Turmeric dye extraction residue for use in bioactive film production: Optimization of turmeric film plasticized with glycerol

Accepted Manuscript Turmeric dye extraction residue for Use in bioactive film production: Optimization of turmeric film plasticized with glycerol B.C. Maniglia, R.L. de Paula, J.R. Domingos, D.R. Tapia-Blácido PII:

S0023-6438(15)30042-6

DOI:

10.1016/j.lwt.2015.07.025

Reference:

YFSTL 4819

To appear in:

LWT - Food Science and Technology

Received Date: 9 March 2015 Revised Date:

2 July 2015

Accepted Date: 11 July 2015

Please cite this article as: Maniglia, B.C., de Paula, R.L., Domingos, J.R., Tapia-Blácido, D.R., Turmeric dye extraction residue for Use in bioactive film production: Optimization of turmeric film plasticized with glycerol, LWT - Food Science and Technology (2015), doi: 10.1016/j.lwt.2015.07.025. 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.

ACCEPTED MANUSCRIPT TURMERIC DYE EXTRACTION RESIDUE FOR USE IN BIOACTIVE FILM

2

PRODUCTION: OPTIMIZATION OF TURMERIC FILM PLASTICIZED

3

WITH GLYCEROL

4

B.C. Maniglia, R. L. de Paula, J. R. Domingos, D.R. Tapia-Blácido*

5

Departamento de Química, Faculdade de Filosofia, Ciências e Letras, Universidade de

6

São Paulo, Av. Bandeirantes, 3.900, Zip Code 14040-901, Ribeirão Preto, SP, Brazil

7

Abstract

8

The turmeric dye extraction residue whole did not form film. Therefore, four turmeric

9

flour fractions (F1, F2, F3, and F4) were produced from the turmeric residue by wet

10

milling and sieving. F2, which presented the lowest lignocellulosic and the highest

11

starch content, afforded the best material for film production. The main objectives of the

12

present study were (i) to evaluate the effect of heating temperature (T) and pH on the

13

mechanical and functional properties of F2 film plasticized with glycerol, and (ii) to

14

optimize the process conditions (T, pH) by response surface method and multi-response

15

analyses. Higher T (>90ºC) and alkaline pH produced a denser and mechanically

16

stronger polymer matrix and more soluble film. In contrast, less soluble films were

17

produced at intermediate T and pH. The effect of T and pH on the mechanical properties

18

and solubility of turmeric films is associated with starch gelatinization and protein

19

solubilization and denaturation. High T produced loss of the curcuminoids and

20

antioxidant activity of turmeric films. The optimal conditions were T of 85.1 ºC and pH

21

of 8.1. These conditions yielded films with high mechanical strength (9 MPa), low

22

solubility (37%), and low water vapor permeability (0.296 g.mm.h-1.m-2.kPa-1).

AC C

EP

TE D

M AN U

SC

RI PT

1

23 *

Corresponding author: Tel: +55 16 36020580, Fax: +55 16 36332660, E-mail address:

[email protected]

1

ACCEPTED MANUSCRIPT 24

Key words: curcumin, turmeric, bioactive film, antioxidant, mechanical properties

25

1. Introduction Edible and/or biodegradable films from biopolymers originating from

27

agricultural sources could substitute non-degradable synthetic plastics in packaging

28

materials. Researchers are currently exploring different materials such as flour extracted

29

from agricultural sources in an attempt to improve the properties of biodegradable films.

30

Because this flour consists of natural blends of starch, protein, and lipid that remain in

31

their original system, it is a new candidate material in the area of biodegradable and

32

edible films (Baldwin et al., 1995; Tapia-Blácido et al., 2007). Researchers have

33

obtained flour from raw materials of plant origin such as fruits, cereals, tubers, and

34

rhizomes to prepare biodegradable films (Andrade-Mahecha et al., 2012; Dias et al.,

35

2010; Gontard, et al., 1993; Pelissari et al., 2013; Tapia-Blácido et al., 2011; Tapia-

36

Blácido et al., 2007). A recent trend has been to employ industrial residues to produce

37

flour from natural mixtures of biopolymers, which may provide the packaging market

38

with a competitive product.

TE D

M AN U

SC

RI PT

26

Curcuma longa L. belongs to the family Zingiberaceae, commonly known as

40

turmeric. This herb is bright yellow because it contains the curcuminoids curcumin,

41

demethoxycurcumin, and bisdemethoxycurcumin. Curcumin is a diphenolic compound

42

that exerts anti-inflammatory action; it can potentially treat cystic fibrosis, Alzheimer's,

43

malarial diseases, and cancer (Joshi et al., 2009; Yallapu et al., 2012). The turmeric

44

resin oil usually contains 30 to 45% of curcuminoids and 15 to 20% of volatile oil.

45

Extraction of this resin oil generates a residue that consists mainly of starch and fibers;

46

this residue may also present residual levels of curcuminoids with antioxidant properties

47

(Braga et al., 2003; Braga et al., 2006; Kuttigounder et al., 2011). In a previous study,

48

our research group demonstrated that the film produced from the turmeric dye

AC C

EP

39

2

ACCEPTED MANUSCRIPT 49

extraction residue and containing sorbitol as plasticizer exhibited antioxidant activity

50

and could constitute an active packaging material. However, these films had poor

51

elongation at break (Maniglia et al., 2014). In this scenario, the present work reports on the development of a bioactive

53

turmeric flour film plasticized with glycerol. The heating temperature and the pH used

54

during turmeric flour film production were optimized. Glycerol was selected because it

55

is a highly hygroscopic molecule, glycerol addition to film-forming solutions prevents

56

film brittleness (Karbowiak et al., 2006), and glycerol is almost always systematically

57

incorporated in most hydrocolloid films (Cuq et al., 1997; Zhang & Han, 2006).

58 59

2. Material and methods

60

2.1. Materials

M AN U

SC

RI PT

52

The dried, milled, and sieved (0.250 mm) turmeric (Curcuma longa L) was

62

supplied by the industry “Flores and Ervas” (Campinas, Brazil). The turmeric residue

63

(R) was obtained from turmeric dye extraction by the Soxhlet method at ∼47 ºC, for 3 h.

64

A mixture of ethanol/isopropanol (1:1 v/v) (Synth-São Paulo, Brazil) was used as

65

solvent during Soxhlet extraction, because this is the mixture employed by the dye

66

industry in Brazil.

EP

Glycerol, used as plasticizer, was purchased from Sigma-Aldrich (São Paulo,

AC C

67

TE D

61

68

Brazil).

69

2.2. Preparation of turmeric flour from the turmeric dye extraction residue

70

Wet milling of R yielded the turmeric flour (Figure 1). First, R was steeped in

71

water at 5 ºC, for 24 h. Next, R was milled and passed through 80-mesh screens four

72

times. Then, the material was sieved through 200- and 270-mesh screens. The solids

73

retained in the 80- (F1), 200- (F2), and 270- (F3) mesh screens were dried at 35 ºC, for

74

24 h, in an oven with forced circulation (MA Q314M, Quimis, Brazil). The liquid 3

ACCEPTED MANUSCRIPT 75

fraction was centrifuged at 6000 × g and 10 ºC, for 20 min, to separate the solids, which

76

were dried at 35 ºC, for 24 h. This material was called F4.

77

2.3. Turmeric flour characterization The turmeric flour moisture, crude protein, and ash contents were analyzed

79

according to standard AOAC methods (AOAC, 1997). The lipids content was calculated

80

by the method of Bligh & Dyer (1959). Cellulose was determined on the basis of the

81

methodology described by Sun (2004). Hemicellulose was determined by HPLC,

82

according to the methodology described by Gouveia et al. (2009). Klason lignin was

83

determined by the TAPPI T 222 om-22 (2002) method, and soluble lignin was analyzed

84

by measuring the absorbance at 280 nm. All the analyses were performed in triplicate.

85

2.4. Film preparation

M AN U

SC

RI PT

78

The turmeric flour films were prepared by casting of the F2 turmeric flour

87

fraction. Initially, a suspension of 5 g of flour/100 g was prepared in deionized water

88

and homogenized for 30 min in a magnetic stirrer (IKA MAG ® C-HS7-Marconi,

89

Brazil). The pH was adjusted, and the solution was heated for 4 h. During this period,

90

the solution was submitted to 2-min homogenization cycles at 12,000 rpm at every

91

hour; an ultra-turrax homogenizer (Ultracleaner 1400, Unique, Brazil) was employed.

92

The pH (6.59, 7.0, 8.0, 9.0, or 9.41) and T (78, 80, 85, 90, or 92 °C) were varied

93

according to a 22 central composite design (star configuration). Next, 22 g of

94

glycerol/100 g of flour was added, and the mixture was heated for 20 min. The solution

95

was poured onto acrylic plates (a weight of 0.15 g.m-² was maintained), and dried at 35

96

°C, for 7 h, in an oven with forced circulation (MA Q314M, Quimis, Brazil). Before

97

characterization, all the films were preconditioned for at least 48 h in desiccators

98

containing saturated NaBr solution (58% RH).

AC C

EP

TE D

86

4

ACCEPTED MANUSCRIPT 99

2.5. Mechanical properties The mechanical tests were performed on a texture analyzer TA TX Plus (TA

101

Instrument, England). The tensile strength (TS) and elongation at break (E) were

102

obtained according to the ASTM D882- 95 method (ASTM, 1995), by taking an

103

average of five determinations in each case. The crosshead speed was 1.0 mm.s-1,

104

respectively. Young’s modulus (YM) was calculated as the inclination of the initial

105

linear portion of the stress versus strain curve, with the aid of the software Texture

106

Expert V.1.22 (SMS).

107

2.6. Solubility in water (S) and moisture content (MC)

SC

RI PT

100

Solubility was calculated as the percentage of dry matter of the solubilized film

109

after immersion in water at 25 ± 2 °C for 24 h (Gontard et al., 1992) as described by

110

Tapia-Blácido et al. (2011). The moisture content in the films was also determined by

111

drying the materials in an oven at 105 °C until constant weight (~24 h). The analyses

112

were performed in triplicate.

113

2.7. Water vapor permeability (WVP)

TE D

M AN U

108

The water vapor permeability (WVP) test was performed at 25 ± 2 ºC, in

115

triplicate; a modified ASTM E96-95 (ASTM, 1995) method was used. Film samples

116

were sealed over the circular opening of a permeation cell containing silica gel. Then,

117

the cells were placed in desiccators containing distilled water. After the samples had

118

reached steady-state conditions (∼20 h), the cell was weighed every one hour, for 9 h,

119

on an analytical scale. The WVP was calculated as described by Maniglia et al. (2014).

120

2.8. Optimal turmeric flour films antioxidant activity

AC C

EP

114

121

The antioxidant activity of the optimal turmeric flour films was determined by

122

DPPH, and the curcuminoids content was quantified by HPLC (Model: CTO-10ASVP,

123

Shimadzu) according to the methodology described by Maniglia et al. (2014). The 5

ACCEPTED MANUSCRIPT remaining DPPH was calculated based on the absorbance at 517 nm (HP Hewlett

125

Packard 8453 spectrophotometer). The control consisted of 500 µL of DPPH solution

126

(0.06 mmol.L-1). The radical scavenging activity (RSA) was determined according to

127

the following equation (Martins et al., 2012):

128

%RSA = 100 × (1 −

A sample A control

)

RI PT

124

(1)

where Asample = absorbance of the solution containing the sample, and Acontrol =

130

absorbance of the DPPH solution without addition of the film.

131

2.9. Scanning electron microscopy (SEM)

SC

129

The turmeric flour fractions and the optimized films obtained from the

133

experimental design were placed in aluminum holders and coated with gold by

134

sputtering (Sputter Coater, model SCD050). A Microscope Scanning Electron mark

135

model ZEISS EVO-50 under accelerating voltage of 20 kV was used to analyze

136

samples.

137

2.10. Experimental design

TE D

M AN U

132

The surface-response methodology was used to study how the heating

139

temperature (T) and the pH of the solution affected TS, E, S, MC, and WVP. To this

140

end, a 22 central composite design (star configuration) was employed. An analysis of variance (ANOVA), a multiple comparison test, and all statistical

AC C

141

EP

138

142

analyses were performed using the Statistica 12 software (StatSoft). The data were fitted

143

to a second order equation (Equation 2) as a function of the independent variables.

144

Yi = b0 + b1X1 + b2X2 + b12X1X2 + b11X12 + b22X22,

145

where bn were constant regression coefficients, Yi were the dependent variables (T, E,

146

S, MC, and WVP), and X1 and X2 were the coded independent variables (T and pH,

147

respectively).

(2)

6

ACCEPTED MANUSCRIPT After the surface-response results were obtained, the process conditions were

149

optimized by multi-response analysis (Derringer & Suich, 1980). This method

150

transformed response variables (Yi) into an individual function of dimensionless

151

desirability (gi) (Equation 4), ranging from 0 (undesirable response) to 1 (desired

152

response). The overall desirability function (G) (Equation 3) was obtained from the

153

geometric means of individual desires. G was later maximized by using the software

154

Mathematic 5.0.

155

G = (g1n1 , g n2 2 ,......g1n K )1 / K ,

156

where:

157

gi =

158

where Ymin was the minimum value of the response, Ymax was the maximum value of

159

the response, k was the number of considered responses, and n was the weight of each

160

response. In the case of solubility, Equation 4 had to be redesigned, to obtain the

161

minimum values for this response (Equation 5).

162

gi =

SC M AN U

TE D

Ymax − Yi Ymax − Ymin

(4)

(5)

Finally, turmeric flour films were prepared in the optimal conditions, to validate the optimized process conditions obtained by the multi-response analysis.

165

3. Results and discussion

166

3.1. Turmeric flour characterization

AC C

164

167

(3)

EP

163

Yi − Ymin , Ymax − Ymin

RI PT

148

Table 1 shows the chemical composition of the turmeric dye extraction residue

168

(R) and of the turmeric flour fractions (F1, F2, F3, and F4). R contained starch, protein,

169

lipid, cellulose, hemicellulose, and soluble and insoluble lignin in its composition;

170

starch was the major component. The methodology shown in Figure 1 yielded four

171

fractions of turmeric flour; each fraction had distinct chemical composition. Fraction F2 7

ACCEPTED MANUSCRIPT presented higher starch content and lower cellulose, hemicellulose, and lignin contents

173

than R, F1, F3, and F4. The different chemical compositions of the R and the flour

174

fractions led to turmeric flour fractions with distinct microstructures (Figure 2). R and

175

F1 exhibited a more irregular, denser, and more closely packed microstructure as

176

compared with F2, F3, and F4 (Figure 2). As observed by Kuttigounder et al. (2011) in

177

cured-dried turmeric powder, heating applied during Soxhlet extraction (~47 ºC) may

178

have caused partial starch gelatinization of R, and posterior starch retrogradation during

179

storage may have modified the structure of R as compared with fresh turmeric rhizomes.

180

Meanwhile, Braga et al. (2006) observed that supercritical extraction of turmeric dye

181

conducted in CO2 in a 1:1 mixture of ethanol/isopropyl alcohol (10% v/v) at 30 ºC and

182

300 bar did not alter the structure of turmeric residue, which allowed for later extraction

183

of starch granules. Therefore, the high temperature used during dye extraction by the

184

Soxhlet method produced a turmeric residue (R) whose closely packed structure

185

retained the starch granules. Indeed, starch granules were not as evident in R as

186

compared with F2, F3, and F4. The structures of R and F1 hindered the access of water

187

into the starch granule, which prevent starch gelatinization during filmogenic solution

188

preparation, consequently these materials did not form films (Figure 3a).

EP

TE D

M AN U

SC

RI PT

172

On the other hand, the micrographs of fractions F2, F3, and F4 clearly showed

190

agglomerated starch granules and some loose starch granules. The starch particles

191

presented irregular surface. The particles were not flat, but elliptical. The starch

192

granules measured between 10 and 30 µm. Other authors have also reported that

193

turmeric starch granules have elliptical shape and sizes ranging between 10 and 35 µm

194

along the major axes (Braga et al., 2006; Kuttigounder et al., 2011).

AC C

189

195

In a preliminary test (results not shown) in which the preparation of films from

196

F1, F2, F3, and F4 was attempted, F1 was the only fraction that did not yield a film. F2

8

ACCEPTED MANUSCRIPT 197

presented the best performance for film formation (Figure 3b). The lower content of

198

lignocellulosic material and the higher content of starch in F2 as compared with F1, F3,

199

and F4 may have contributed to this outcome.

200

3.2. Optimization of the turmeric film Table 2 summarizes the results of the 22 central composite design (star

202

configuration) performed to evaluate how the heating temperature (T) and pH

203

influenced the turmeric flour films properties. Table 3 depicts the analysis of variance

204

(ANOVA) and the regression coefficients of the second order polynomials models for

205

the tensile strength (TS), elongation at break (E), Young’s modulus (YM), solubility

206

(S), moisture content (MC), and water vapor permeability (WVP) of turmeric flour

207

films. The analysis of variance (ANOVA) showed that, except for MC, the fitted

208

models were statistically significant (p < 0.10) for all the studied responses (Fcalculated >

209

Flisted). However, only the models developed for TS, E, YM, and S were predictive,

210

because their F ratio (Fcalculated/Flisted) was higher than the corresponding Flisted (Table 3).

211

Hence, we generated the response surfaces for these properties.

212

3.2.1 Mechanical properties of films

TE D

M AN U

SC

RI PT

201

Figure 4 (a-c) depicts the response surfaces for TS, E, and YM of the turmeric

214

flour films plasticized with glycerol. The response surface for TS revealed that

215

increasing heating temperature (T) improved the mechanical resistance of the turmeric

216

films, with maximum TS values occurring at temperatures > 90 ºC for pH lying between

217

7.5 and 9 (Figure 4a). Therefore, higher T and alkaline pH produced a denser and

218

mechanically stronger polymer matrix. This behavior also emerged in the case of the

219

achira and amaranth flour films (Andrade-Mahecha et al., 2012; Tapia-Blácido et al.,

220

2011).

AC C

EP

213

9

ACCEPTED MANUSCRIPT On the other hand, a maximum region for E appeared for T between 82 and 88

222

ºC and pH between 8 and 9.4 (Figure 4b). This behavior was different for the turmeric

223

flour film plasticized with 30 g of sorbitol/100 g of flour (Maniglia et al., 2014), for

224

which higher temperatures (85 to 92ºC) and lower pH (7.5 to 8.5) furnished larger E

225

values.

RI PT

221

Figure 4c revealed that YM was strongly influenced by T and regardless of pH.

227

The effect of T on the mechanical properties turmeric flour films is associated with

228

starch gelatinization and protein denaturation. Braga et al. (2006) studied the turmeric

229

starch gelatinization. These authors reported that the values of onset temperature (To),

230

peak temperature (Tp), and conclusion temperature (Tc) were 70.8, 81.0, and 97.0 ºC,

231

respectively. Since that turmeric flour films more resistant mechanically were produced

232

at higher T values (>85ºC) corresponding to temperature values above Tp is expected

233

that in this condition most of the starch granules should be gelatinized. Thus, higher

234

heating temperature must have increased the number of interactions of amylose and

235

amylopectin with other components of the polymeric matrix (proteins, lipids, and

236

fibers). The high temperature used to prepare the films should elicit a closely packed

237

state where extensive intermolecular bonding occurs, thereby inhibiting further

238

orientation and better alignment of the protein and starch chains (Arvanitoyannis et al.,

239

1998).

M AN U

TE D

EP

AC C

240

SC

226

On the other hand, higher temperature promotes also protein denaturation, which

241

may also have increased the number of interactions such as protein-protein and starch-

242

protein. Since that the flour turmeric contents fibers, protein-fiber and polyphenol-

243

protein interactions can also be found in the turmeric film matrix. Lignin is a

244

polyphenolic compound located on the fiber surface, plyphenol-protein interactions can

245

be formed in protein/fiber matrix (Montaño-Leyva et al., 2013).

10

ACCEPTED MANUSCRIPT The effect of pH on TS and E is also associated with starch gelatinization.

247

Builders et al. (2014) observed that the tigernut starch gelatinization temperature was

248

remarkably sensitive to pH: the gelatinization temperature was higher in neutral pH as

249

compared with dispersions in pH 9.2 and 4.0. In other words, pH can change the starch

250

gelatinization temperature and consequently modify film properties. When the starch

251

granule becomes less structured, different interactions among the polymers (amylose,

252

amylopectin, proteins, and fibers) present in the turmeric flour arise, leading to a more

253

homogeneous matrix. Figure 4 shows that alkaline pH (7.5 to 9) affords films that are

254

more resistant to rupture, which suggests that these pH conditions require a higher

255

temperature for turmeric flour starch gelatinization. Higher pH may also have

256

solubilized the turmeric proteins, improving their dispersion within the filmogenic

257

matrix.

258

3.2.2. Film solubility

M AN U

SC

RI PT

246

Figure 5 shows that it is possible to produce less soluble turmeric flour film at

260

heating temperatures between 78 and 85 ºC and in pH between 7 and 8.4. The lower

261

solubility of the turmeric film obtained at lower T and intermediate pH values may be

262

associated with incomplete starch gelatinization in the turmeric flour. The turmeric flour

263

film plasticized with sorbitol also presented a region of minimum S as a function of pH

264

(>7.5) and T (80–88 ºC) as compared with the turmeric flour film plasticized with

265

glycerol (Maniglia et al., 2014). Hence, the effect of process conditions on turmeric

266

film solubility may vary depending on the plasticizer.

267

3.3. Experimental validation of optimization

AC C

EP

TE D

259

268

The models calculated for TS, E, and S of the turmeric flour films (Table 3)

269

furnished the desirability function (G). The gi function was obtained by considering the

270

minimum and maximum values of each response variable derived from the

11

ACCEPTED MANUSCRIPT 271

experimental results obtained in the experimental design (Table 2). The optimal process

272

conditions leading to the maximum global desirability of the G function were T = 85.1

273

ºC and pH = 8.1. The optimization methodology was validated for production of turmeric flour

275

films under the optimized process conditions. TS, E, S, and WVP of the optimized

276

turmeric flour films were compared with values predicted by the models (Table 4). The

277

relative deviation values revealed that the predicted and experimental data correlated

278

well.

279

3.4. Characterization of the optimized turmeric flour film

SC

RI PT

274

The turmeric flour films prepared with the optimal formulation and bearing

281

glycerol as plasticizer were less resistant (8.9 MPa) than the turmeric flour film

282

plasticized with sorbitol (17.99 MPa) (Maniglia et al., 2014). Meanwhile, the TS value

283

was close to the TS value reported for banana flour films (9.2MPa) (Pelissari et al.,

284

2013) and higher than the TS value of other flour films like achira, quinoa, and

285

amaranth (Andrade Mahecha et al., 2012, Tapia-Blácido et al., 2011, Araujo-Farro et

286

al., 2010). However, E was much lower (1.97%) than the E values achieved for rice

287

flour and cellulose fiber films (4.0%) (Dias et al., 2011), amaranth flour film of

288

caudatus and cruentus species (83.7 and 51.9%) (Tapia-Blácido et al., 2011), achira

289

flour films (22%) (Andrade-Mahecha et al., 2012), and banana flour films (24.2%)

290

(Pelissari et al., 2013). Regardless of the plasticizer type (glycerol or sorbitol), the

291

turmeric flour films were less elongable than other flour films. The presence of

292

hemicellulose and lignin and incomplete starch gelatinization could make incorporation

293

of glycerol and sorbitol into the polymeric matrix difficult.

AC C

EP

TE D

M AN U

280

294

WVP (0.980 x 10−10gm−1s−1Pa−1) of turmeric flour film plasticized with glycerol

295

was greater than WVP of the film plasticized with sorbitol (0.415 x 10−10g m−1 s−1 Pa−1)

12

ACCEPTED MANUSCRIPT (Maniglia et al., 2014). This result was expected: glycerol can interact with water,

297

which facilitates permeation of this plasticizer through the film (Galdeano et al., 2009).

298

Sorbitol hygroscopicity is low, because it can crystallize at room temperature and

299

displays high relative humidity (Talja et al., 2007). On the other hand, WVP of the

300

turmeric flour film plasticized with glycerol was greater than WVP of amaranth flour

301

films (0.7 x 10−10gm−1s−1Pa−1) (Tapia-Blácido et al., 2011) and lower than WVP of

302

banana flour films (2.1 x 10−10gm−1s−1Pa−1) (Pelissari et al., 2013).

RI PT

296

The turmeric flour film had lower S (37.23%) than achira flour films (Andrade-

304

Mahecha et al., 2012), but larger S than banana (18.7%) and quinoa (18.7%) flour films

305

(Araujo-Farro et al., 2008, Pelissari et al., 2013). Although glycerol is more hydrophilic

306

than sorbitol, S values were similar to the S values achieved for turmeric flour films

307

plasticized with sorbitol. The matrix must have incorporated the fibers present in the

308

turmeric flour more easily, which suggested good degree of fiber-matrix interactions.

309

Such strong interaction between the matrix and the fibers meant that the matrix and

310

water interacted less, leading to lower S. Müller et al. (2009) also found that S

311

decreased in cassava starch films with increasing amount of fibers (cellulose).

TE D

M AN U

SC

303

Figure 6 depict the turmeric flour films microstructure as analyzed by SEM at

313

500x. The turmeric films showed a homogeneous and rough surface. The starch-fiber,

314

protein-fiber and fiber-fiber interactions into film matrix may have afforded the rougher

315

surface causing discontinuity of the film matrix and consequently lower elongation at

316

break of films (Figure 6a). Since that the turmeric film presents a complex composition

317

due to the presence of fiber, protein, starch, lipid and glycerol, the cross-section

318

revealed a dense and irregular structure (Figure 6b). It may have been related to the

319

different interactions such as starch-fiber, starch-protein, starch-starch, protein-protein,

320

protein-fiber that forming the turmeric film matrix. This type of structure was also

AC C

EP

312

13

ACCEPTED MANUSCRIPT 321

observed in amaranth and chia flour films (Colla et al., 2006, Tapia-Blacido et al., 2007,

322

Dick et al., 2015)

323

3.5. Antioxidant activity HPLC analysis of the optimized turmeric flour film evidenced peaks with

325

retention times similar to those that emerged in the chromatogram of turmeric extract

326

(Figure

327

demethoxycurcumin, and bisdemetoxycurcumin) in the turmeric film (Table 5).

328

However, the curcuminoids concentration in the film was lower as compared with the

329

curcuminoids concentration in R and F2, indicating that the film production process

330

caused loss of curcumunoids (Table 5). Because film production was conducted at high

331

heating temperature for 4 h to achieve starch gelatinization, this process caused loss of

332

78% curcumin, 91% demethoxycurcumin, and 60% bisdemethoxycurcumin. The

333

turmeric flour fraction F2 presents lower curcuminoids concentration as compared to

334

turmeric residue (R), indicating that the flour production process by wet milling and

335

sieving caused loss of 32% curcumin, 34% demethoxycurcumin, and 33%

336

bisdemethoxycurcumin.

This

confirmed

the

presence

of

curcuminoids

(curcumin,

EP

TE D

M AN U

SC

7).

RI PT

324

DPPH analysis of the residue and turmeric flour also confirmed loss of

338

antioxidant activity of turmeric flour (Table 5). The optimized films also had lower

339

antioxidant activity than turmeric flour and residue. Curcuminoids loss due to heating

340

during film production should have decreased the antioxidant activity of films.

341

Together, these results demonstrated that shorter heating time should reduce

342

curcuminoids loss.

343

4. Conclusion

344

The temperature used during turmeric dye extraction by the Soxhlet method promoted

345

partial starch gelatinization and subsequent starch retrogradation in the turmeric

AC C

337

14

ACCEPTED MANUSCRIPT rhizomes, which modified the structure of the turmeric dye extraction residue. This

347

prevented film production from turmeric residue and required preparation of turmeric

348

flour to obtain the desired film. The turmeric flour originating from the turmeric dye

349

extraction residue contained starch, cellulose, lignin, hemicellulose, proteins, lipids, and

350

remnants of curcuminoids, being potentially applicable in the preparation of active

351

biodegradable films. The presence of these components influenced the film structure

352

and afforded less elongable films as compared with other flour films. The heating

353

temperature and pH affected film properties, solubility, and water vapor permeability.

354

The response surface methodology and multi-response analyses allowed optimization of

355

process conditions, to produce turmeric flour films with better mechanical properties

356

and lower solubility. These conditions were T = 85.1 °C and pH = 8.1.

357

Acknowledgement

358

The authors would like to thank Fundação de Amparo à Pesquisa do Estado de São

359

Paulo (São Paulo Research Support Foundation–FAPESP) and CAPES (Coordenacão

360

de Aperfeiçoamento de Pessoal de Nível Superior) for financial support.

361

References

362

Andrade-Mahecha,

363

Development and optimization of biodegradable films based on achira flour.

364

Carbohydrate Polymers, 88(2), 449-458.

365

AOAC. (1997). Official Methods of Analysis of AOAC International. (16th ed).

SC

M AN U

TE D

EP M.M.,

Tapia-Blácido,

&

D.R.,

Menegalli,

F.C.

(2012).

AC C

366

RI PT

346

Association of Official Analytical Chemists, Washington.

367

Araujo-Farro, P.C., Podadera G., Sobral; P.J.A., & C. Menegalli, F. C. (2010).

368

Development of films based on quinoa (Chenopodium quinoa, Willdenow) starch.

369

Carbohydrate Polymers, 81, 839–848.

15

ACCEPTED MANUSCRIPT 370

Arvanitoyannis, I., Nakayama, A., & Aiba, S. (1998). Edible films made from

371

hydroxypropyl starch and gelatin and plasticized by polyols and water. Carbohydrate

372

Polymers, 36(2/3), 105–119. ASTM. (1995). Standard test method for tensile properties of thin plastic sheeting

374

(D882-95) and Standard method for water vapor transmission of materials (E 96-95).

375

In: Annual book of ASTM standards. Philadelphia: American Society for Testing and

376

Materials.

378

Baldwin, E.A.; Nisperos – Carriedo, M.O.; & Baker, R.A. (1995). “Edible Coatings for

SC

377

RI PT

373

Lightly Processed Fuits and Vegetables”, HortScience, 30, 35 – 38.

Bitencourt, C.M., Fávaro-Trindade, C.S., Sobral, P.J.A., & Carvalho, R.A. (2014).

380

Gelatin-based films additivated with curcuma ethanol extract: Antioxidant activity

381

and physical properties of films. Food Hydrocolloids, 40, 145-152.

383

Bligh, E.G., & Dyer, W.J. (1959). A rapid method of total lipid extraction and purification. Canadian Journal of Biochemistry and Physiology, 37, 911–917.

TE D

382

M AN U

379

Braga, M.E., Leal, P.F., Carvalho, J.E., & Meireles, M.A. (2003). Comparison of yield,

385

composition, and antioxidant activity of turmeric (Curcuma longa L.) extracts

386

obtained using various techniques. Journal of Agricultural and Food Chemistry, 51,

387

6604–6611.

EP

384

Braga, M.E.M., Moreschi, S.R.M., & Meireles, M.A.A. (2006). Effects of supercritical

389

fluid extraction on Curcuma longa L. and Zingiberofficinale R. starches.

390

Carbohydrate Polymers, 63, 340-346.

AC C

388

391

Builders, P.F., Mbah, C.C, Adama, K.K., & Audu, M.M. (2014). Effect of pH on the

392

physicochemical and binder properties of tigernut starch. Starch/Stärke. 66, 281-293.

393

Colla, E., Sobral, P. J. A., & Menegalli, F. C. (2006). Amaranthus cruentus flour edible

394

films: influence of stearic acid addition, plasticizer concentration, and emulsion stirring

16

ACCEPTED MANUSCRIPT speed on water vapor permeability and mechanical properties. Journal of Agriculture

396

and Food Chemistry, 54, 6645-6653.

397

Cuq, B., Gontard, N., Cuq, J. L., et al. (1997). Selected functional properties of fish

398

myofibrillar protein-based films as affected by hydrophilic plasticizers. Journal of

399

Agricultural and Food Chemistry, 45(3), 622-626.

401 402 403

Derringer, G. & Suich, R. (1980). Simultaneous optimization of several response variable. Journal of Quality Technology, 12(4), 214-219.

Dias, A.B., Muller, C.M.O., Larotonda, F.D.S., & Laurindo, J. (2010). Biodegradable

SC

400

RI PT

395

films based on rice starch and rice flour. Journal of Cereal Science, 51, 213- 219. Dias, A.B., Muller, C.M.O., Larotonda, F.D.S., & Laurindo, J. (2011). Mechanical and

405

barrier properties of composite films based on rice flour and cellulose fibers. LWT-

406

Food Science Technology, 44, 535-542.

408

Dick et al. (2015). Edible films based on chia flour: Development and characterization. Journal of Applied Polymer Science, 132, 42455-42464.

TE D

407

M AN U

404

Galdeano, M.C., Mali, S., Grossmann, M.V.E., Yamashita, F., & García, M.A. (2009).

410

Effects of plasticizers on the properties of oat starch films. Materials Science and

411

Engineering: C – Journal, 29, 532-538.

EP

409

Gontard, N., Guilbert, S., & Cuq, J.L. (1992). Edible wheat gluten films: Influence of

413

the main process variables on film properties using response surface methodology.

414

Journal of Food Science, 57(1), 190–195.

AC C

412

415

Gontard, N., Guilbert, S., & Cuq, J. L. (1993). Water and glycerol as plasticizers affect

416

mechanical and water vapor barrier properties of an edible wheat gluten film. Journal

417

of Food Science, 58(1), 206–211.

17

ACCEPTED MANUSCRIPT 418

Gouveia, E.R., Do Nascimento, R.T., & Souto-Maior, A.M. (2009). Validação de

419

metodologia para a caracterização química de bagaço de cana-de-açúcar. Química

420

Nova, 32(6), 1500-1503.

422

Joshi, P., Jain, S., & Sharma, V. (2009). Turmeric a natural source of edible yellow colour. Int. J. Food Sci. Tech. 4, 2402-2406.

RI PT

421

Karbowiak, T., Hervet, H., Leger, L., et al. (2006). Effect of plasticizers (water and

424

glycerol) on the diffusion of a small molecule in iota-carrageenan biopolymer films

425

for edible coating application. Biomacromolecules, 7(6), 2011-20119.

SC

423

Kuttigounder, D., Rao L., & Bhattacharya, S. (2011). Turmeric Powder and Starch:

427

Selected Physical, Physicochemical, and Microstructural Properties. Journal of Food

428

Science, 76, C1284-C1291.

M AN U

426

Maniglia, B.C., Domingos, J.R; de Paula, R.L.; & Tapia-Blácido, D.R. (2014).

430

Development of bioactive edible film from turmeric dye solvent extraction residue.

431

LWT- Food Science and Technology, 56, 269-277.

TE D

429

Martins, J.T., Cerqueira, M.A., & Vicente, A.A. (2012). Influence of -tocopherol on

433

physicochemical properties of chitosan – based films. Food Hydrocolloids, 27, 220 –

434

227.

EP

432

Montaño-Leyva, B., da Silva, G.G., Gastaldi, E., Torres-Chávez, P., Gontard, N., &

436

Angellier-Coussya, H. (2013). Biocomposites from wheat proteins and fibers:

437

Structure/mechanical properties relationships. Industrial Crops and Products, 43,

438

545-555.

AC C

435

439

Müller, C.M.O., Laurindo, J.B., & Yamashita, F. (2009). Effect of cellulose fibers

440

addition on the mechanical properties and water vapor barrier of starch-based films.

441

Food Hydrocolloids, 23 (5), 1328–1333.

18

ACCEPTED MANUSCRIPT 442

Pelissari, F.M., Andrade-Mahecha, M.M., & Sobral, P.J.A. (2013). Optimization of

443

process conditions for the production of films based on the flour from plantain

444

bananas (Musa paradisíaca). LWT - Food Science Technoogy, 52, 1-11.

446

Sun, J. (2004). Isolation and characterization of cellulose from sugarcane bacasse. Polymer Degradation and Stability, 84, 2, 331-339.

RI PT

445

Talja, R.A., Helén, H., Roos, Y.H., & Jouppila, K. (2007). Effect of various polyols and

448

polyol contents on physical properties of potato starch-based films. Carbohydrate

449

Polymers, 67(3), 288-295.

SC

447

Tapia-Blácido, D., Mauri, A.N., Menegalli, F.C., Sobral, P.J.A., & Añón, M.C. (2007).

451

Contribution of the Starch, Protein, and Lipid Fractions to the Physical,Thermal, and

452

Structural Properties of Amaranth (Amaranthus caudatus) Flour Films. Journal of

453

Food Science 72, E293-E300.

M AN U

450

Tapia-Blácido, D., Sobral, P.J.A., & Menegalli, F.C. (2011). Optimization of amaranth

455

flour films plasticized with glycerol and sorbitol by multi-response analysis. LWT-

456

Food Science Technoogy, 44(8), 1731-1738.

TE D

454

TAPPI. T222 om-02. (2002). Acid - insolub lignin in wood and pulp. 5p.

458

Yallapu, M.M., Jaggi, M., & Chauhan, S.C. (2012). Curcumin nanoformulations: A

460 461

future nanomedicine for cancer. Drug Discovery Today, 17, 71-80. Zhang, Y., & Han, J. H. (2006). Mechanical and thermal characteristics of pea. Journal

AC C

459

EP

457

of Food of Science, 71(2), 109-118.

19

ACCEPTED MANUSCRIPT Figure captions Figure 1. Procedure employed to produce turmeric flour fractions. Figure 2. Microstructure of the turmeric residue (R) and the turmeric flour fractions (F1, F2, F3, and F4). The arrows show starch granules.

RI PT

Figure 3. Pictures of turmeric films produced with (a) turmeric residue (R) and (b) turmeric flour (F2).

Figure 4. Tensile strength (a), elongation at break (b), and Young’s modulus (c) of

SC

turmeric flour films plasticized with glycerol as a function of heating temperature and pH.

M AN U

Figure 5. Solubility in water of turmeric flour films plasticized with glycerol as a function of heating temperature and pH.

Figure 6. SEM micrographs (magnification 500x) of turmeric flour films produced in the optimized process conditions. (a) Cross section, and (b) Surface.

TE D

Figure 7. HPLC profile for (a) the extract from turmeric pie and (b) turmeric flour film prepared in the optimized conditions. Peaks relative to bisdemethoxycurcumin (1),

AC C

EP

demethoxycurcumin (2), and curcumin (3).

20

ACCEPTED MANUSCRIPT

Table 1

Lipid

Ashes

5.44±0.09d

9.33±0.01b

1.35±0.03b

F1

7.52±0.28a

8.56±0.11d

0.68±0.02e

2.54±0.09f

F2

7.73±0.42a

9.25±0.11b

0.88±0.02d

5.80±0.26d

F3

6.93±0.35b

9.14±0.09c

0.72±0.02e

5.40±0.06e

F4

6.48±0.25c

10.60±0.02a

4.10±0.10a 11.97±0.09a

Cellulose

EP

Soluble

Non-soluble

lignina

lignina

5.37±0.29d

2.18±0.04a

9.14±0.59c

48.16±0,15d 10.93±0.34a

15.14±0.19a

2.23±0.08a

11.76±0.50a

69.02±0,29a

2.61±0.24e

3.87±0.95e

1.83±0.05c

6.75±0.57e

60.76±0,41b

3.43±0.36d

10.28±1.35b

1.97±0.07b

8.30±0.38d

47.38±0,15e

7.93±0.32b

6.92±0.47c

1.99±0.03b

9.10±0.30c

Values with different letters in the same column are significantly different (P<0.05). * Moisture content expressed in wet basis (g/100 g w.t) T, dye extraction residue, F1 (80 mesh); F2 (200 mesh); F3 (270 mesh); F5 (centrifuged fraction).

AC C

Hemicellulose

8.75±0.05c

11.51±0.14b 52.37±0,18c

TE D

R

Starch

SC

Protein

M AN U

Moisture*

RI PT

Chemical composition (g/100 g db) of the different turmeric flour fractions

ACCEPTED MANUSCRIPT

Table 2

7 (-1)

2

80 (-1)

3

765.9±15.2

875.0±15.7

34.7 ±0.5

0.283 ±0.06

15.2 ±1.4

1027.9 ±84.1

37.7 ±0.9

0.261 ±0.08

16.5 ±0.7

1.4±0.2

814.2 ±15.0

42.1 ±0.5

0.360 ±0.02

16.4 ±1.1

1.1±0.1

849.5±13.6

35.1 ±0.3

0.414 ±0.01

19.4 ±1.5

YM (MPa)

6.3±0.6

1.2 ±0.1

9 (+1)

9.2±0.3

1.5±0.3

90 (+1)

7 (-1)

10.1±0.5

1.2±0.3

4

90 (+1)

9 (+1)

9.5±1.9

5

78 (-1.414)

8 (0)

7.8±0.2

6

92(+1.414)

8 (0)

7

85 (0)

6.59(-1.414)

8

85 (0)

9.41(+1.414)

9

85 (0)

8 (0)

10

85 (0)

8 (0)

11

85 (0)

8 (0)

S (%)

SC

80 (-1)

MC (g/100g) 15.6 ±0.6

E (%)

M AN U

1

WVP (g.mm.h-1.m-2.kPa-1) 32.3±0.3 0.483 ±0.07

TSc (MPa)

TE D

pH (X2b)

12.7±2.3

1.1±0.5

1105.9 ±65.0

41.6 ±1.5

0.450 ±0.08

14.6 ±1.1

3.6±0.4

1.2±0.5

388.7±68.5

38.3 ±0.6

0.335 ±0.05

16.6 ±1.6

4.2±0.5

3.2±0.1

246.8 ±46.8

41.8 ±1.9

0.236 ±0.03

16.6 ±1.2

8.8±0.5

2.3±0.1

539.7 ±68.1

33.5 ±1.6

0.300 ±0.12

16.9 ±2.0

8.9±1.0

2.0±0.2

643.7 ±65.5

32.7 ±0.6

0.300 ±0.04

16.5 ±1.2

8.3±1.9

2.1±1.0

566.5±23.4

32.5 ±1.0

0.303 ±0.11

16.0 ±1.1

EP

Ta (X1b)

AC C

Test

RI PT

Mechanical properties, solubility in water, moisture content, and water vapor permeability (WVP) of turmeric flour film plasticized with glycerol according to the full-factorial design 22 central composite designs.

a T (ºC). b Independent variables values (the values between brackets are the coded variables). c Tensile strength (TS), elongation at break (E), Young’s modulus (YM), solubility (S), water vapor permeability (WVP), moisture content (M), opacity (OP).

ACCEPTED MANUSCRIPT Table 3 Regression coefficients and analysis of variance (ANOVA) for responses variables

Coefficients

Moisture WVP (g.mm.h content 1 -2 -1 .m .kPa ) (g/100 g) Y5 Y6

RI PT

Tensile Elongation Young's Solubility strength at break modulus in water (MPa) (%) (MPa) (%) Y1 Y2 Y3 Y4 9.88*

2.09*

549.92*

32.90*

β1

1.38*

-0.012

70.48

2.75*

β2 quadratic

0.39

0.42*

-38.16

1.47*

β11

1.15*

-0.58*

287.46*

2.10*

0.063*

0.15

β22 interactions

-2.36*

-0.047

-60.35

2.95*

-0.010*

-0.05

β12 R2

-0.88

-0.025

-80.70

0.50

0.074*

0.08

0.72

0.80

0.72

0.90

0.92

0.24

11.74

14.65

22.72

12.67

10.88

2.10

Flistedb

3.07

3.11

3.36

3.18

3.45

3.36

Fcalculated/Flisted

3.82

4.71

6.76

3.98

3.15

0.62

AC C

EP

M AN U

TE D

Fcalculated

a

0.30*

16.27*

-0.012*

-0.58*

-0.030*

-0.06

SC

βo linear

ACCEPTED MANUSCRIPT

Predicted and experimental values of the responses in the optimum conditions.

Tensile strengh(MPa) Elongation at break (%) -1

Experimental

RDa (%)

9.87

8.88 ± 1.66

-11.15

1.97 ± 0.33

-8.63

0.352 ± 0.05

15.91

37.23 ± 1.95

10.82

2.14 -1

0.296

Solubility (g/100 g)

33.20

M AN U

WVP (g.mm.h .m KPa )

EP

TE D

Relative deviation (RD): [(experim ental value-predicted value)/experimental value]x100.

AC C

a

-2.

Predicted

SC

Properties

RI PT

Table 4

ACCEPTED MANUSCRIPT Table 5 Curcuminoids concentration and antioxidant activity (AA) for turmeric residue, flour (F2), and optimized film. Curcumin (mg/L)

Demethoxycurcumin Bisdemethoxycurcumin (mg/L) (mg/L)

Turmeric residue

15.02 ±1.11

25.50 ± 2.47

Flour

10.26 ± 0.10

16.75 ± 0.31

Optimized film

2.18 ± 0.02

1.41 ± 0.00 -3

-1

AA (%)

RI PT

Samples

22.19 ± 0.31

74.4 ± 0.42

14.86 ± 0.91

66.6 ± 1.28

5.81 ± 0.11

46.0 ± 1.34

AC C

EP

TE D

M AN U

SC

Calibration curve: Curcumin: 17.19 – 0.54 (10 g.L ), R² = 0.992; Bisdemethoxycurcumin: 35.00 – 0.27 . (10 -3 g/L-1), R² = 0.999; Demethoxycurcumin: 48.40 – 0.19 (10 -3 g. L-1), R² = 0.999

ACCEPTED MANUSCRIPT Turmeric residue

Steeping time

H2O, 24 h, 5ºC

Milling

Sieving 200 mesh screen

M AN U

Sieving 270 mesh screen

F1

Drying

F2

Drying

F3

Drying

SC

Sieving 80 mesh screen

RI PT

Recycle four times

Liquid fraction

TE D

Centrifugation

F4

AC C

EP

Drying

Figure 1. Procedure employed for production of turmeric flour fractions

ACCEPTED MANUSCRIPT

RI PT

R

M AN U

SC

F1

AC C

EP

F3

TE D

F2

F4

Figure 2. Microstructure of turmeric residue (R) and turmeric flour fraction (F1, F2, F3 and F4). The arrows show starch granules.

ACCEPTED MANUSCRIPT

SC

RI PT

(a)

TE D

M AN U

(b)

AC C

EP

Figure 3. Pictures of turmeric films produced with (a) turmeric residue, and (b) turmeric flour (F2).

SC

RI PT

ACCEPTED MANUSCRIPT

(b)

AC C

EP

TE D

M AN U

(a)

(c)

Figure 4. Tensile strength (a), elongation at break (b), and Young’s modulus (c) of turmeric flour films plasticized with glycerol as a function of heating temperature and pH.

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

Figure 5. Solubility in water of turmeric flour films plasticized with glycerol as a function of heating temperature and pH.

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

(a)

TE D

(b)

Figure 6. SEM micrographs (magnification 500x) of turmeric flour films produced in

AC C

EP

the optimized process conditions. (a) Surface and (b) Cross section.

(b)

AC C

EP

TE D

M AN U

(a)

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 7. HPLC profile for (a) the extract from the turmeric pie, and (b) turmeric flour film prepared in the optimized conditions. Peaks relative to bisdemethoxycurcumin (1), demethoxycurcumin (2), and curcumin (3).

ACCEPTED MANUSCRIPT

Highlight

Turmeric dye extraction by Soxhlet modified the starch granules

•

Glycerol has little plasticizing effect on the turmeric flour film

•

The heating temperature and pH affect the properties of turmeric flour film

•

The optimal process conditions to obtain turmeric flour film have been established

•

The films contain curcuminoids and display effective antioxidant activity

AC C

EP

TE D

M AN U

SC

RI PT

•