Emissions of Secondary Formed ZnO Nano ... - ACS Publications

Subscriber access provided by READING UNIV

Article

Emissions of secondary formed ZnO nano-objects from the combustion of impregnated wood. An on-line size-resolved elemental investigation. Debora Foppiano, Mohamed Tarik, Elisabeth Müller Gubler, and Christian Ludwig Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b03584 • Publication Date (Web): 19 Dec 2017 Downloaded from http://pubs.acs.org on December 20, 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 24

Environmental Science & Technology

1

Emissions of secondary formed ZnO nano-objects

2

from the combustion of impregnated wood. An on-

3

line size-resolved elemental investigation.

4

Debora Foppiano* a,c, Mohamed Tarik a, Elisabeth Müller Gubler b, Christian Ludwig* a,c.

5

a

6

(ENE), Paul Scherrer Institut (PSI), CH 5232 Villigen PSI, Switzerland

7

b

8

Scherrer Institut (PSI), CH 5232 Villigen PSI, Switzerland

9

c

Bioenergy and Catalysis Laboratory (LBK), Energy and Environment Research Division

Laboratory of Biomolecular Research (LBR), Biology and Chemistry Division (BIO), Paul

Environmental Engineering Institute (IIE), School of Architecture, Civil and Environmental

10

Engineering (ENAC), École Polytechnique Fédérale de Lausanne (EPFL), CH 1015 Lausanne,

11

Switzerland.

12

KEYWORDS

13

ZnO nanoparticles, Elemental analysis, On-line measurement, Aerosol, Wood combustion,

14

Waste biomass, Particle emission, Thermal desorption, Gas-borne nanoparticles, Redox-sensitive

15

Zn, Incineration

16

ABSTRACT. The release of secondary nano-objects formed during waste combustion processes

17

is becoming a matter of concern, considering their known toxicity and the fact that the 100%

ACS Paragon Plus Environment

1


Page 2 of 24

18

efficiency of filtering systems is not always ensured. An increased cytotoxicity and genotoxicity

19

on human peripheral blood lymphocytes is known particularly in the case of ZnO, which is often

20

contained in paints and waterproof agents, heading to a relevant quantity present in the waste-

21

wood material. In this study, the behavior of ZnO nanoparticles during wood combustion and the

22

effect of the reduction potential of generated carbon species on the release of secondarily formed

23

ZnO-containing nano-objects were investigated. By hyphenating a modified scanning mobility

24

particle sizer (SMPS) and inductively coupled mass spectrometry (ICP-MS), it was possible to

25

obtain simultaneously size-resolved and chemical information on the emitted nanoparticles.

26

Through the established correlation between SMPS and ICP-MS signals, Zn-containing particles

27

were efficiently resolved from the combustion generated particles. Transmission electron

28

microscopy (TEM) and energy-dispersive X-ray spectroscopy (EDS) on size-selected particles

29

confirmed the SMPS and ICP-MS data. The use of electron diffraction allowed determining the

30

structure of the crystalline materials as hexagonal ZnO. A possible mechanism of reduction of

31

ZnO to Zn and further reformation as secondary nano-objects is proposed.

32

2. INTRODUCTION

33

Engineered nanoparticles (ENPs) and nano-objects are widely used in many applications such as

34

motor- or vehicle- related products, electronics, paints, coatings and adhesives, household

35

products, and personal care and cosmetic products. Increasingly, with the heightened interest in

36

reducing atmospheric particle emissions, the release of such nanoparticles from waste

37

incineration and combustion processes has grown in importance. During waste combustion,

38

metallic ultrafine particle emissions have been reported (dominated by K, Zn and Ca, Fe, Mg,

39

and Na).1


2

Page 3 of 24


40

A recent study carried out by various institutions in Switzerland showed how modern solid-waste

41

incinerators can retain nanoparticles through the filtering system.2 However, the occurrence of

42

unintentionally produced nanomaterials cannot be excluded. The so-called incidental

43

nanoparticles may volatilize in the incinerator during incomplete combustion processes and they

44

can re-form after filtration of the flue gas.3 Moreover, new nano-objects can be formed also as

45

result of disaggregation of the already present nano-object and re-formation of new nano-sized

46

compounds and materials.4

47

Depending on their surface area, the evaporation temperature of metal ENPs is significantly

48

lower than that of the respective bulk material. However, when heterogeneous reactions and

49

agglomeration occur during combustion, a generally decreased surface area of nanoparticles

50

allows lower temperatures of evaporation. Moreover, the stability of metal ENPs is enhanced for

51

the oxidized form, known to be less soluble and with a higher flash point than the corresponding

52

metal. Therefore, certain metal oxide ENPs such as cerium oxide have a higher tendency to

53

remain in the bottom ashes without being volatilized during incineration.5 Regarding the matrix

54

effect, Jakob et al. showed how thermal treatment of fly ashes, involving thermochemical

55

reactions of the heavy metals with other constituents (e.g. alumino-silicates compound, alkali

56

metals chlorides and carbon species), provoke their evaporation.6

57

Zinc oxide is one of the relevant metal oxide NPs regarding significant exposures

58

often contained in paints and impregnation or waterproofing sprays, which are used for wood

59

preservation. Particular redox conditions could influence the evaporation of zinc since it can be

60

reduced in the presence of carbon from its oxide to its metallic state, which evaporates at a much

61

lower temperature.9 Thus, zinc represents an interesting example of unintentionally produced

62

nanomaterials during the combustion process, in which a large amount of carbon is present.

7,8

and it is


3


Page 4 of 24

63

Generally, the volatility of Zn decreased in the presence of oxygen and water but is strongly

64

enhanced in presence of C contained in the feedstock, leading to the increase in volatility from

65

20% at 750 °C to 100% at about 950-1130 °C.10

66

Different techniques can be used to determine on-line particle size distribution (PSD) and

67

concentration of gas-borne particles. A comprehensive selection of PSD determination

68

techniques is reported in several reviews,11,12 but among them, the scanning mobility particle

69

sizer (SMPS) is one of the most established techniques in particle emission measurements. On

70

the other hand, multi-elemental techniques as Inductively Coupled Plasma Mass Spectrometry

71

(ICPMS),13 are often used to determine the chemical composition off-line. However,

72

contamination or morphology alteration of the particles for off-line analytical methods must not

73

be neglected.

74

The presence of Zn in waste wood combustion off-gases was measured by an online elemental

75

analysis of process gases, performed with ICP-OES (inductively coupled plasma optical

76

emission spectrometer).14 Furthermore, data from several studies have been identifying a Zn-

77

enrichment in the fraction of fine particles;15 while in these studies the particle size distribution

78

was measured on-line, the chemical composition of the collected fly ashes was determined off-

79

line.

80

In the current work, the behavior of ZnO during wood incineration has been investigated through

81

the use of a recently developed on-line technique, namely SMPS-ICP-MS, using a rotating disk

82

diluter (RDD) as introduction system.16,17 The analysis was focused on a size range between 14

83

and 300 nm. In the early 2000s, a known survey was published on the health impact of the

84

average PM2.5 (the particulate matter or particle pollution with a diameter of 2.5 µm or below),


4

Page 5 of 24


85

showing how particles in this size range have a direct connection with lung cancer and

86

respiratory mortality.18 Another study reported that even within this size range the transport of

87

NPs to human lung fibroblast can anyway vary a lot for different particle sizes.19 In the case of

88

particles in between 25-50 nm, the tendency to rapidly form larger agglomerates limits the

89

diffusion, while particles between 250-500 nm are mostly unagglomerated. Therefore, larger NPs

90

resulted in a higher uptake of different metal oxides compared to smaller particles, even at a

91

concentration of 100 ng g-1. Furthermore, not only the size but also the chemical composition

92

and morphology of NPs or nano-objects is affecting the inflammatory response. For example,

93

Dandley et al. showed how much an acute pulmonary response is shown for in-vitro human cells

94

and mice after exposure to ZnO coated carbon nanotubes, while reduced fibrosis is associated

95

with Al2O3 coated carbon nanotubes.20

96

ZnO dried particles, beech wood sawdust and, ZnO impregnated sawdust were combusted

97

following two different temperature profiles in a lab-scale incinerator. The potential release of

98

ZnO nanoparticles in sawdust samples formed in the reducing process gases was of particular

99

interest.

100

3. EXPERIMENTAL SECTION

101

3.1. Chemicals. A commercial zinc oxide nanopowder (5810MR, 1314-13-2, AUER-REMY),

102

was used to prepare suspensions of ZnO NPs. The average particle size provided by the

103

manufacturer was verified with SEM analysis. Characterizations and further specifications of the

104

powder are reported in the supporting information (SI) in Figures S1 and S2 and Table S1. For

105

the colloidal stability of ZnO particles dispersion, a solution 0.1% wt of Poly Acrylic Acid


5


Page 6 of 24

106

(PAA, MW2000, 50% wt H2O, electronic grade, 535931-100G, lot # 02115HIV, Sigma-Aldrich)

107

was required.

108

Nitric acid (HNO3, 65%, Z0346841519, 1.00441.1000, Merck) and hydrochloric acid HCl (30%,

109

Z0344318516, 1.00318.1000, Merck), were used to dissolve the deposited material and to clean

110

the liner and the quartz tube of the tubular furnace after each experiment. The resulting solutions

111

were further diluted and analyzed by ICP-MS. Zn calibration solutions for analysis of the

112

collected samples were prepared by the dilution of Zn standard solution 1000 µg/ml (Merck,

113

1.19806.0100) with ultrapure water (MilliQ, 18.2 MΩ cm).

114

3.2. Samples. ZnO suspensions. A suspension 0.3% wt of zinc oxide (Suspension A) was

115

prepared by dispersing 333.3 mg of the commercial powder in 100 g of dispersant solution. The

116

amount of nano-powder used was chosen according to the minimum weight Wmin needed to have

117

a representative sample of the bulk lot.21 An ultrasonic treatment was applied three times for two

118

minutes to homogeneously disperse the suspension; ultrasonic horn was preferred to bath

119

immersion since it is reported to be more effective for concentrated dispersions.22 The second

120

suspension with a concentration of 600 µg/ml of ZnO (suspension B) was prepared by dilution of

121

suspension A with the same dispersant solution, under constant magnetic stirring and ultrasonic

122

bath for six minutes. The suspensions were characterized by using a Malvern Zetasizer Nano

123

series, a Brookhaven X-ray disc centrifuge, and ICP-MS to verify the effective concentration.

124

The results of these analyses are reported in supporting information in Tables S2, S3 and S4.

125

Beech wood sawdust. The samples of beech wood were collected in the forest area of Villigen in

126

March 2012 and further characterized by Bahrle.23 The multi-element analysis was performed in

127

ICP-MS after microwave assisted acid digestion with HNO3. The results are reported in


6

Page 7 of 24


128

supporting info, in Tables S5 and S6. Multi–element calibration solutions were prepared by

129

dilution of a commercial multi-elemental standard solution (Bernd Kraft, 10931.2000, batch-no.:

130

1530569). Non-treated samples and samples impregnated with ZnO suspension were used for

131

incineration experiments.

132

Beechwood sawdust impregnated with ZnO suspension. In order to obtain a synthetic nano-waste

133

sample, 2 g of beech wood sawdust were impregnated with zinc oxide suspension A (0.3% wt).

134

The sawdust samples were kept in contact with 15 ml of ZnO suspension for three hours under

135

magnetic stirring. After a drying step of 90 minutes at 105°C in a thermobalance (PM100, LP-

136

16, Mettler-Toledo), the samples were carefully removed from the Petri capsule. In every

137

experiment, an amount of 400 mg of the prepared batch was transferred into a boat-shaped

138

alumina crucible and introduced into the tubular furnace for the thermal treatment experiments.24

139

3.3. Instrumental arrangement. The experimental setup consisted of a tubular furnace, used as

140

a lab-scale incinerator (Heraeus, RE 1.1), connected to the SMPS-ICP-MS, together with a

141

rotating disk diluter (RDD) equipped with a heating tube as introduction system. A detailed

142

description of RDD-SMPS-ICP-MS coupling strategy and instruments specifications were

143

already presented.17,25

144

Several mass flow controllers (MFC) were used to precisely adjust the flows. The different flows

145

up- and downstream the tubular furnace were verified by a standard primary flow calibrator

146

(Gilibrator-2, Sensidyne) at the beginning of every measurement.


7


Page 8 of 24

Figure 1. Flow concept of the thermal treatment and the SMPS-ICPMS setup. The red arrows show the main path of the measurement, while orange arrows show the parallel measurement with a reference SMPS. The different gases provided to the system are instead indicated as light-blue arrows. Legend: MFC = mass flow controller; RDD = rotating disc diluter, ASET =evaporation tube; SMPS = scanning mobility particle sizer, DMA =differential mobility analyzer, CPC =condensation particle counter; ICP-MS= inductively coupled plasma mass spectrometry.

147

Figure 1 above shows a drawing of the experimental setup. The argon-borne particles were

148

generated starting from a ZnO suspension by the aerosol generator by applying a relative gas

149

pressure at the inlet of the nebulizer of approx. 1 bar. The aerosol was then passed through the

150

silica gel dryer at ambient pressure, to reduce the relative humidity. A total flow of 1.5 L min-1

151

was introduced into the tubular furnace by adding 50 mL min-1 flow of oxygen (or hydrogen) and

152

additional argon to the aerosol generated flow. In order to avoid losses due to electrostatic

153

deposition in the system, all the instruments were connected using electrically conductive tubes.

154

At the outlet of the furnace, the aerosol is diluted with a mixture of Air/Ar and introduced into

155

the RDD.


8

Page 9 of 24


156

The RDD is consisting of a dilution head and a control unit as shown in Figure 1. The dilution

157

argon flow provided to the dilution head has to be defined by an MFC because it determines also

158

the flow rate of the diluted aerosol directed to the DMA. Only the RDD MD193E connected to

159

the modified SMPS was also equipped with an evaporation tube, operated at 350°C in all

160

experiments. The SMPS is operated such as a size range between 13.1 and 299.6 nm is covered.

161

A second SMPS instrument (Electrostatic Classifier 3080, condensation particle counter 3776,

162

TSI) was used as a reference, operated with a separate RDD, as shown in Figure 1.The classified

163

aerosol leaving the DMA was split between the CPC and the ICP-MS with a ratio 3:7.

164

A flow of 4 mL min-1 of a mixture of xenon in argon (Xe 100 ppm), precisely adjusted with an

165

MFC, was added to the monodisperse aerosol flow directed into the ICP-MS. This allowed a

166

proper tuning of ICP-MS parameters before each measurement to obtain high sensitivity based

167

on the recorded signal of

168

during the overall analysis.

169

3.4. Temperature program of the lab-scale incinerator. In order to determine the optimal

170

dilution ratio for the generated emissions and to protect the instrument's detectors, a first

171

temperature program was performed with a constant heating rate (5°C min-1) between 200°C and

172

500°C. A standard incineration procedure was later applied to simulate the combustion processes

173

in an industrial plant (drying step, fast pyrolysis from 300°C to 900°C, combustion at 900°C and

174

afterburning from 900°C to 400°C).

175

The acquisitions of both SMPS and ICP-MS signals were started simultaneously when a

176

temperature of 105°C was reached inside the tubular furnace in the case of the first temperature

177

program (Figure S3a in SI). For the incineration program lasting only 24 minutes, the

124

Xe, and an eventual plasma drift could, therefore, be monitored


9


Page 10 of 24

178

measurements were started simultaneously with the furnace’s temperature program but the RDD

179

channel was opened only for combustion and afterburning steps, 6 minutes after the start (see

180

Figure S3b in SI).

181

3.5. Size-selected particles collection and sample preparation for TEM. The size-selected

182

particles from 13 nm to 340 nm were collected at the outlet of the DMA, using an aspiration EM

183

grid sampler26 (detail view and schematic sampling point reported in Figure S15 in SI). The

184

particles were filtered through a TEM grid (Quantifoil R 1.2/1.3 Cu 400 mesh) from the tube

185

normally directed to the ICP-MS at the outlet of DMA, with a flow rate of 0.3 L min-1 by using a

186

membrane pump. The particles were collected in separate replicates of the experiments reported

187

in result and discussion for which the ICP-MS detection was not included. The samples were

188

later analyzed with a JEOL JEM-2010 transmission electron microscope (200kV, point

189

resolution of 0.25nm), to investigate particle size and morphology. The chemical composition

190

was clarified through the use of EDS and electron diffraction pattern of the crystalline material.

191

4. RESULTS AND DISCUSSION

192

4.1. ZnO injected particles during sawdust thermal treatment. Dried ZnO particles generated

193

with the use of an aerosol generator were constantly injected into a lab-scale incinerator during

194

thermal treatment of beech sawdust under oxidizing conditions (O2 50 mL min-1). The particle

195

size distribution and the intensity of 66Zn were measured with SMPS-ICP-MS. The temperature

196

program applied is reported below in brown (Figure 2 a, b and d) and gray line (Figure 2c).


10

Page 11 of 24


Figure 2. 2D plot of the volume-related PSD of combusted sawdust and injected ZnO particles (a) and the sizerelated ICP-MS intensity of 66Zn (b). Raw ICP-MS intensity for 66Zn and 124Xe (c) showing increasing intensity from the black line; size-related ICP-MS intensity of 66Zn (d) corrected proportionally to the 124Xe signal.

197

Figure 2 (b) shows a particle-size-related ICP-MS intensity covering a 100-290 nm size range,

198

with intensity more or less constant during the total time of analysis. The contribution of Zn in

199

the overall PSD was efficiently resolved through the established correlation between ICP-MS

200

and SMPS shown in previous works,17 even though the considerable amount of particles

201

generated from wood treatment gives rise to a different PSD plot (Figure 2a). As shown in

202

Figure 2c the raw signal of the ICP-MS follows the size selection applied in the SMPS. At a

203

specific DMA voltage particles having the same mobility diameter will reach the ICP. By


11


Page 12 of 24

204

increasing the voltage the next size class is selected. This iterative process allows to have a

205

corresponding signal between each size class and ICP intensity over a time range of 150 s (up

206

scan), followed by a fast down scan from high to low voltage, in order to start the next up scan.

207

The size range of particle-size-related ICP-MS intensity is confirmed by experiments performed

208

in the same conditions on the ZnO suspension alone, showing a broad distribution ranging from

209

50 nm up to 290 nm, with a higher intensity between 100 and 290 nm (Figure S5 and Tables S2

210

and S3 in Supporting Information).

211

According to TGA studies, it is around 230°-380°C that hemicellulose, cellulose, and lignin

212

decompose (forming C, CO, CO2, H2O, etc.), followed by cellulose pyrolysis up to nearly

213

480°C.27 Similarly, the experiment with the tubular furnace confirmed the combustion of the

214

sawdust within 280°C and 380°C, where a maximum in the particle concentration occur (Figure

215

2a). This PSD is confirmed by experiments on the non-treated sawdust in which the same

216

temperature profile was applied but no additional stream of particles was injected into the system

217

(see SI, Figure S11). In correspondence with this temperature interval, an unexpected increased

218

intensity in

219

injected ZnO particles (Figure 2b). This behavior may have two possible explanations. First,

220

through the combustion, the concentration of C increased resulting in an increased ionization

221

efficiency of the plasma, and then in higher sensitivity of the different elements present in the

222

plasma, including Zn. Fliegel et al. reported how carbon addition is consistently improving

223

sensitivity for all elements from Li to Bi by the use of a mixed gas plasma with methane

224

addition.28 This seems to be confirmed by the corresponding increase in the

225

particular time frame with respect to the overall analysis (Figure 2c). Second, alterations of the

226

ZnO particles occur due to the reducing potential of the carbon compound generated by the

66

Zn ICP-MS signal and a decrease in mobility diameter were detected for the

124

Xe signal, in this


12

Page 13 of 24


227

sawdust combustion, resulting in a decrease of the particle size diameter in comparison to the

228

size of the injected particles. This hypothesis is not confirmed by TEM analysis on the size-

229

selected particles collected during this time frame (SMPS scan from 30 min to 40 min, Figure

230

2c). Figure 3b shows that not only the particle size is comparable but also the morphology is

231

consistent with the one analyzed for the ZnO powder (Figure 3a), that is, the carbon generated

232

atmosphere is not having a relevant effect on the ZnO nanoparticles but only on the

233

corresponding ICP-MS signal intensities, at this specific temperature interval. Therefore, the

234

particle-size-related ICP-MS signal was corrected proportionally to the drift registered for the

235

124

Xe signal (Figure 2d).

Figure 3. TEM micrograph on ZnO commercial nano-powder dispersed in ethanol (a); TEM micrograph on injected ZnO particles, which were collected after thermal treatment at about 340°C during the sawdust combustion (b) and after size-selection through the DMA.

236

4.2. ZnO impregnated sawdust thermal treatment. As a nano-waste model, a sample of ZnO

237

impregnated sawdust was investigated to monitor sub-micron particles release during thermal

238

treatment, applying the same conditions of the previously presented experiment (in excess of

239

oxygen and with the same temperature profile). For the 66Zn only a few peaks were detectable in

240

ICP-MS (Figure 4b), which can be most probably due to physical transport of some ZnO


13


Page 14 of 24

241

particles. Nevertheless, ZnO contained in the sawdust is to great extent retained in the bottom

242

ashes. These results might be explained by the fact that ZnO is very stable at the oxidizing

243

conditions up to high temperatures.29 Although carbo-thermal reduction of ZnO is a process

244

commonly used in metallothermic extraction30, this reaction cannot be taken into consideration

245

as a possible scenario, since elemental C and CO (g) generated by decomposition of carbon

246

species occur before reaching the temperature at which ZnO can be reduced (at about 500°C in

247

case of excess of H2, as shown in Figure S10 in SI).

Figure 4. Volume-related PSD of ZnO-impregnated sawdust (a) and ICP-MS mass intensity of 66Zn (b), in presence of O2 (50 ml min-1). Temperature profile in brown line with a constant heating rate (5°C min-1) between 200°C and 500°C.

248

On the other hand, the investigation of the behavior of the ZnO nano-powder under reducing

249

atmosphere (excess of H2 in an argon atmosphere) confirmed the evaporation of Zn(g) and the

250

change in morphology and chemical composition of the reformed Zn-containing particles. At

251

nearly 800°C the maximum in intensity is detected by SMPS-ICP-MS (Figure S10 in supporting

252

information). TEM analysis and electron diffraction patterns of the size-selected particles

253

collected at the DMA outlet suggest a core-shell structure with inner Zn core and outer


14

Page 15 of 24


254

passivation layer generated by re-oxidation of Zn surface (Figure 5b below and Figure S11d-f in

255

SI). In this experiment, a substantial deposition on the quartz tube was observed, either as a

256

white or gray film, in different areas of the tube. Figure 5a shows the presence of Zn in two

257

different morphologies, a lamellar one and a flower-like type, most probably associated with

258

ZnO. The Zn deposited in form of lamellar or hexagonal prismatic crystals, probably where the

259

temperature in the quartz tube dropped from 600°C to 400°C, as reported in previous

260

experiments by Weidenkaff.31 Further characterizations are reported in SI (Figure S11b-c).

Figure 5. SEM micrograph of ZnO powder (a), after thermal treatment in presence of H2 (50 ml min-1); TEM micrograph of size-selected particles collected at the outlet of the DMA(b), after thermal treatment in presence of H2 (50 ml min-1).

261

4.3. ZnO impregnated sawdust incineration. Another sample from the same batch of ZnO

262

impregnated sawdust was used for a standard incineration experiment. According to the

263

similarity laws described by Wochele et al.,32 the typical 100 min program of a municipal solid

264

waste incinerator was reduced to 24 min for a lab-scale incinerator (brown line in Figure 6), in

265

order to observe the complete conversion of carbon compound to CO2. The eventual release of

266

ZnO particles at these particular conditions is investigated by RDD-SMPS-ICP-MS.


15


Page 16 of 24

267

In opposition to the previous experiments, a clear PSD ranging from 50 nm to 240 nm was

268

obtained for Zn-containing particles, as reported in Figure 6b. Additionally, what stands out in

269

Figure 6 is the good correlation between the shape of volume-related PSD and

270

mass intensity.

66

Zn ICP-MS

Figure 6. Volume related particle size distribution of ZnO impregnated sawdust (a) and ICP-MS mass intensity of 66

Zn (b), in presence of O2 (50 ml min-1). Temperature profile (in brown line) lasting only 24 minutes: drying step,

fast pyrolysis from 300°C to 900°C, combustion at 900°C and afterburning from 900°C to 400°C. Measurements started simultaneously with the furnace’s temperature program but RDD channel opened only for combustion and afterburning steps, 6 minutes after the start.

271

These results indicate that with this temperature program applied, an effective reduction occurs.

272

As suggested in the initial paragraph of this section, reduction of ZnO by C or CO is a quite

273

known process that leads to the formation of Zn. However, from the above presented SMPS-ICP-

274

MS measurement, it is only possible to relate the sub-micron particles emitted during combustion

275

to generally Zn-containing particles, without having specific information on the molecular

276

composition and morphology. Therefore, the size-selected particles contributing to the intensity

277

measured in Figure 6b were collected at the outlet of the DMA through aspiration EM sampler

278

and further characterized with TEM.


16

Page 17 of 24


279

First of all, the intercorrelation between PSD measured with TEM and SMPS appears clear. The

280

size distribution is confirmed in a range of smaller particles of 20 nm to the statistically

281

significant presence of 50, 70, and 85 nm particles, together with bigger agglomerate with an

282

aerodynamic diameter up to around 240 nm probably formed in the cold zone of the quartz tube

283

or during particle collection. In Figure 7, EDX spectra and indexation of electron diffraction

284

pattern of crystalline material interestingly correspond to ZnO with hexagonal crystalline

285

structure. In Figure 7c the intensity of the elements labeled in light blue are attributed to the EM

286

grid itself.

Figure 7. TEM micrograph (a), indexation of electron diffraction pattern (b) and EDX spectrum (c) on size-selected particles collected after the DMA.

287

Different experiments confirmed the results shown in the TEM characterization. The bigger

288

agglomerates in Figure 7a and 8a generates powder like electron diffraction pattern (Figure 7b

289

and 8b), while single crystals (Figure 8c) show diffraction only on certain planes indicated by the

290

reported Miller indexes (Figure 8d), which confirm interplanar distance matching hexagonal

291

ZnO structure.

292

In the supporting information we suggested a mechanism of reduction and further re-oxidation to

293

give ZnO as the final product, together with thermochemical calculations in which the input

294

concentrations are based on real ratios between C and Zn in impregnated wood (Figures S12-13).


17


Page 18 of 24

Figure 8. TEM micrographs of agglomerated particles (a) and (c); indexation of electron diffraction patterns corresponding to selected area in orange (b) and (d).

295

4.5. Wood fly ashes. Elled et al. detected a bi-modal PSD for wood fly ashes with submicron

296

and micrometer range, respectively. The first PSD mode (submicron particles) is attributed to

297

inorganic matter while the second one derives from fragmentation of mineral and char.29 The

298

formation of ash particles and aggregates has been previously shown to occur for condensation

299

of organic material or inorganic vapors on the generated ZnO particles. Alkali salts are generally

300

found on the surface of these newly formed particles if efficient combustion occurs.15 Since we

301

focus in our experiments principally on the first peak of this bi-modal distribution only smaller


18

Page 19 of 24


302

inorganic particles were detected, while bigger agglomerates of organic carbon are directed to

303

the excess gas during DMA selection.

304

4.6. Fate of ZnO generated particles - Environmental relevance. The current data highlight

305

the importance of determining the fate of nano-object in fly ashes. Although the work presented

306

by Walser et al. might suggest the entrapment in the bottom ashes of most of the nanoparticles

307

(CeO2) injected in a waste incinerator,2 that case is not considering metal nanoparticles and non-

308

nano particulate metals in waste which may re-form as incidental nanoparticles.3 As explained by

309

Roes et al. secondary nano-objects might be formed depending on the conditions present in the

310

incinerator. Moreover, in the case of PSD below 100 nm, a complete removal from flue gas is

311

not ensured.4

312

This study confirms the presence of incidental ZnO nano-objects in the fine particles fraction

313

upon combustion of a nano-waste model, with dimensions also below 100 nm detected by the

314

use of an on-line analytical technique. Moreover, this finding proves the possibility of formation

315

of secondary nano-objects that can decompose and reform as new species, such as oxide. Due to

316

the demonstrated toxicity of ZnO,33,34,35 a particular attention should be addressed to emissions

317

by small and medium scale plants below 1 MW, in which efficient filtering systems are not

318

always implemented. Therefore, new experiments with larger burners and in particular, test on

319

particle removal are suggested.

320

ASSOCIATED CONTENT

321

Supporting Information. Zinc oxide powder characterization; Zinc oxide suspension

322

characterization; sawdust multi-elemental analysis in ICP-MS; ZnO impregnated sawdust

323

measurements with ICP-MS and calculation on efficiency of the impregnation; description of the


19


Page 20 of 24

324

two temperature programs applied for thermal treatment and incineration experiments; data

325

processing and contour plots explanation; comparison between PSD of ZnO suspension and PAA

326

solution; ZnO suspension thermal treatment in oxidizing and reducing conditions; ZnO powder

327

thermal treatment in reducing conditions; HSC thermochemical calculations; beech wood saw

328

dust incineration experiment; sampling point and detailed view of particle collection device, ZnO

329

suggested mechanism of reduction and further re-oxidation. This material is available free of

330

charge via the Internet at http://pubs.acs.org.

331

AUTHOR INFORMATION

332

Corresponding Author:

333

*E-mail: [email protected]

334

*E-mail: [email protected]

335

Author Contributions: The paper was written with contributions from all authors. Debora

336

Foppiano took the lead in compiling the first draft of the manuscript. All authors have given

337

approval to the final version. The incineration experiments and the RDD-SMPS-ICP-MS

338

measurements were carried out equally by Mohamed Tarik and Debora Foppiano. Elisabeth

339

Müller performed the TEM analysis and supervised Debora Foppiano for the indexation of the

340

electron diffraction pattern. Christian Ludwig gave his contribution as the principal investigator

341

of the research project.

342

Notes The authors declare no competing financial interest.

343

ACKNOWLEDGMENT


20

Page 21 of 24


344

Financial support was obtained from the Swiss National Science Foundation (Projects 139136

345

and 136890), the Swiss Nanoscience Institute (Argovia, Project NanoFil), and the Swiss

346

Competence Center for Energy Research – Biomass for Swiss Energy Future (SCCER

347

BIOSWEET) in cooperation with the Commission for Technology and Innovation CTI. The

348

authors would like to thank Marcel Hottiger for the construction of the particle collection cell.

349

Special thanks go to Dr. Andrea Testino, for his help during the ZnO suspension preparation and

350

characterization, and to Dr. Adrian Hess and Albert Schuler for the introductory explanation on

351

the RDD-SMPS-ICP-MS coupling and incineration experiments, respectively.

352

ABBREVIATIONS

353

PSD, particle size distribution; PAA, polyacrylic acid; SMPS, scanning mobility particle sizer;

354

ICP-MS, inductively coupled mass spectrometer; RDD, rotating disk diluter; SEM, scanning

355

electron microscopy; EDS, energy-dispersive X-ray spectroscopy; TEM, transmission electron

356

microscopy; SI, supporting information.

357

REFERENCES

358

(1)

359 360

Sanderson, P.; Delgado-Saborit, J. M.; Harrison, R. M. A review of chemical and physical characterisation of atmospheric metallic nanoparticles. Atmos. Environ. 2014, 94, 353–365.

(2)

Walser, T.; Limbach, L. K.; Brogioli, R.; Erismann, E.; Flamigni, L.; Hattendorf, B.; Juchli, M.;

361

Krumeich, F.; Ludwig, C.; Prikopsky, K.; et al. Persistence of engineered nanoparticles in a

362

municipal solid-waste incineration plant. Nat. Nanotechnol. 2012, 7 (8), 520–524.

363

(3)

364 365

nanoparticles. Nat. Nanotechnol. 2012, 7 (8), 487–488. (4)

366 367

370

Roes, L.; Patel, M. K.; Worrell, E.; Ludwig, C. Preliminary evaluation of risks related to waste incineration of polymer nanocomposites. Sci. Total Environ. 2012, 417–418, 76–86.

(5)

368 369

Wiesner, M. R.; Plata, D. L. Environmental, health and safety issues: Incinerator filters

Johnson, D. R. Nanometer-sized emissions from municipal waste incinerators: A qualitative risk assessment. J. Hazard. Mater. 2016, 320, 67–79.

(6)

Jakob, A.; Stucki, S.; Struis, R. P. W. J. Complete heavy metal removal from fly ash by heat treatment: Influence of chlorides on evaporation rates. Environ. Sci. Technol. 1996, 30 (11), 3275–


21


371 372

Page 22 of 24

3283. (7)

Merdzan, V.; Domingos, R. F.; Monteiro, C. E.; Hadioui, M.; Wilkinson, K. J. The effects of

373

different coatings on zinc oxide nanoparticles and their influence on dissolution and

374

bioaccumulation by the green alga, C. reinhardtii. Sci. Total Environ. 2014, 488–489 (1), 316–324.

375

(8)

376 377

potential exposure and hazard. Nanotoxicology 2010, 4 (1), 15-41. (9)

378 379

(10)

Jakob, A.; Stucki, S.; Kuhn, P. Evaporation of Heavy Metals during the Heat Treatment of Municipal Solid Waste Incinerator Fly Ash. Environ. Sci. Technol. 1995, 29 (9), 2429–2436.

(11)

382 383

Ludwig, C.; Lutz, H.; Wochele, J.; Stucki, S. Studying the evaporation behavior of heavy metals by thermo-desorption spectrometry. Anal. Bioanal. Chem. 2001, 371 (8), 1057–1062.

380 381

Osmond, M. J.; McCall, M. J. Zinc oxide nanoparticles in modern sunscreens: An analysis of

Burtscher, H. Physical characterization of particulate emissions from diesel engines: A review. J. Aerosol Sci. 2005, 36 (7), 896-932.

(12)

Petrović, V. S.; Janković, S. P.; Tomić, M. V.; Jovanović, Z. S.; Knežević, D. M. The possibilities

384

for measurement and characterization of diesel engine fine particles - A review. Therm. Sci. 2011,

385

15 (4), 915-938.

386

(13)

Krystek, P.; Ulrich, A.; Garcia, C. C.; Manohar, S.; Ritsema, R. Application of plasma

387

spectrometry for the analysis of engineered nanoparticles in suspensions and products. J. Anal. At.

388

Spectrom. 2011, 26 (9), 1701–1721.

389

(14)

Wellinger, M.; Wochele, J.; Biollaz, S. M. A.; Ludwig, C. Online elemental analysis of process

390

gases with ICP-OES: A case study on waste wood combustion. Waste Manag. 2012, 32 (10),

391

1843–1852.

392

(15)

Tissari, J.; Sippula, O.; Torvela, T.; Lamberg, H.; Leskinen, J.; Karhunen, T.; Paukkunen, S.;

393

Hirvonen, M. R.; Jokiniemi, J. Zinc nanoparticle formation and physicochemical properties in

394

wood combustion - Experiments with zinc-doped pellets in a small-scale boiler. Fuel 2015, 143,

395

404–413.

396

(16)

397 398

Hess, A.; Tarik, M.; Ludwig, C. A hyphenated SMPS–ICPMS coupling setup: Size-resolved element specific analysis of airborne nanoparticles. J. Aerosol Sci. 2015, 88, 109–118.

(17)

Tarik, M.; Foppiano, D.; Hess, A.; Ludwig, C. A Practical Guide on Coupling a Scanning Mobility

399

Sizer and Inductively Coupled Plasma Mass Spectrometer ( SMPS-ICPMS ). J. Vis. Exp. 2017,

400

No. June, 1–9.

401

(18)

Laden, F.; Schwartz, J.; Speizer, F. E.; Dockery, D. W. Reduction in fine particulate air pollution

402

and mortality: Extended follow-up of the Harvard Six Cities Study. Am. J. Respir. Crit. Care Med.

403

2006, 173 (6), 667–672.

404

(19)

Limbach, L. K.; Li, Y.; Grass, R. N.; Brunner, T. J.; Hintermann, M. A.; Muller, M.; Gunther, D.;


22

Page 23 of 24


405

Stark, W. J. Oxide Nanoparticle Uptake in Human Lung Fibroblasts: Effects of Particle Size,

406

Agglomeration, and Diffusion at Low Concentrations. Environ. Sci. Technol. 2005, 39 (23), 9370–

407

9376.

408

(20)

Dandley, E. C.; Taylor, A. J.; Duke, K. S.; Ihrie, M. D.; Shipkowski, K. A.; Parsons, G. N.;

409

Bonner, J. C. Atomic layer deposition coating of carbon nanotubes with zinc oxide causes acute

410

phase immune responses in human monocytes in vitro and in mice after pulmonary exposure.

411

Part. Fibre Toxicol. 2015, 13 (1), 29.

412

(21)

413 414

Allen, T.Dispersion of powders. In Particle size measurement;Springer: Netherlands, 1990, pp. 285-309.

(22)

Chung, S. J.; Leonard, J. P.; Nettleship, I.; Lee, J. K.; Soong, Y.; Martello, D. V.; Chyu, M. K.

415

Characterization of ZnO nanoparticle suspension in water: Effectiveness of ultrasonic dispersion.

416

Powder Technol. 2009, 194 (1–2), 75–80.

417

(23)

418 419

Bährle, C. Formation , Structure , and Reactivity of Organic Radicals in Lignin and its Depolymerization Products, ETH Zurich, 2016.

(24)

Hess, A.; Tarik, M.; Foppiano, D.; Edinger, P.; Ludwig, C. Online Size and Element Analysis of

420

Aerosol Particles Released from Thermal Treatment of Wood Samples Impregnated with Different

421

Salts. Energy and Fuels 2016.

422

(25)

Hess, A.; Tarik, M.; Foppiano, D.; Edinger, P.; Ludwig, C. Online Size and Element Analysis of

423

Aerosol Particles Released from Thermal Treatment of Wood Samples Impregnated with Different

424

Salts. Energy and Fuels 2016, 30 (5), 4072–4084.

425

(26)

426 427

Sampling by TEM Grid Filtration. Aerosol Sci. Technol. 2013, 47 (7), 767–775. (27)

428 429

R’ mili, B.; Le Bihan, O. L. C.; Dutouquet, C.; Aguerre-Charriol, O.; Frejafon, E. Particle

Tancredi, N.; Gabús, M.; Yoshida, M. I.; Suúrez, A. C. Thermal studies of wood impregnated with ZnCl2. European Journal of Wood and Wood Products. 2016, pp 1–6.

(28)

Fliegel, D.; Frei, C.; Fontaine, G.; Hu, Z.; Gao, S.; Günther, D. Sensitivity improvement in laser

430

ablation inductively coupled plasma mass spectrometry achieved using a methane/argon and

431

methanol/water/argon mixed gas plasma. Analyst 2011, 136 (23), 4857–5092.

432

(29)

433

Elled, A. L.; Åmand, L. E.; Eskilsson, D. Fate of zinc during combustion of demolition wood in a fluidized bed boiler. Energy and Fuels 2008, 22 (3), 1519–1526.

434

(30)

Shamsuddin, M. Physical Chemistry of Metallurgical Processes; 2016; Vol. 44.

435

(31)

Weidenkaff, A.; Steinfeld, A.; Wokaun, A.; Auer, P. O.; Eichler, B.; Reller, A. Direct solar

436

thermal dissociation of zinc oxide: condensation and crystallization of zinc in the presence of

437

oxygen. Sol. Energy 1999, 65 (1), 59–69.

438

(32)

Wochele, J.; Ludwig, C.; Stucki, S. Heavy metal volatilization during MSW incineration. In


23


439 440

Municipal Solid Waste Management; Springer: New York, Berlin, Heidelberg, 2003; pp 211-220. (33)

Sliwinska, A.; Kwiatkowski, D.; Czarny, P.; Milczarek, J.; Toma, M.; Korycinska, A.; Szemraj, J.;

441

Sliwinski, T. Genotoxicity and cytotoxicity of ZnO and Al2O3 nanoparticles. Toxicol. Mech.

442

Methods 2015, 25 (3), 176–183.

443

(34)

Uski, O.; Jalava, P. I.; Happo, M. S.; Torvela, T.; Leskinen, J.; Mäki-Paakkanen, J.; Tissari, J.;

444

Sippula, O.; Lamberg, H.; Jokiniemi, J.; et al. Effect of fuel zinc content on toxicological

445

responses of particulate matter from pellet combustion in vitro. Sci. Total Environ. 2015, 511,

446

331–340.

447

Page 24 of 24

(35)

Sharma, V.; Anderson, D.; Dhawan, A. Zinc oxide nanoparticles induce oxidative DNA damage

448

and ROS-triggered mitochondria mediated apoptosis in human liver cells (HepG2). Apoptosis

449

2012, 17 (8), 852–870.

450 451

For Table of Contents Only

452


24

Emissions of Secondary Formed ZnO Nano ... - ACS Publications

Recommend Documents