the Ag concentrations derived from ICP-OES measurements. Sample preparation Ag0 fraction ICP-OES Denom.a Atm.b T/◦C t/min from LCFc Ag mg kg−1 Ash 1 N2 800 1 72 % 339 Ash 2 N2 800 120 77 % 1503 Ash 3 Air 800 1 70 % 857 Ash 4 Air 800 120 86 % 1502 Ash 5.1 FG 800 1 72 % 455 Ash 5.2 FG 800 1 60 % 2681 Ash 6.1 FG 800 10 76 % 2700 Ash 6.2 FG 800 10 81 % 2627 Ash 7.1 FG 800 120 48 % 2466 Ash 7.2 FG 800 120 47 % 2466 Ash 8 FG 900 120 69 % 436 Ash 9 FG 600 120 23 % 2395 Ash 10 Oven 900 120 31 % 351 Ash 11.1 Oven 800 120 29 % 507 Ash 11.2 Oven 800 120 23 % 1902 Ash 12 Oven 700 120 39 % 501 a Samples in duplicate (e.g., 5.1 and 5.2) were prepared independently. b Atmosphere Oven = samples prepared in a muffle oven, where oxidizing conditions prevail. All other atmospheres refer to fluidized bed incineration, where N2 = pressurized Nitrogen, Air = pressurized air from in-house supply system (dry), FG = synthetic flue gas mixture (from bottle, 4 % O2 , 12 % CO2 , 300 ppm SO2 , rest N2 ) humidified to a dew point temperature of 70 ◦C. c EXAFS LCF using Ag0 and Ag PO as reference spectra. Fit statistics shown in Table S5. 3 4

21




Page 22 of 26

Page 23 of 26


A

B

C

500 nm

500 nm D

100 nm

E Sludge

F

Ash

200 nm

100 nm

Figure 2: A-C: STEM images of a µm sized ash particle (Ash 5.1, 1 min exposure at 800 ◦C). A: SE image (topography), B: HAADF image of the same area as A, C: higher magnified HAADF image, image area is indicated by the yellow rectangle in B. D: Ag-NP in dried sludge, E: Ag-NP in Ash 5.1, F: EDX spectra from the designated areas in D and E.

23




Page 24 of 26

Page 25 of 26





Page 26 of 26

Transformation of Silver Nanoparticles in Sewage ... - ACS Publications

Recommend Documents