Accepted Manuscript Title: THERMAL AND ELECTRICAL PROPERTIES OF STARCH-GRAPHENE OXIDE NANOCOMPOSITES IMPROVED BY PHOTOCHEMICAL TREATMENT Author: Priscilla P. Peregrino Maria J.A. Sales Mauro F.P. da Silva Maria A.G. Soler Luiz F.L. da Silva Sanclayton G.C. Moreira Leonardo G. Paterno PII: DOI: Reference:

S0144-8617(14)00128-3 http://dx.doi.org/doi:10.1016/j.carbpol.2014.02.008 CARP 8574

To appear in: Received date: Revised date: Accepted date:

11-1-2014 1-2-2014 3-2-2014

Please cite this article as: Peregrino, P. P., Sales, M. J. A., Silva, M. F. P., Soler, M. A. G., Silva, L. F. L., Moreira, S. G. C., & Paterno, L. G.,THERMAL AND ELECTRICAL PROPERTIES OF STARCH-GRAPHENE OXIDE NANOCOMPOSITES IMPROVED BY PHOTOCHEMICAL TREATMENT, Carbohydrate Polymers (2014), http://dx.doi.org/10.1016/j.carbpol.2014.02.008 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

THERMAL AND ELECTRICAL PROPERTIES OF STARCH-GRAPHENE

2

OXIDE

3

TREATMENT

NANOCOMPOSITES

IMPROVED

BY

PHOTOCHEMICAL

4 Priscilla P. Peregrinoa, Maria J. A. Salesa, Mauro F. P. da Silvab, Maria A. G. Solerc,

6

Luiz F. L. da Silvad, Sanclayton G. C. Moreirad, Leonardo G. Paternoa*

7 8

a

9

b

10

c

11

d

12

PA 66075-900, Brazil

cr

Universidade de Brasília, Instituto de Química, Brasília DF 70910-900, Brazil

Pontifícia Universidade Católica de São Paulo, FCET, São Paulo SP 01303-050, Brazil

us

Universidade de Brasília, Instituto de Física, Brasília DF 70910-900, Brazil

Universidade Federal do Pará, Instituto de Ciências Exatas e Naturais (ICEN), Belém

an

13 *corresponding author.

15

Tel.: +55 61 3107 3869; FAX: +55 61 3107 3900

16

E-mail address: [email protected]

21 22 23 24 25 26 27 28

te

20

Ac ce p

19

d

17

M

14

18

ip t

5

29 30 31 32 33

Highlights for Review ! bionanocomposites of starch filled with graphene oxide

1 Page 1 of 30

34 35 36 37 38

! reagentless photochemical treatment to reduce graphene oxide and improve thermal and electrical properties of nanocomposites ! assessment of molecular-level interaction of starch and graphene oxide by ATR-FTIR and thermogravimetric analysis

39

ip t

40 ABSTRACT

42

Bionanocomposite films have been prepared by casting an aqueous suspension of

43

acetylated starch (ST) and poly(vinyl alcohol) (PVA) loaded with graphene oxide (GO).

44

A photochemical and reagentless method has been successfully performed to convert

45

the GO phase into reduced graphene oxide (RGO). The nanocomposites have displayed

46

improved thermal and electrical properties when the amount of the GO phase is

47

increased and properly converted to RGO. The molecular-level interactions between

48

components are mainly hydrogen-bonding type according to attenuated total

49

reflectance-Fourier transform infrared (ATR-FTIR) and Raman spectroscopies, as well

50

as thermogravimetric analysis (TGA). Scanning electron microscopy (SEM) has

51

confirmed the effective mixing between GO and the ST-PVA matrix. The thermal

52

diffusivity and electrical resistivity of ST-GO nanocomposites have increased one order

53

and decreased two orders of magnitude, respectively, after the photochemical treatment.

54

These findings have confirmed the effectiveness of the proposed approach to produce

55

starch-based nanocomposites with improved thermal and electrical properties.

us

an

M

d

te

Ac ce p

56

cr

41

57

Key-words: bionanocomposites, starch, graphene oxide, thermal properties, electrical

58

properties, photochemical reduction.

59 60

1. Introduction

61

Bionanocomposites represent today an environmentally friendly strategy for the

62

production of novel materials while employing raw components from the plenty source

63

provided by biomass. These systems are comprised by a continuous biopolymer matrix

64

loaded with nanofillers, usually inorganic in nature. While the biopolymer matrix

65

ensures the material’s biodegradability, nanofillers provide the required thermal,

66

mechanical, and electrical properties, usually very poor in the biopolymer alone

67

(Darder, Aranda, & Ruiz-Hitzky, 2007).

2 Page 2 of 30

Among several biopolymers for bionanocomposites starch is very appealing

69

because of its low cost and availability from different natural sources such as vegetables

70

and cereals (Paraginski et al., 2014). Besides the nourishing value, the film forming

71

ability has given to starch a more remarkable role, for example in food packaging.

72

Starch-based films are biodegradable and edible and, therefore, potential substitutes for

73

many synthetic packaging materials and pose negligible harm to the environment (Tang

74

et al., 2012).

ip t

68

On the other hand, starch exhibits poor thermal and electrical properties, which

76

restrict its widespread use. This apparent limitation can be, however, circumvented by

77

adding nanofillers such as graphene oxide (GO) sheets. GO can be considered a

78

precursor for the preparation of graphene-based materials (Zhu et al., 2010). Besides

79

that, GO is highly hydrophilic and dispersible in water, features that readily enable

80

mixing and compatibility of GO and starch, as well as with other polymeric matrixes

81

(Jaber-Ansari & Hersam, 2012). In fact, this possibility has been recently verified by

82

different research groups. For example, Li et al. (Li, Liu & Ma, 2011) and Ma et al. (Ma

83

et al., 2012) have carried out a pioneer research on nanocomposites made of starch and

84

GO when they have found that GO increases the Young modulus and the thermal

85

stability of pure starch. A subsequent investigation has been conducted by He and

86

colleagues (He et al., 2013) when they have observed that GO prevents starch from

87

degradation in both acidic and alkaline environments. Rodriguez-Gonzalez et al.

88

(Rodríguez-González et al., 2012) have observed that the mechanical properties (storage

89

modulus) of a chitosan-starch matrix is substantially improved (about 900 %) with only

90

0.5 wt % of GO. Ashori's group have introduced the melting processing of

91

polypropylene composites with either sugar cane bagasse or wood flours, both enriched

92

with graphene nanoplatelets (Sheshmani, Ashori & Fashapoyeh, 2013; Chaharmahali et

93

al., 2014). They have observed that there is an optimum graphene content for the

94

bionanocomposites to reaching the best mechanical properties.

us

an

M

d

te

Ac ce p

95

cr

75

However, another main interest on introducing GO into common polymeric

96

matrices is to provide or enhance their optoelectronic properties. GO is usually obtained

97

after chemical exfoliation of graphite which is a more affordable and readily available

98

carbon source (Jaber-Ansari & Hersam, 2012). GO resembles graphene by its

99

monoatomic dimension but it also differs significantly from it because of oxygen

100

bearing functionalities such as hydroxyl, carbonyl, carboxyl, and epoxy. These

101

functionalities disrupt part of the sp2 conjugation and turn GO into an electrical 3 Page 3 of 30

insulating material. However, part of optical, electronic and many other properties of

103

pristine graphene can be restored after proper reduction of GO (Stankovich, et al., 2007;

104

Tung et al., 2009). Concerning that, we have produced starch-based nanocomposite

105

films in which GO is introduced and photochemically converted into its reduced form,

106

namely reduced graphene oxide (RGO). The reduced nanocomposite has thermal and

107

electrical properties improved after the photochemical process, which is easy to perform

108

and does not demand for hazardous chemicals such as hydrazine or sodium

109

borohydride.

cr

ip t

102

Herein, we describe the preparation and investigate the properties of

111

bionanocomposite cast films made of starch filled with GO sheets (ST-GO). UV-vis,

112

attenuated total reflectance-Fourier transform infrared (ATR-FTIR) and Raman

113

spectroscopies are employed to determine the structure and infer about the molecular-

114

level interaction between the starch matrix and GO sheets. An additional study

115

performed with thermogravimetric analysis (TGA) provides further evidence for the

116

interaction of bionanocomposites' components. In order to restore part of the

117

outstanding properties of graphene we have submitted the bionanocomposite films to

118

UV radiation (254 nm). After a proper irradiation time, the thermal diffusivity and

119

electrical properties of ST-GO films are significantly improved, due to partial reduction

120

of GO, as confirmed by Raman and UV-vis spectroscopies.

121

te

d

M

an

us

110

2. Experimental

123

2.1. Materials

124

Graphite powder (Union Carbide, SP-1 grade, 100 μm), acetylated starch (Avebe Guaíra

125

Amidos, Brazil), poly(vinyl alcohol)-PVA (Sigma-Aldrich, 89,000-98,000 g.mol-1),

126

H2SO4 98%, HNO3 67%, KClO3, KHCO3, HCl 37%, NH4OH 47%, were purchased

127

from Sigma-Aldrich, Merck and Carlo Erba, all analytical grade or better and used as

128

received. Ultrapure water (18.2 M! .cm) purified by a Milli-Pore Milli-Q system was

129

used throughout all procedures. Graphene oxide (GO) was obtained via a two step

130

procedure. In the first, graphite powder was oxidized in a mixture of 1:2 H2SO4/HNO3

131

(2:1 v/v) and KClO3 as originally described by Staudenmaier (Staudenmaier, 1898) with

132

some modification as described elsewhere (Silva et al., 2010). In the second step, the as-

133

obtained graphitic oxide was suspended in NH4OH aqueous solution (pH 10) under

134

sonication (150 W, 1 h). The resulting GO suspension was centrifuged at 15,000 rpm

135

for 20 min to remove remaining solids and stored for the nanocomposites’ preparation.

Ac ce p

122

4 Page 4 of 30

136 137

2.2. Preparation of Starch-GO (ST-GO) nanocomposites. 2.5 g of acetylated starch (ST) and 0.15 g of PVA were suspended together in 40

139

mL of deionized water. PVA was used as an additional plasticizer (besides water) for

140

starch. The milky ST-PVA suspension was kept under magnetic stirring at 90 oC for 1 h

141

in order to gelatinize starch and ensure their complete mixing. In another vessel,

142

additional 0.05 g of PVA were dissolved in deionized water (70 oC) under magnetic

143

stirring until reaching its complete dissolution. A fixed volume of the GO aqueous

144

suspension was then added to this PVA solution under magnetic stirring at 70 oC until

145

obtaining a homogenous solution. This procedure was adopted to enhance compatibility

146

between GO and starch and to avoid GO flocculation. Four different PVA-GO solutions

147

with different GO concentrations were prepared. A fifth PVA solution absent of GO

148

was prepared and used for the preparation of the control film. For the preparation of ST-

149

GO nanocomposite films, the PVA-GO solution was added slowly to the ST-PVA

150

suspension under magnetic stirring at 60 oC. The stirring and temperature were kept for

151

more 1 h after addition of the PVA-GO solution. The resulting pale yellow suspension

152

was then cast onto Petri dishes (diameter: 10 cm) and left drying in oven (60 oC) for 24

153

h. The control film (absent of GO) was prepared under the same procedure. Five

154

nanocomposite samples were prepared, labeled as: ST-C (control), ST-GO-010, ST-

155

GO-025, ST-GO-050, and ST-GO-075. The numbers refer to the weight percentage of

156

GO in the nanocomposite (regarding the starch mass).

158 159

cr

us

an

M

d

te

Ac ce p

157

ip t

138

2.3. UV treatment of ST-GO samples In order to induce the photochemical reduction of GO, ST-GO samples (1.0 x

2

160

2.5 cm ) were exposed to UV light (254 nm) (Boitton Instrumentos, Brazil) for different

161

periods of time up to 2 hours. UV-vis spectra of films were obtained after different

162

irradiation time to monitor the photochemical effect. For comparison, the control film

163

(absent of GO) underwent the same treatment and monitoring. The ATR-FTIR and

164

Raman spectra, thermal diffusivity, and current vs. potential curves of un- and treated

165

samples were measured in order to evaluate the effectiveness of the photochemical

166

treatment.

167 168

2.4. Film characterizations

5 Page 5 of 30

169

The chemical structure of nanocomposite films was assessed by UV-vis (Varian

170

Cary 5000), ATR-FTIR (Varian 640IR) and Raman (Jobin Yvon T 6400, 514 nm laser

171

line, 0.3 mW) spectroscopies. Film’s morphology (surface and cross-section areas) was evaluated by scanning

173

electron microscope (JEOL 440A). Film’s cross-section was made after submitting

174

films to liquid nitrogen and further cracking. Samples were attached to brass stubs with

175

the aid of conductive carbon tape and gold-coated by sputtering.

ip t

172

Thermogravimetry (TGA) and derivative thermogravimetry (DTG) analysis

177

were conducted in a DTG-60H Shimadzu thermoanalyzer. Powdered samples (~ 5 mg)

178

were filled in platinum pans and evaluated under nitrogen atmosphere (30 mL.min-1), at

179

heating rate of 10 °C.min-1.

us

cr

176

The thermal diffusivity of film samples was determined by photopyroelectric

181

spectroscopy (PPS) as described by Mandelis and Zver (Mandelis & Zver, 1985), with

182

the aid of a home-made experimental setup. The PPS working principle bases on the

183

irradiation of a solid sample with a monochromatic source and measurement of that part

184

of optical energy that is converted into heat by non-radiative deexcitation processes

185

within the solid. According to the model described by Mandelis and Zver, a non-

186

uniform variation on temperature leads to a voltage drop in the pyroelectric transductor.

187

The voltage drop is related to the sample’s thermal diffusivity, according to equation 1

188

as follows:

M

d

te

Ac ce p

189

an

180

190

where ! s is the sample’s nonradiative conversion efficiency; !

191

diffusivity of the pyroelectric transducer and sample, respectively; kp is the transducer’s

192

thermal conductivity; bsp is the ratio between the product of sample’s and transducer’s

193

thermal conductivity and diffusivity (= ks! s/kp! p), ! is the angular frequency of the

194

electrical field, and Ls is the sample`s thickness. In addition, the parameter A (=

195

pI0/2k! 0), where p is the pyroelectric coefficient of the transducer, I0 is the incident light

196

source irradiance, k is the transducer’s dielectric constant, and ! 0 is the vacuum

197

permittivity (= 8.854 x 10-12 C V-1m-1). In the present study, samples` thicknesses were

198

evaluated with a Mitutoyo micrometer, model M110-25. The sample`s thicknesses were

199

about 12.0 ± 0.5 ! m.

p

and ! s are the thermal

6 Page 6 of 30

The electrical properties of the ST-GO samples, before and after photochemical

201

reduction, were determined by current versus potential curves obtained with the HP

202

4156A Semiconductor Parameter Analyzer. The ST-GO and ST-RGO films (1.0 x 2.5

203

cm2) were coated with four equidistant silver epoxy contacts (MG Chemicals 8331

204

Conductive Epoxy Adhesive) through which gold microprobes were connected to

205

perform electrical measurements. A fixed current was applied through the two external

206

contacts and the voltage drop across the two inner contacts was thus measured.

ip t

200

208

cr

207 3. Results and Discussion

us

209

3.1. Structure and morphology of ST-GO nanocomposite films

211

Figure 1 provides UV-vis absorption spectra along with digital snapshots of ST-GO

212

nanocomposites with different amounts of GO. Similarly to the GO aqueous suspension

213

(not shown), the spectra of nanocomposites (Figure 1a) exhibit two maxima absorption

214

peaks located at 230 nm and 300 nm, which are ascribed to the ! ! ! * transition in

215

aromatic C=C bonds and the n! ! * transition in carbonyl groups, respectively (Li et al,

216

2008; Guardia et al., 2012). The nanocomposites also display the same brownish color

217

of GO suspension and they become darker as more GO is incorporated (Figure 1b). In

218

addition, the transparency as estimated by the film transmittance (400-700 nm) of

219

nanocomposites drastically reduces with incorporation of GO. For example, the ST-GO-

220

0.10 sample has a transmittance of 75%, which then reduces to 35 % in the ST-GO-

221

0.25, and then to 20% (ST-GO-0.50) and less than 20% for the ST-GO-0.75.

Ac ce p

te

d

M

an

210

222 7 Page 7 of 30

ip t

Figure 1. UV-vis absorption spectra (a) and digital snapshots (b) of ST-GO

225

nanocomposites with different GO loadings (in weight percentage), as indicated.

226

The ATR-FTIR spectra of control, pure GO and ST-GO-0.50 nanocomposite are

227

presented in Figure 2. The band referred to the stretching of O-H bond is clearly seen in

228

the three samples. While in the control and in the ST-GO-0.50 nanocomposite this band

229

peaks at 3360 cm-1, in pure GO the same band is more asymmetric and exhibits a

230

maximum at shorter wavenumber (3235 cm-1). In the range below 2000 cm-1, the

231

fingerprints of starch component are readily recognized in both control and ST-GO-0.50

232

nanocomposite. In fact, the ST-GO-0.50 nanocomposite spectrum (in green) is almost

233

indistinguishable from the control (in black) and shows none of the bands seen in the

234

GO spectrum (in red). The main vibration modes in the control sample (in black) have

235

been ascribed in accordance to literature (Flores-Morales, Jiménez-Estrada & Mora-

236

Escobedo, 2012) as follows: C=O stretching of the acetylated groups (1742 cm-1); O-H

237

bending (1652 cm-1); C-O stretching (1154 cm-1); C-O-H stretching (1082 cm-1); C-O-H

238

bending (1022 cm-1); and C-O-C bending of ! -1,4-glycosidic bond (998 cm-1). In the

239

spectrum of the ST-GO-0.50 nanocomposite (in green) most of bands have been

240

systematically shifted to shorter wavenumbers. This trend suggests the interaction of

241

starch and GO via hydrogen bonding. The assignments for bands in the ST-GO-0.50

242

sample have been made as follows: C=O stretching of the acetylated groups (1742 cm-

243

1

); O-H bending (1640 cm-1); C-O stretching (1150 cm-1); C-O-H stretching (1078 cm-

244

1

); C-O-H bending (1022 cm-1); and C-O-C bending of ! -1,4-glycosidic bond (997 cm-

245

1

). The greatest shift has been observed for the O-H bending, which is about 12 cm-1.

246

The spectrum of pure GO displays the oxygen-bearing functionalities that have been

247

introduced after graphite oxidation. In accordance to literature (Hontoria-Lucas et al.,

248

1995) the assignments for such bands have been made as follows: C=O stretching (1714

249

cm-1); C=C stretching (1612 cm-1); C-O-H bending (1417 cm-1); C-O-C stretching

Ac ce p

te

d

M

an

us

cr

223 224

8 Page 8 of 30

250

(1060cm-1); and =C-H bending (980 cm-1). The spectra of other ST-GO nanocomposites

251

have showed similar features and are not presented.

d

M

an

us

cr

ip t

252

Figure 2. ATR-FTIR spectra of control (black), pure GO (red), and ST-GO-0.50

255

nanocomposite (green) samples.

257

Ac ce p

256

te

253 254

258

Raman spectra of control, pure GO, and ST-GO-0.50 nanocomposite samples are shown

259

in Figure 3. As a general observation, GO bands screen any other bands from the starch

260

+ PVA matrix. The same trend has been observed for all other ST-GO compositions and

261

for that reason their corresponding spectra are not shown. This effect is ascribed to the

262

use of 514 nm excitation that matches better GO vibrations than starch ones. The

263

spectrum of the control sample features the C-O and C-C stretching as well as the C-O-

264

C deformation modes typical of the glycosidic bonds in starch. In accordance to

265

literature (Almeida et al., 2010; Rodriguez-Gonzalez et al., 2012) the main vibrations

266

(wavenumbers assigned in the spectrum) have been tentatively ascribed as follows:

267

bending of C-C-H and C-O-C bonds (858 and 925 cm-1); stretching of C-O and C-C

268

bonds (1129 and 1209 cm-1); bending of C-C-H, O-C-H and C-O-H bonds (1266 cm-1); 9 Page 9 of 30

stretching of C-O and bending of C-O-H bonds (1340 cm-1); bending of C-O-H bond

270

(1387 cm-1), and bending of C-H and C-O-H bonds (1462 cm-1). The distinct bands at

271

925 and 858 cm-1 are referred to vibrations originated from α-1,4 glycosidic linkages

272

and provide a fingerprint for the amylose component of starch. The spectrum of pure

273

GO (in red) displays two intense bands at 1340 cm-1 and 1600 cm-1. The first, the D-

274

band, corresponds to the breathing mode of k-point phonons of A1g symmetry while the

275

second one, the G-band, corresponds to the first-order scattering of the E2g phonon of

276

sp2 carbons. Both bands are regarded to the graphene structure according to literature

277

(Saito et al., 2011). The spectrum of the ST-GO-0.50 nanocomposite (in green)

278

exclusively features bands regarding GO. This screening effect has been discussed

279

above. The relative Raman D/G intensity ratio (ID/IG) is inversely proportional to

280

average size of the sp2 domains (Saito et al., 2011). In the case of our pure GO, ID/IG =

281

1.19, while for the ST-GO-0.50 nanocomposite, ID/IG = 0.84. The slight decrease of

282

ID/IG when going from pure GO to ST-GO-0.50 nanocomposite has no explanation so

283

far. As to other nanocomposites compositions, we have found no correlation between

284

ID/IG and GO content, which has remained almost constant: 0.86 (ST-GO-0.10); 0.81

285

(ST-GO-0.25); 0.84 (ST-GO-0.50); and 0.87 (ST-GO-0.75). After the photochemical

286

treatment, which will be discussed in detail at section 3.3, the ID/IG ratio in the

287

nanocomposite sample with 0.50 % wt of GO and now labeled ST-RGO-0.50,

288

underwent a slight increase to about 0.92. This increase corresponds to reduction

289

reactions of the GO phase, as already proposed by Stankovich and coworkers

290

(Stankovich et al., 2007).

Ac ce p

te

d

M

an

us

cr

ip t

269

10 Page 10 of 30

ip t cr us an Figure 3. Raman spectra of control (starch + PVA, in black), pure GO (in red), ST-GO-

293

0.50 nanocomposite (in green), and ST-RGO-0.50 nanocomposite (in blue). The

294

spectrum in blue refers to the one obtained after the ST-GO-0.50 has been submitted to

295

UV exposition for 160 min., as described in section 2.3.

te

d

M

291 292

296

Figure 4 presents SEM images of the surface (top) and cross section (bottom) of

298

nanocomposites. While the surface of all samples, including that of the control, is

299

relatively flat and smooth, the cross section after fracture is more irregular on the

300

nanocomposites with GO. This is because GO is well dispersed and strongly adhered to

301

the starch-PVA matrix.

Ac ce p

297

11 Page 11 of 30

ip t cr us an

302

Figure 4. SEM images of surface (top images; scale bar: 100 ! m) and cross section

304

(bottom images; scale bar: 10 ! m) of nanocomposites: control (a) and (d); ST-GO-0.10

305

(b) and (e); and ST-GO-0.50 (c) and (f).

M

303

307

d

306

3.2. Thermal properties

309

Figure 5 shows TG curves of the control and GO-containing nanocomposites. In Figure

310

5a, the thermal behavior of the ST-GO-0.50 nanocomposite is compared to that of the

311

control and pure GO. For pure GO (curve in black), one observes the decomposition at

312

200 oC associated to a mass loss of about 60 %. This event has been associated to the

313

dehydration of oxygenated groups accompanied by the partial reestablishment of the pi-

314

conjugated system (Silva et al., 2011). The latter event maintains the partially reduced

315

GO stable so that the residue after reaching 600 oC is about 30 %, which is greater than

316

that left after decomposition of pure starch or the ST-GO-0.50 nanocomposite. As to

317

pure starch, the TG curve (in blue) presents two main thermal events, the first one at ~

318

100 oC that corresponds to the water elimination and a second event, at 300 oC, which is

319

regarded to elimination of polyhydroxyl groups and starch depolymerization

320

(Schlemmer, Sales & Resck, 2010; Schlemmer & Sales, 2010). At 600 oC, almost all

321

starch is decomposed. The nanocomposites’s TG curve (in red) displays the elimination

Ac ce p

te

308

12 Page 12 of 30

322

of water as well as the decomposition of starch. However, the residual mass at 600 oC is

323

slightly greater than the control, most likely due to the residual, more stable GO.

324

In Figure 5b, TG curves of ST-GO nanocomposites are presented for evaluating

325

the effect of GO on their thermal stability. As one notes, as more GO is incorporated,

326

the onset for the main decomposition event occurs at lower temperature whereas the

327

residual mass at 600

328

decomposition is attributed to the intercalation of GO sheets in between starch chains.

329

This certainly reduces the intermolecular forces in starch as well as provides more

330

oxygen reactive species that contribute for the starch decomposition. This result has

331

been shared by other investigators working on similar systems (Ma et al., 2013; He et

332

al., 2013; Li et al., 2011). This hypothesis is corroborated further by data collected in

333

Table 1 (second column) that show that the more GO is introduced the lower is the

334

enthalpy for the water elimination. This observation along with ATR-FTIR data (Figure

335

2) allows one to conclude that GO sheets replace water molecules in the nanocomposite

336

while establishing hydrogen bonds with starch and PVA chains. Additionally, Table 1

337

(third column) collects thermal diffusivities measured accross the samples. The data

338

show that samples containing GO are better thermal conductors than the control sample.

339

This improvement has been ascribed to the few graphene domains remaining in GO

340

after graphite oxidation. The outstanding thermal conduction in deffect free graphene is

341

ascribed to the in-plane scattering of phonons (in the diffusion limit regime) (Pop,

342

Varshney & Roy, 2012).

te

d

M

an

us

cr

ip t

C increases. The lowering on temperature of the main

Ac ce p

343

o

344 13 Page 13 of 30

ip t cr us Figure 5. TG curves of (a) pure starch, pure GO, and ST-GO-0.50 nanocomposite and

347

(b) ST-GO nanocomposites with different loadings of GO, as indicated.

an

345 346

M

348 349

Table 1. Thermal properties of control and ST-GO nanocomposites. Thermal Diffusivity (x 10-4 cm2.s-1) Sample Enthalpy (J.g-1)

352

Before Reduction

After Reduction

Control

465.11

1.86

-

ST-GO-0.10

491.15

1.92

7.1

ST-GO-0.25

196.27

3.08

8.3

ST-GO-0.50

102.12

3.59

11.5

ST-GO-0.75

84.36

7.19

17.5

Ac ce p

351

te

d

350

353

3.3. Photochemical reduction of ST-GO nanocomposites

354

The network structure of GO sheets is highly attractive and potentialy promising for the

355

improvement of mechanical properties of biopolymers such as starch. However, its

356

optoelectronic properties are still very poor since the pi-conjugated system is partially

357

disrupted after the graphite oxidation. Thus, the restoring of such conjugated structure is

358

mandatory for the attainment of electrical conducting composites. Generally, reductive

359

chemicals such as hydrazine and sodium borohydride, have been successfully employed

360

for this task even though they involve severe health and environmental concerns. In this 14 Page 14 of 30

sense, milder reduction approaches have been pursued and photochemical procedures

362

have called special attention. The photochemical reduction of GO to RGO by UV light

363

has been made possible because GO sheets strongly absorb energy at this wavelength

364

and subsequent relaxation processes could trigger a local heating of the sheets that

365

favors their deoxygenation (Kumar et al., 2012; Guardia et al., 2012; Fabbri et al.,

366

2012). Such a reduction method would provide the means by which one can easily

367

produce RGO with relatively less impurity.

ip t

361

Considering those aspects, we have exposed the ST-GO-0.50 nanocomposite to

369

an UV light source (354 nm) and followed by UV-vis spectroscopy the structural

370

changes caused by the irradiation (see section 2.3). According to Figure 6, the ratio of

371

UV absorbances at 250 and 300 nm, which are ascribed to n! ! * (C=O) and ! ! ! *

372

(C=C) transitions, respectively, decreases with the exposure time. Moreover, the

373

snapshopts in the inset show that the nanocomposite initially brown have become

374

deeply dark at the end of treatment (after 160 min. of exposure). A control experiment

375

performed in the same conditions with the control film has not showed any darkening

376

effect of the starch matrix.

Ac ce p

te

d

M

an

us

cr

368

377 378 379

Figure 6. Monitoring of absorbance ratios (A250/A300) of the ST-GO-0.50

380

nanocomposite as a function of the exposure time to UV light (254 nm). Inset:

381

snapshots of nanocomposite sample at the beggining (0 min.) and after 160 minutes of

382

exposure. 15 Page 15 of 30

383 To confirm the effectiveness of the proposed photochemical reduction method we have

385

carried out Raman spectroscopy as well as measured the thermal diffusivity and

386

electrical properties of nanocomposites. As displayed by Figure 3, the ID/IG ratio

387

increases from 0.84 (ST-GO-0.50, in green) to 0.91 (ST-RGO-0.50, in blue) after the

388

photochemical treatment. It is known that the ID/IG ratio is inversely proportional to the

389

average size of sp2 domains (Tuinstra & Koenig, 1970). Thus, the increase that has been

390

observed herein suggests that new graphitic domains have been created and that they are

391

smaller in size to the ones present in GO before reduction, but more numerous. In

392

summary, part of the pi-conjugated system is recovered, which is consistent with the

393

GO ! RGO reduction. The photochemical treatment has increased thermal diffusivities

394

one order of magnitude, as quoted in Table 1 (fourth column), since RGO is a better

395

thermal conductor than GO.

an

us

cr

ip t

384

The reduction of GO phase within the nanocomposites was also probed by

397

electrical measurements. Figure 7 displays current versus voltage curves for the ST-GO-

398

0.50 nanocomposite before and after the photochemical treatment. For the same voltage

399

range, the current measured within the nanocomposite has increased two orders of

400

magnitude after the photochemical treatment. The different colors refer to curves

401

aquired through different contact pairs on the sample. It is possible to note that data drift

402

is very low when measurement contacts are changed because samples and contacts are

403

quite uniform. The overpotential seen for the sample before reduction (bottom curves)

404

has been credited to its highly insulating behavior. Based on these results, one can

405

conclude that the photochemical treatment is sufficient to reduce GO to RGO and to

406

improve the thermal and electrical properties of the ST-GO nanocomposite.

d

te

Ac ce p

407

M

396

16 Page 16 of 30

ip t cr us Figure 7. Current versus voltage curves for ST-GO-0.50 nanocomposite before and after

410

photochemical reduction. The different colors refer to curves aquired between two

411

different pairs of contacts on the sample.

M

an

408 409

412 413 4. Conclusions

415

Starch-graphene oxide nanocomposites (ST-GO) exhibit better thermal and electrical

416

properties when the amount of the GO phase is increased and properly converted to

417

reduced graphene oxide (RGO), respectively. Because water molecules are readily

418

replaced by GO sheets, the latter interact with the ST matrix through hydrogen bonds,

419

the enthalpy for water release from the nanocomposites become significant lower as

420

more GO is introduced. However, GO sheets also reduce the interaction between ST

421

domains and induce their degradation at lower temperatures. A reagentless

422

photochemical method for conversion of the GO phase into RGO increases the thermal

423

diffusivity and lowers the electrical resistance of nanocomposites to about one and two

424

orders of magnitude, respectively. Raman and UV-vis absorption spectroscopies

425

provide strong evidence for the partial restoring of the pi-conjugated system after the

426

photochemical treatment. In summary, the ST-GO nanocomposites and the

427

photochemical approaches described herein enable one for the production of low cost

428

and environmentally friendly nanomaterials with potential thermal and electrical

429

properties.

Ac ce p

te

d

414

430 17 Page 17 of 30

431

Acknowledgements

432

The financial support of Brazilian agencies CNPq (process # 308038/2012-6), CAPES,

433

and Finatec is greatfully acknowledged. The authors also wish to thank Dr. Roberto

434

Cavallari (PSI-EPUSP) for the electrical characterizations.

436

ip t

435 References

437

Almeida, M. R., Alves, R. S., Nascimbem, L. B., Stephani, R., Poppi, R. J. & Oliveira,

439

L. F. (2010). Determination of amylose content in starch using Raman spectroscopy and

440

multivariate calibration analysis. Analytical Bioanalytical Chemistry, 397, 2693-26701.

us

cr

438

441

Chaharmahali, M., Hamzeh, Y., Ebrahimi, G., Ashori, A., Ghasemi, I. (2014). Effects of

443

nano-graphene on the physico-mechanical properties of bagasse/polypropylene

444

composites. Polymer Bulletin, 71, 337–349.

M

445

an

442

Darder, M., Aranda, P. & Ruiz-Hitzky E. (2007). Bionanocomposites: A new concept of

447

ecological, bioinspired, and functional hybrid materials. Advanced Materials, 19, 1309–

448

1319.

449

te

d

446

450

Fabbri, P., Valentini, L., Bon, S. B., Foix, D., Pasquali, L., Montecchi, M. &

451

Sangermano,

452

photopolymerization of graphene oxide/acrylic resins mixtures. Polymer, 53, 6039-

453

6044.

(2012).

In-situ

Ac ce p

454

M.

graphene

oxide

reduction

during

UV-

455

Flores-Morales, A., Jiménez-Estrada, M. & Mora-Escobedo, R. (2012). Determination

456

of the structural changes by FTIR, Raman, and CP/MAS

457

retrograded starch of maize tortillas. Carbohydrate Polymers, 87, 61– 68.

458

13

C NMR spectroscopy on

459

Guardia, L., Villar-Rodil, S., Paredes, J. I., Rozada, R., Martinez-Alonso, A. & Tascon,

460

J. M. D. (2012). UV light exposure of aqueous graphene oxide suspensions to promote

461

their direct reduction, formation of graphene–metal nanoparticle hybrids and dye

462

degradation. Carbon, 50, 1014–1024.

463

18 Page 18 of 30

464

He, Y., Wang, X., Wu, D., Gong, Q., Qiu, H., Liu, Y., Wu, T., Ma, J. & Gao, J. (2013).

465

Biodegradable amylose films reinforced by graphene oxide and polyvinyl alcohol.

466

Materials Chemistry and Physics, 142, 1-11.

467 Hontoria-Lucas, C, Lopez-Peinado, A. J., Lopez-Gonzalez, J. D, Rojas-Cervantes, M.

469

L. & Martin-Aranda, R. M. (1995). Study of oxygen-containing groups in a series of

470

graphite oxides: physical and chemical characterization. Carbon, 33, 1585–1592.

ip t

468

cr

471

Jaber-Ansari, L. & Hersam, M. C. (2012). Solution-processed graphene materials and

473

composites. MRS Bulletin, 37, 1167-1175.

us

472 474

Kumar, P., Das, B., Chitara, B., Subrahmanyam, K. S., Gopalakrishnan, K., Krupanidhi,

476

S. B. & Rao, C. N. R. (2012). Novel radiation-induced properties of graphene and

477

related materials. Macromolecular Chemistry and Physics, 213, 1146−1163.

an

475

M

478

Li, D, Muller M. B., Gilje, S., Kaner R. B. & Wallace, G. G. (2008). Processable

480

aqueous dispersions of graphene nanosheets. Nature Nanotechnology, 3, 101–105.

481

d

479

Li, R., Liu, C. & Ma, J. (2011). Studies on the properties of graphene oxide reinforced

483

starch biocomposites. Carbohydrate Polymers, 84, 631–637.

Ac ce p

484

te

482

485

Ma, J., Liu, C., Li, R. & Wang, J. (2012). Properties and structural characterization of

486

oxide starch/chitosan/graphene oxide biodegradable nanocomposites. Journal of

487

Applied Polymer Science, 123, 2933–2944.

488 489

Ma, T., Chang, P. R., Zheng, P. & Ma, X. (2013). The composites based on plasticized

490

starch and graphene oxide/reduced graphene oxide. Carbohydrate Polymers, 94, 63–70.

491 492

Mandelis, A. & Zver, M. M. (1985).Theory of photopyroelectric spectroscopy of solids.

493

Journal of Applied Physics, 57, 4421-4430.

494 495

Paraginski, R. T., Vanier, N. L., Moomand, K., de Oliveira, M., Zavareze, E. R., Silva,

496

R. M., Ferreira, C. D. & Elias, M. C. (2014). Characteristics of starch isolated from

497

maize as a function of grain storage temperature. Carbohydrate Polymers, 102, 88–94. 19 Page 19 of 30

498 499

Pop, E., Varshney, V. & Roy, A. K. (2012). Thermal properties of graphene:

500

fundamentals and applications. MRS Bulletin, 37, 1273-1281.

501 Rodríguez-González, C., Martínez-Hernández, A. L., Castaño, V. M., Kharissova, O.

503

V., Ruoff, R. S. & Velasco-Santos, C. (2012). Polysaccharide nanocomposites

504

reinforced with graphene oxide and keratin-grafted graphene oxide. Industrial &

505

Engineering Chemistry Research, 51, 3619−3629.

cr

ip t

502

506

Saito, R. Hofmann, M., Dresselhaus, G., Jorio, A. & Dresselhaus, M. S. (2011). Raman

508

spectroscopy of graphene and carbon nanotubes. Advances on Physics, 60, 413-550.

us

507 509

Schlemmer, D., Sales, M. J. A. & Resck, I. S. (2010). Preparação, caracterização e

511

degradação de blendas PS/TPS usando glicerol e óleo de buriti como plastificantes.

512

Polímeros: Ciência e Tecnologia, 20, 6-13.

M

an

510

513

Schlemmer, D., & Sales, M. J. A. (2010). Thermoplastic starch films with vegetable oils

515

of Brazilian Cerrado, Journal of Therma. Analysis and Calorimetry, 99, 675-679.

te

516

d

514

Sheshmani, S., Ashori, A., Fashapoyeh, M. A. (2013). Wood plastic composite using

518

graphene nanoplatelets. International Journal of Biological Macromolecules, 58, 1–6.

519

Ac ce p

517

520

Silva, M. F. P. H., Costa, C. J. F., Triboni, E. R., Politi, M. J. & Isolani, P. C. (2010).

521

Synthesis and characterization of CeO2–graphene composite. Journal of Thermal

522

Analysis and Calorimetry, 102, 907-913.

523

Staudenmaier, L. (1898). Verfahren zur Darstellung der Graphitsäure, Berichte der

524

Deutschen Chemischen Gesellschaft, 31, 1481-1487.

525 526

Stankovich, S., Dikin, D. A., Piner, R. D., Kohlhaas, K. A., Kleinhammes, A., Jia, Y.,

527

Wu, Y., Nguyen, S. B. T. & Ruoff, R. S. (2007). Synthesis of graphene-based

528

nanosheets via chemical reduction of exfoliated graphite oxide. Carbon, 45, 1558–1565.

529

20 Page 20 of 30

530

Tang, X. Z., Kumar, P., Alavi, S. & Sandeep, K. P. (2012). Recent advances in

531

biopolymers and biopolymer-based nanocomposites for food packaging materials,

532

Critical Review in Food Science and Nutrition, 52, 426-442.

533 Tuinstra, F. & Koenig, J. L. (1970). Raman spectrum of graphite. Journal of Chemical

535

Physics, 53, 1126–1130.

ip t

534 536

Tung, V. C., Allen, M. J., Yang, Y. & Kaner, R. B. (2009). High-throughput solution

538

processing of large-scale graphene, Nature Nanotechnology, 4, 25-29.

cr

537

us

539

Zhu, Y. , Murali, S., Cai, W., Li, X., Suk, J. W., Potts, J. R. & Ruoff, R. S. (2010).

541

Graphene and graphene oxide: Synthesis, properties, and applications. Advanced

542

Materials, 22, 3906–3924.

an

540

Ac ce p

te

d

M

543

21 Page 21 of 30

Table 1. Thermal properties of control and ST-GO nanocomposites. Sample Enthalpy (J.g-1) Thermal Diffusivity (x 10-4 cm2.s-1) Before Reduction

After Reduction

465.11

1.86

-

ST-GO-0.10

491.15

1.92

7.1

ST-GO-0.25

196.27

3.08

8.3

ST-GO-0.50

102.12

3.59

ST-GO-0.75

84.36

7.19

ip t

Control

11.5

17.5

cr

543

544

te

d

M

an

us

Peregrino et al.

Ac ce p

545 546

22 Page 22 of 30

546

Figure Captions

547 548 549

Figure 1. UV-vis absorption spectra (a) and digital snapshots (b) of ST-GO nanocomposites

550

with different GO loadings (in weight percentage), as indicated.

ip t

551 552

Figure 2. ATR-FTIR spectra of control (black), pure GO (red), and ST-GO-0.50 nanocomposite

553

(green) samples.

cr

554

Figure 3. Raman spectra of control (starch + PVA, in black), pure GO (in red), ST-GO-

556

0.50 nanocomposite (in green), and ST-RGO-0.50 nanocomposite (in blue). The

557

spectrum in blue refers to the one obtained after the ST-GO-0.50 has been submitted to

558

UV exposition for 160 min., as described in section 2.4.

an

us

555

559

Figure 4. SEM images of surface (top images; scale bar: 100 ! m) and cross section (bottom

561

images; scale bar: 10 ! m) of nanocomposites: control (a) and (d); ST-GO-0.10 (b) and (e); and

562

ST-GO-0.50 (c) and (f).

M

560

563

Figure 5. TG curves of (a) pure starch, pure GO, and ST-GO-0.50 nanocomposite and (b) ST-

565

GO nanocomposites with different loadings of GO, as indicated.

te

d

564 566

Figure 6. Monitoring of absorbance ratios (A250/A300) of the ST-GO-0.50

568

nanocomposite as a function of the exposure time to UV light (254 nm). Inset:

569

snapshots of nanocomposite sample at the beggining (0 min.) and after 160 minutes of

570

exposure.

571

Ac ce p

567

572

Figure 7. Current versus voltage curves for ST-GO-0.50 nanocomposite before and after

573

photochemical reduction. The different colors refer to curves aquired between two different

574

pairs of contacts on the sample.

575 576

23 Page 23 of 30

ip t us

cr 581 582 583

Figure 1

Ac ce p

580

te

578 579

d

M

an

577

24 Page 24 of 30

ip t cr us an d

588

te

587

Figure 2

Ac ce p

586

M

584 585

25 Page 25 of 30

ip t cr us 593 594 595 596 597

d

M

an 592

Figure 3

Ac ce p

591

te

589 590

26 Page 26 of 30

ip t cr us an

598 599 Figure 4

M

600 601

605 606 607 608 609

te

604

Ac ce p

603

d

602

27 Page 27 of 30

ip t us

cr 611 612 613

Ac ce p

te

d

M

an

610

Figure 5

28 Page 28 of 30

ip t us

cr 616

an

614 615 Figure 6

M

617 618

622 623 624 625

te

621

Ac ce p

620

d

619

29 Page 29 of 30

ip t cr us an

626 627 628 Figure 7

M

629

Ac ce p

te

d

630

30 Page 30 of 30

Thermal and electrical properties of starch-graphene oxide nanocomposites improved by photochemical treatment.

Bionanocomposite films have been prepared by casting an aqueous suspension of acetylated starch (ST) and poly(vinyl alcohol) (PVA) loaded with graphen...
4MB Sizes 1 Downloads 4 Views