Effect of vacuum spray drying on the physicochemical properties, water sorption and glass transition phenomenon of orange juice powder

Accepted Manuscript Effect of Vacuum Spray Drying on the Physicochemical Properties, Water Sorption and Glass Transition Phenomenon of Orange Juice Powder M. Zohurul Islam, Yutaka Kitamura, Yoshitsugu Yamano, Mai Kitamura PII:

S0260-8774(15)00377-5

DOI:

10.1016/j.jfoodeng.2015.08.024

Reference:

JFOE 8301

To appear in:

Journal of Food Engineering

Received Date: 17 May 2015 Revised Date:

7 August 2015

Accepted Date: 11 August 2015

Please cite this article as: Zohurul Islam, M., Kitamura, Y., Yamano, Y., Kitamura, M., Effect of Vacuum Spray Drying on the Physicochemical Properties, Water Sorption and Glass Transition Phenomenon of Orange Juice Powder, Journal of Food Engineering (2015), doi: 10.1016/j.jfoodeng.2015.08.024. 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 1

Effect of Vacuum Spray Drying on the Physicochemical Properties, Water Sorption and

2

Glass Transition Phenomenon of Orange Juice Powder

3

M. Zohurul Islama d, Yutaka Kitamurab* Yoshitsugu Yamanoc, Mai Kitamuraa Graduate School of life and Environmental Sciences, University of Tsukuba, Japan 3058572

5 b

6

Faculty of life and Environmental Sciences, University of Tsukuba, Japan 305-8572

7

9

d

Tanabe Engineering Co. Ltd, Himeji Technology Center, Japan 671-1123

M AN U

c

8

RI PT

a

SC

4

Department of Food Engineering and Technology, Shahjalal University of Science and Technology, Sylhet 3114, Bangladesh

11

*Corresponding Author. Tel.: +81 298 53 4655; Fax: +81 298 53 4655

12

E-mail: [email protected]

15

16

EP

14

AC C

13

TE D

10

17

18

19

1

ACCEPTED MANUSCRIPT Abstract

21

Vacuum spray drying is a new technique used to produce concentrated orange juice powder

22

using maltodextrin as a drying agent. The dryer was developed for the low-temperature (40-

23

50°C) drying powderization of liquefied food using superheated steam (200°C) as a heating

24

medium. The physicochemical properties of orange juice powder with four different

25

combinations of juice solids were determined: maltodextrin solids at 60:40, 50:50, 40:60, and

26

30:70. The moisture content, hygroscopicity, water activity, particle size, particle

27

morphology, color characteristics, rehydration and ascorbic acid retention were significantly

28

affected by the maltodextrin concentration and drying conditions. Water sorption and the

29

glass transition temperature of orange juice powder conditioned at various water activities

30

were also evaluated. The data obtained were well fitted to both BET and GAB models. A

31

strong plasticizing effect of water on Tg was predicted by the Gordon-Taylor model, where

32

Tg was greatly reduced by the increasing moisture content of the powder.

33

Keywords: orange juice, vacuum spray drying, physicochemical properties, sorption

34

isotherm, glass transition

EP

35

TE D

M AN U

SC

RI PT

20

1. Introduction

37

The most widely grown citrus fruit in Japan is orange (Citrus sinensis). Orange are the great

38

sources of phytochemicals, vitamins C and bioactive compounds. However, it is a highly

39

perishable fruit in shorter shelf life, really difficult to ensure the availability of fresh fruits in

40

the whole year. Various techniques have been used to process and preserve orange; among

41

them, drying has been commonly used as an ancient technique that removes moisture and

42

reduces water activity. Spray drying is widely used to transform liquids into shelf-stable

43

products. Spray dried powder have many benefits and economic potential over the liquid

AC C

36

2

ACCEPTED MANUSCRIPT counterparts. Fruit juice powder can easily handle and transfer due to its reduced weight or

45

volume as well as a reduced packaging system and have a much longer shelf life (Shrestha et

46

al., 2007). Spray drying parameters affect the quality of spray dried powder and drying

47

condition they were the best way to describe the quality change factors of orange products

48

(Brennan et al., 1971). However, the major drawbacks and complex in spray drying of orange

49

juices is the matter of stickiness and flow problems of the powder (Tonon et al., 2008; Goula

50

and Adamopoulos, 2010). Fructose, glucose and citric acid are the main components of

51

orange juice with very low Tg values (5, 31 and 16°C, respectively) in a pure, dry state,

52

which decrease drastically when moisture is absorbed. During spray drying, fruit juice forms

53

an amorphous solid or a syrupy/sticky powder; sticky products are typically produced

54

because the low glass transition temperature (Tg), high viscosity, low melting point and high

55

hygroscopicity (Adhikari et al., 2003; Bhandari and Howes, 1997) lead the juice components

56

to adhere to the drying chamber and agglomerate on the conveying system (Bhandari and

57

Howes, 2005). Glass transition temperature (Tg) can use as a reference parameter to

58

characterize the properties, quality and stability of food (Levine and Slade, 1990). Another

59

thermodynamic tool, the water sorption isotherm, is used to predict the interaction between

60

the water and food components. Glass transition and water activity have been extensively

61

used to predict the storage stability of dried products.

62

To overcome the stickiness problem, various methods that are able to produce free-flowing

63

fruit juice powder have been suggested: using an adjunct or a carrier agent (maltodextrin,

64

gum, starch or gelatin) as an additive in the feed material during spray drying (Saénz et al.,

65

2009), scrapping the drying surfaces, spray freezing or cooling the drying chamber wall

66

(Chegini and Ghobadian, 2005; Chegini et al., 2008). Carrier agents with higher molecular

67

weights increase the glass transition temperature and reduce the hygroscopicity and stickiness

AC C

EP

TE D

M AN U

SC

RI PT

44

3

ACCEPTED MANUSCRIPT of the powder, which may in turn increase the final product yield and efficiency (Lee et al.,

69

2013). Goula and Adamopoulos (2010) produced spray dried concentrated orange juice

70

powder by using dehumidified air as drying medium and maltodextrin DE 6, DE 12, DE 21

71

as carrier agent. Shrestha et al. (2007) conducted spray drying of orange juice with various

72

levels of maltodextrin solids as carrier agent. Tsourouflis et al. (1976) stated that when they

73

were used as drying aids for orange juice, low-dextrose equivalent maltodextrins were found

74

to give higher collapse temperatures than high-DE maltodextrins at the same

75

concentrations.The major limitations of the use of drying aids are causing the subsequent

76

change in the product properties and the cost. For this reason, maltodextrin is commonly used

77

as a carrier agent because it is relatively inexpensive and highly soluble in water with low

78

viscosity and a bland flavor (Carolina et al., 2007 and Saénz et al., 2009); it is also more

79

efficient at protecting bioactive compounds from adverse conditions (Ferrari et al., 2012).

80

Another drawback of spray drying at high temperatures (150-250°C) is that heat sensitive

81

functional ingredients are nearly destroyed.

82

In the present study, a vacuum spray drying (VSD) method is proposed as a new technique

83

for vacuum pressure spray drying of concentrated orange juice powder. The dryer was

84

developed to turn liquefied food into a powder at a low temperature drying using superheated

85

steam as the heat source. The evaporation temperature is a saturation temperature based on

86

the degree of vacuum; therefore, because the heat exchange occurred under vacuum, the

87

drying product temperature could be maintained at a low temperature (40-60°C), which will

88

suppress the damage or loss of thermosensitive functional ingredients. Ultrasonic vacuum

89

spray drying has been used to produce dried probiotic powder (Semyonov et al., 2011).

90

Kitamura et al. (2009) and Kitamura and Yanase (2011) who were developed and conducted

91

vacuum spray drying of probiotic foods at 40-60oC. However, no data exists regarding this

AC C

EP

TE D

M AN U

SC

RI PT

68

4

ACCEPTED MANUSCRIPT 92

vacuum spray drying technique at a low temperature treatment for orange juice powder

93

without freezing.

94

Therefore, the objectives of this study were to produce orange juice powder by VSD with

96

different combinations of maltodextrin; to determine the effect of vacuum spray drying on the

97

physicochemical and functional properties of orange juice powder and to provide

98

experimental data on water sorption isotherms and glass transition temperatures of the

99

concentrated orange juice powder with a carrier agent for insight on powder stability. In order

100

to predict the sorption characteristics by widely used BET and GAB models, and the Gordon-

101

Taylor model used to determine the glass transition temperature and water plasticizing effect

102

on the concentrated orange juice powder.

103

M AN U

SC

RI PT

95

2. Materials and Methods

105

2.1. Raw materials

106

Concentrated orange juice (Citrus sinensis) was obtained from Ehime beverage Ltd., Japan.

107

The basic composition of the concentrated orange juice was total soluble solid (TSS)

108

62±0.45%, citric acid 4.8-5.7%, and ascorbic acid 320 mg/100g. Maltodextrin (MD) 12DE

109

(Showa Sangyo Co. Ltd., Japan) with a moisture content of 4.15±0.02% was used as a carrier

110

agent. Considering the TSS contents of concentrated orange juice (COJ) and maltodextrin

111

solids at 60:40, 50:50, 40:60, and 30:70 (concentrated orange juice: maltodextrin) by weight

112

were chosen for vacuum spray drying. A total of 1000 g of COJ and MD were prepared with

113

33 Brix% solutions for VSD. Moreover, in a previous work, it was found that the proportions

114

of orange juice : maltodextrin by weight at 40:60, 35:65, 30:70 and 25:75 were used during

115

spray drying at 160oC (Shrestha et al., 2007). And Goula and Adamopoulos (2010)

116

conducted spray drying of concentrated orange juice with matodextrin as carrier agent at a

AC C

EP

TE D

104

5

ACCEPTED MANUSCRIPT 117

solid ratio of 4, 2, 1, and 0.25. The present study considered these conditions and also other

118

higher concentration of orange juice:maltodextrin 60:40, 50:50 by weight for vacuum spray

119

drying.

120

2.2. Vacuum Spray Drying (VSD) System

122

The experimental VSD (Tanabe Engineering Corporation, Japan) diagram with its working

123

principles is shown in Fig 1. The COJ/MD mixture is sprayed at 300 mL/h through a

124

metering pump with an upward flow and atomized into the drying chamber by two fluid

125

nozzles. The nozzles use compressed air at a rate of 40 N-L/min to atomize the mixture into

126

droplets with small diameters of 10-50 µm. Heat exchange occurs as the atomized droplets

127

come into contact with steam from another steam nozzle that has been superheated to 200°C

128

by an electric heater. The latent heat required to evaporate water in the droplet is supplied by

129

the sensible heat of the superheated steam. Therefore, superheated steam, whose thermal

130

capacity is higher than that of dry air, can supply more heat capacity per unit area. However,

131

an average droplet is not completely dried instantly, so the product temperature does not

132

reach the saturated steam temperature, which is 40°C for a vacuum of approximately 5 kPa in

133

the evaporator. In addition, the main body of the drying chamber is decompressed by a

134

vacuum pump, and the evaporator jacket temperature is maintained at 50°C by a supply of

135

hot water-1 and a return hot water-1at the bottom of the evaporator and the dry powder is

136

drained from the drying chamber by a flow of saturated steam. The dry powder it is

137

completely captured by the first or second cyclone. Usually most of the powder was captured

138

in the first cyclone. To prevent the formation of wall deposits and classification in the

139

cyclone, a hot water-2 and a return hot water-2 at the end of the first cyclone were

140

sipplied.The low-pressure dry air was used to collect the powder in a receiver at 45 oC and

141

here also hot water-3 supplied in the bottom to prevent wall deposition in the receiver and a

AC C

EP

TE D

M AN U

SC

RI PT

121

6

ACCEPTED MANUSCRIPT returned hot water-3 was collected at top. The similar procedure was maintained at the

143

second cyclone. However, superheated steam can greatly reduce the capacity by cooling with

144

a condenser by supplying cool water; the steam can then be collected as water in a water

145

reservoir, and the vacuum pump can be downsized to more easily maintain the pressure.

146

Fig 1. Schematic flow diagram of the vacuum spray drying process with superheated

147

steam

148

2.3 Assessment of the Physicochemical Properties of Orange Juice Powder

149

2.3.1. Moisture content

150

The moisture contents of the orange juice powder were determined by drying in an oven at

151

70°C until consecutive constant weights were obtained, followed by 2 h intervals and which

152

gave variation less than 0.3%. Moisture content expressed as % of moisture in wet basis

153

(Goula and Adamopoulos, 2010; and AOAC, 1990).

154

2.3.2. Bulk density

155

The bulk density of the orange powder was measured by gently adding 2 g of powder to an

156

empty 10 mL graduated cylinder and holding the cylinder in a vibrator for one minute (Goula

157

et al., 2004). The volume was then recorded and used to calculate the bulk density in g/mL.

158

2.3.3. Rehydration

159

Rehydration of orange powder was determined according to the method described by Goula

160

and Adamopoulos (2010) with slight modifications. For measurement 2 g of powder were

161

added in a 50 mL distilled water at 26oC in a 100 mL low form glass beaker. The mixture was

162

agitated in a digital hot plate/stirrer at 900 rpm, using a magnetic stirrer bar with a size of 2

163

mm x 7 mm. Time required in sec for the powder to be completely rehydrated was recorded.

164

2.3.4. Hygroscopicity

165

For hygroscopicity, 1.5g of the orange juice powder was kept in Conway unit (Model:

166

060310-02A, Shibata Co. Ltd., Tokyo, Japan) containing saturated salt solution of NaCl

AC C

EP

TE D

M AN U

SC

RI PT

142

7

ACCEPTED MANUSCRIPT (75.29% RH). Samples was weighed after one week and hygroscopicity was expressed as the

168

weight of the hygroscopic moisture per 100 g of dry solid (Cai and Corke, 2000).

169

2.3.5. Water activity

170

The water activity of the orange powder was determined with the standard water activity of

171

the saturated salt solution such as Lithium Chloride (PubChem CID: 433294), Potassium

172

acetate (PubChem CID:517044), Magnesium Chloride (PubChem CID: 5360315),

173

Magnesium nitrate hexahydrate (PubChem CID: 202877), Strontium chloride hexahydrate

174

(PubChem CID: 6101868), Sodium chloride (PubChem CID: 5234), and Potassium Chloride

175

(PubChem CID: 4873) of 0.11 0.22, 0.33, 0.53, 0.71, 0.75 and 0.84water activities

176

respectively at 25°C (A. ArabHosseini, 2005; Greenspan, 1977; Young, 1967) by Conway

177

unit (Model: 060310-02A, Shibata Co. Ltd., Tokyo, Japan) and the graph intersection method

178

described by Kitamura and Yanase (2011). One gram of powder was kept in aluminium case

179

with aluminium foil and weighed. Then 3 g of each salt placed into the conway unit. After

180

that a few drops of water were added in the salt and the sample containing aluminium case

181

kept inside of the conway unit and conditioned with the saturated salt solution. Finally the lid

182

of Conway unit was covered and kept in 25oC for 2h conditioned with the saturated salt

183

solution. After 2 h, the weight of the sample containing aluminium case was measured. Then

184

the differences of sample weight (Y-axis) were plotted against the known water activity of

185

the saturated salt (X-axis). A straight line was drawn against each data set and

186

intersection with the X-axis, which corresponds to zero sample weight variation, and

187

which was defined as the water activity of the sample. For each of the sample triplicate

188

measurements were conducted.

AC C

EP

TE D

M AN U

SC

RI PT

167

189 190

2.3.6. Particle size and particle size distribution

8

ACCEPTED MANUSCRIPT Particle size (D50 and D75) and its size distribution were determined by a laser diffraction

192

particle size analyzer (SALD-2200, Shimadzu Corporation, Japan) in dry measurement mode

193

and according to Koyama and Kitamura (2014). Working conditions are as follows: sample

194

suction type is hand shot ; Pressure is at 0.6-0.8 KPa, filtering rate is 5µm or larger;

195

Operating temperarure and the humidity are 25oC and 60% respectively, and particle size was

196

taken as D50 and D75 values expressed as µm. D50 is defined as the median diameter, and

197

D75 is the particle size corresponding to 75% of the particles being under size by mass. The

198

particle size distribution profile was constructed by the particle size and frequency of the

199

distribution.

200

M AN U

SC

RI PT

191

2.3.7. Particle morphology

202

Particle morphology was evaluated by field emission scanning electron microscopy (FE-

203

SEM) according to Mishra et al. (2014) and slite modifications. The powder were mounted

204

onto stubs with double-sided adhesive tape, and the samples were coated with a thin layer of

205

gold under vacuum. Then, the samples were examined on an FE scanning electron

206

microscope (SU8020, HITACHI, Japan) at 5 kV and 5000-fold magnification. The taking

207

microphotographs wasss carried out with a camera coupled to the microscopic (Cano-Chauca

208

et al., 2005)

209

2.3.8. Color characteristics

210

The color characteristics of the orange juice powder were determined by Colorimeter (CR-

211

200, Minolta Co., Japan). The results were expressed by the Hunter color value L, a and b

212

where +L indicates lightness and –L denotes darkness, +a denotes redness and –a as

213

greenness, +b as yellowness and –b as blueness. The color intensity denotes as Chroma value

214

and calculated by the formula = (a2+b2)1/2 and the Ho , hue angle was calculated by Ho = tan-

215

1

AC C

EP

TE D

201

(b/a). These parameters characterize the perception of the color of the orange juice powder. 9

ACCEPTED MANUSCRIPT The pure color of red, yellow and blue denotes by the values of the hue angle as 0°, 90°, 180°

217

and 270° respectively (Quek et al., 2007; Santhalakshmy et al., 2015). Before the sample

218

analysis the instrument was calibrated with a white ceramic plate (L = 91.0, a = +0.3165, b =

219

+0.3326) (Quek et al., 2007). The samples were analyzed as triplicates.

220

2.3.9. Ascorbic acid determination

221

Total ascorbic acid was determined by liquid chromatography according to Rodriguez et al.

222

(1992) with slight modifications. Forsample preparation,5-10 g of orange juice powder were

223

homogenized with 60-80 mL of 3% HPO3 (Wako, Japan)(PubChem CID:3084658) for one

224

minute. The obtained extract was filtered through Whatman No.542 paper, and brought to

225

100 mL with 3% HPO3 in a 100-mL volumetric flask. Prior to HPLC, the sample was filtered

226

through a 0.25 µm membrane filter. Each sample was run in triplicate.

227

A standard solution of L-ascorbic acid (Wako, Japan) was prepared with 3% HPO3 at

228

concentrations of 10, 30, 60, 90, 120, and 150 mg/L.

229

Ascorbic acid was analyzed by HPLC (LC 20AD Shimadzu Corporation, Japan) consisted of

230

UV detection at 254 nm, a Shim-Pack-ODS (75 x 4.6 mm) analytical column, a mobile phase

231

of methanol:/water (5:95, v/v), pH = 3 (H3PO4), a flow rate of 0.75 mL/min and an injection

232

volume of 20 µL.

233

The retention % of ascorbic acid content of orange juice powder was calculated by formula:

234

[(ascorbic acid content of orange juice powder (mg/100g) after drying) х 100/ ascorbic acid

235

content of COJ/MD mixture (mg/100g) before drying]

236

2.3.10. Water sorption isotherm

237

The equilibrium moisture contents of the orange juice powder were evaluated at different

238

values of water activity by isopiestic method (Rockland, 1987; Speiss and Wolf, 1987).

239

Seven saturated salt solutions were prepared as mentioned before. One gram of each of the

240

three orange powder were weighed in aluminum vials and equilibrated in the saturated salt

AC C

EP

TE D

M AN U

SC

RI PT

216

10

ACCEPTED MANUSCRIPT solutions in a dessicator at 25°C with H2SO4 to maintain proper relative humidity. Sample

242

equilibration took 3-4 weeks based on the change in weight, which did not exceed 0.01%.

243

The equilibrium moisture content was determined by oven drying at 70°C unit a constant

244

weight was obtained (AOAC, 1990).

245

Sorption isotherms are generally explained by the Brunauer–Emmett–Teller (BET) (Mrad et

246

al., 2012) and Guggenheim–Anderson–de Boer (GAB) (Bizot, 1983) mathematical models

247

based on the following empirical and theoretical parameters:

248

BET Model:
 = (

249

GAB Model:
 = (

250

In the present study, sorption isotherm data were fitted to the BET and GAB models and

251

analyzed by Origin Pro 8.5 software. The goodness of fit was evaluated by determining the

252

coefficient R2, and mean deviation modulus was calculated by the following formula (Goula

253

et al., 2008):

254

 =

255

2.4. Glass Transition Temperature

256

The glass transition temperatures, Tg, of the orange powder were determined by differential

257

scanning calorimetry (DSC) (DSC-60, Shimadzu Corporation, Japan) according to Shrestha

258

et al. (2007) . Five to ten grams of the orange powder were scanned in a hermetically sealed

259

20 µL DSC aluminum pan. All samples had previously been equilibrated over a saturated salt

260

solution with different relative humidities at 25°C. An empty aluminum pan was used as a

261

reference. The rate of thermal scanning was carried out in the following order: (1) isothermal

262

at -30°C for 1 min, (2) heating at a rate of 10°C/min from -30°C to a temperature just over

263

the apparent or predetermined Tg, (3) rapidly cooling at 50°C/min to -30°C and (4) heat

264

scanning at a rate of 10°C/min from -30°C to 200°C. All analyses were conducted in

M AN U

………………… ……(2)

  

……………………..(3)

AC C

EP



∑!

      )(     )

TE D



.................................................... (1)

SC

  

)(   )

RI PT

241

11

ACCEPTED MANUSCRIPT triplicate. The midpoint temperature of the DSC thermogram was considered the transition

266

temperature (Goula et al., 2008).

267

The glass transition temperature of a binary water-solid mixture depends on the plasticizing

268

effects of the water and is described by the Gordon-Taylor model (Gordon and Taylor, 1952).

269

Tg =

270

Where Tg, Tgs, and Tgw are the glass transition temperatures of the mixture, solids, and water,

271

respectively, xw is the mass fraction of water, and k is the Gordon-Taylor parameter.

($% )·'() *∙$% ∙'(% ($% )*∙

%$ ………………………(4)

SC

272

RI PT

265

3. Results and discussion

274

3.1. Effects of VSD and Maltodextrin Concentration on the Physical Properties of

275

Orange Juice Powder

276

To evaluate the physical conditions of the powder, water activity (aw), moisture content, glass

277

transition temperature (Tg) and product recovery were determined as shown in Table 1. The

278

moisture content of the powder had a great impact on the flowability, stickiness and storage

279

stability because of its plasticizing effects and crystallization behavior. The availability of

280

free water in a food component is expressed as water activity, which is responsible for

281

microbiological and biochemical reactions. All of the powder samples had water activity (aw)

282

values of 0.25±0.00-0.25±0.01 and moisture contents of 2.29 to 3.35%. whereas Shrestha et

283

al., (2007) found the wetness of the powder was high as indicated the moisture contents and

284

water activity of 4.3 - 4.5% and 0.30 to 0.40. The present research had the lower moisture

285

content and water activity than spray dried of concentrated orange juice by Goula and

286

Adamopoulos (2010). This was due to the higher moisture removal during vacuum spray

287

drying, and where a grater the heat gradient occurred in VSD between the steam and the

288

particles than hot air, so that there was a higher heat transfer into the particles, which

AC C

EP

TE D

M AN U

273

12

ACCEPTED MANUSCRIPT supplied the driving forces of the moisture removal. The moisture contents and other

290

parameters of powder produced by a COJ/MD ratio of 60:40 were not evaluated because of

291

the stickiness, low glass transition temperature of this powder. The moisture content and

292

water activity (aw) of powder decreased with increasing maltodextrin concentration likely

293

because the sugar and acid in juice-rich powder is highly hygroscopic and maltodextrin

294

increases the total solid of the feed and reduces the amount of water evaporation. This result

295

agrees with the findings of Shrestha et al. (2007) in spray-dried orange juice powder. The

296

water activity values of all powder samples were below 0.30, which improves powder

297

stability. The glass transition temperature Tg has a great impact on the surface stickiness of

298

the amorphous powder during spray drying (Lloyd et al., 1996). The Tg values of orange

299

juice/maltodextrin powder were investigated by DSC. The characteristic parameters Tgonset

300

and Tgmidpoint (oC) were observed from the DSC thermogram as shown in Table 1. The glass

301

transition temperature of the powder increased with increasing the amounts of maltodextrin

302

solids because of the higher molecular weight of maltodextrin. Roos and Karel (1996) stated

303

that the similar phenomenon, Tg value of binary mixture (sucrose:maltodextrin) increases

304

with increasing the maltodextrin concentration. Bhandari and Howes (2005) mentioned that

305

the recovery of spray dried sugar rich product was the direct functions of the temperature and

306

moisture of the outlet air. The increases in the maltodextrin solids also increased the cyclone

307

recovery because of low Tg sticks or collapses occur during drying, that’s why in the present

308

study COJ/MD 60:40 was failed due the low Tg (39.76°C) and higher moisture content

309

(≥ 4%) . In general several authors stated that the recovery of feed solid in the final product

310

increases with increasing the maltodextrin (Bhandari and Howes, 2005; Papadakis and Bahu,

311

1992). The pattern of the findings were similar to Shrestha et al. (2007). The highest cyclone

312

recovery, 63.3%, was obtained with 30:70 COJ/MD powders, which had the highest Tg of

313

approximately 86.63°C, and the lowest recovery of 53.0% was observed with 50:50 COJ/MD

AC C

EP

TE D

M AN U

SC

RI PT

289

13

ACCEPTED MANUSCRIPT powder at 61.78°C. These results were differed from Shrestha et al. (2007), who found that

315

50:50 orange juice/maltodextrin powder had Tg values of 66.4oC and no powder output in the

316

cyclone. These differences are caused because the present study was conducted on a low

317

temperature in a vacuum condition and superheated heated steam, as the drying medium,

318

which has a higher heat capacity than dry air, as a result the COJ/MD 50:50 powder were

319

produced at a low Tg (61.78°C).

320 321

Table 1. Effects of vacuum spray drying on the physicochemical properties of orange juice powder

322

Table 2 shows the hygroscopicity of the VSD powder in relations with the maltodextrin

323

concentration. Hygroscopicity of powder decreased from 0.195±0.02 to 0.143±0.01 gH₂O/g

324

with increasing maltodextrin concentration in Table 2. This was in an agreement with Tonon

325

et al. (2008). This trend was also similar to that of Cai and Corke (2000) and Rodríguez-

326

Hernández et al. (2005) for spray-dried Amaranthus and cactus pear juice powder. According

327

to Table 2 it is indicated that by increasing the maltodextrin, solubility of the powder

328

increases and rehydration index were greatly reduced. Because the maltodextrin has superior

329

solubility in water (Cano-Chauca et al., 2005), the fast rehydration may be due to the low

330

moisture contents of the powder (Goula et al., 2008).The solubility index of the potato flour

331

was increased with increasing the maltodextrin solids (Grabowski et al., 2006). The bulk

332

densities of the powder were not significantly different at p≤ 0.05. The determined bulk

333

density of VSD orange juice powder agreed with that of Chegini and Ghobadian (2005).

334

Table 2. Physical properties of vacuum spray - dried orange juice powder

335

Table 2 shows that the VSD orange juice powder particle diameters were significantly

336

different at p≤ 0.05. The median (D50) values of COJ/MD 30:70 powder had the lowest peak

337

with a smaller volume distribution and smaller particle diameter as shown in Figure 2.

AC C

EP

TE D

M AN U

SC

RI PT

314

14

ACCEPTED MANUSCRIPT Powder containing a high amount of orange juice are rich in acid and sugar, which imparts a

339

high hygroscopicity and stickiness to the liquid feed; as a result, larger particles are produced

340

during spray drying (Shrestha et al., 2007; Chegini and Ghobadian, 2005). Powders with

341

50:50 COJ/MD had significantly (at p≤ 0.05) higher particle size and size distributions. These

342

trends are similar to the findings described by Shrestha et al. (2007); however, powder

343

particle sizes in the present study were much smaller because of the difference atomization at

344

a lower temperature. Tonon et al. (2011) stated that when the inlet temperature is low, the

345

particle remains shrunk and maintains a smaller diameter, whereas at higher temperatures,

346

faster drying rates and higher swelling form larger particles. D75 powder particles with

347

COJ/MD ratios of 40:60 and 30:70 were not significantly different at p≤ 0.05.

348

Fig 2. Particle size distributions of Vacuum spray-dried orange juice powder

349

3.2. Effects of VSD and maltodextrin concentration on ascorbic acid retention

350

Table 3 shows the ascorbic acid contents of the VSD spray-dried orange juice powder with

351

different concentration of maltodextrin before and after vacuum spray drying. The retention

352

percentage of vitamin C was also calculated based on the vitamin C content of the orange

353

juice powder before and after vacuum spray drying. Vitamin C is an important water-soluble

354

and highly heat-sensitive natural substance. According to Goula and Adamopoulos (2006)

355

stated that vitamin C losses in tomato pulps occur in spray drying due to the thermal

356

degradation and oxidation. A goal of the present study was to retain the maximum amount of

357

vitamin C in the powder. The powder produced from COJ/MD 50:50 had significantly higher

358

amounts of vitamin C and also had the highest retention percentage (71.01%) as shown in

359

Table 3. As conventional spray drying of gooseberries had a 62.1% loss of vitamin C at

360

140°C (Thankitsunthorn et al., 2009) and the percentage of vitamin C retention varied from

361

28 to 51% during spray drying of cactus pear juice powder at 205oC and 225oC (Rodríguez-

AC C

EP

TE D

M AN U

SC

RI PT

338

15

ACCEPTED MANUSCRIPT Hernández et al., 2005). Kaya et al. (2010) also reported that 27.47% of vitamin C was

363

retained during drying of kiwifruits without carrier agent at 60oC. During spray drying of

364

passion fruit with matodextrin at 180oC and 190oC, retained percentage of vitamin C were

365

from 39.73 to 56.89 % (Angel et al., 2009). It means that vitamin C is highly sensitive to heat

366

and oxidation. That is why low temperature treatment and carrier agent retain maximum

367

vitamin C in the present study. In this study VSD operates with superheated steam, and

368

powder were captured under depression conditions at a very low temperature. With the

369

addition of maltodextrin, the vitamin C contents of the orange juice powder decreased from

370

126.4 -74.94mg/100g, because of the higher amount of solid come from maltodextrin. This

371

agrees with the findings of Mishra et al. (2014) on the spray drying of Amla juice with

372

different concentrations of maltodextrin.

373

Table 3. Effects of vacuum spray drying on Ascorbic acid contents of orange juice

374

powder

375

3.3. Effects of VSD and maltodextrin concentration on the color properties of orange

376

juice powder

377

The color of the powder had great impact on the reconstituted juice. In general, the color

378

attributes significantly depend on the drying conditions and carrier agents. The color

379

parameters L, a, b, chroma and Hue angle (Ho) are shown in Table 3. In the present study,

380

maltodextrin was white in color, whereas the VSD produced orange powder had a bright

381

yellow color. The color parameters of the 50:50 COJ/MD powder, highest yellowness (+b)

382

and color intensity (chroma), were significantly different from the other powder at p≤ 0.05.

383

With increases in the maltodextrin concentration, the yellowness and chroma values of the

384

40:60 and 30:70 COJ/MD powder decreased while lightness (+L), greenness (-a), and hue

385

angle (Ho) increased because of the lower amount of orange juice solids present in the

AC C

EP

TE D

M AN U

SC

RI PT

362

16

ACCEPTED MANUSCRIPT powder. This agrees with the findings of Mishra et al. (2014) on the spray drying of Amla

387

juice with different concentrations of maltodextrin.(Quek et al., 2007) observed that if the

388

powder were produced with 10% or higher concentration of maltodextrin, they lose their

389

attractive red-orange color. The results of this study contrast with that of Shrestha et al.

390

(2007) and Saikia et al. (2014), who were found that orange powder had low yellowness (+b)

391

and high redness (+a) and low lightness (+L) in their color, which may be due to the higher

392

inlet temperature (160°C) and oxidation effects during the drying process. In the present

393

study, the low drying temperature of 50-60°C and in a vacuum condition of VSD minimized

394

the color loss of the powder.

395

Table 4. Color characteristics of VSD concentrated orange juice/maltodextrin powder

396

3.4. Particle Morphology of the VSD Orange Juice Powder

397

Vacuum spray-dried powder prepared with different concentrations of orange juice and

398

maltodextrin had spherical shapes with different sizes, which was observed using FE-SEM

399

given in Figure 3. The micrograph shows that the morphology of the particles depends on the

400

concentration of the carrier agent and the drying conditions. The 30:70 COJ/MD powder

401

produced with higher amounts of maltodextrin had the smoothest surface with smaller

402

spherical shapes and no shrinkage. This agrees with the results published by Ravindra et al.

403

2013, who stated that a higher maltodextrin concentration was more susceptible to shrinkage

404

during the spray drying of white-flowered O. stamineus plant extract and observed that the

405

particle surface becomes smoother when the maltodextrin concentration increases from

406

0.53% to 10.67%. The 40:60 COJ/MD powder had smoother spherical shapes, but some

407

smaller particles adhered to the larger ones, making the particle size larger than that of the

408

30:70 COJ/MD powder. The VSD powder with 50:50 COJ/MD had dented surfaces with

409

wrinkles and deformation compared to the other powder. These results agree with Ferrari et

AC C

EP

TE D

M AN U

SC

RI PT

386

17

ACCEPTED MANUSCRIPT al. (2012), in which the spray drying of blackberry powder with lower maltodextrin

411

concentrations resulted in particles that strongly adhered to the surfaces and subsequently

412

formed agglomerates with larger particles, which increased the particle sizes.

413 414 415

Fig 3. Micrographs of VSD orange juice powder particles at different maltodextrin concentrations obtained by FE-SEM (a) OJC/MD 50:50, (b) OJC/MD 40:60 and (c) OJC/MD 30:70

RI PT

410

416

3.5. Water Sorption Isotherms of the Vacuum Spray-Dried Orange Juice Powder

418

Water sorption isotherms were constructed for the vacuum spray-dried orange juice powder,

419

and the experimental moisture sorption data for the equilibrium moisture contents (EMC, Xe

420

g H2O/g of dry solid) are given in Table 5. With increases in water activity (aw), the

421

equilibrium moisture contents (Xe) of the powder also increased. Experimental data were also

422

fitted to the BET and GAB models. The coefficients of determination and mean deviation

423

modulus were calculated to find the best fit model. The sorption isotherms for the orange

424

juice powder predicted by the BET and GAB models showed a typical sigmoid curve and a

425

type III (J shape) and typical of sugar-rich products according to Brunauer’s classification as

426

shown in Fig 4. According to Gabas et al. (2007) observed similar types of curve for vacuum-

427

dried pineapple juice powder with maltodextrin or gum Arabic as carrier agents. The

428

nonlinear regression data for the GAB and BET models are presented in Table 6. GAB

429

models resulted in the highest R2 and lowest Me values, similar to Goula et al. (2008) Tonon

430

et al. (2009) for spray-dried Acai juice and tomato pulp powder, respectively. Both the BET

431

and GAB models had satisfactory fitted values for powder produced by COJ/MD 30:70

432

powder. According to Table 6, GAB model was highly fitted to the experimental data of the

433

orange juice powder, because this model has the third parameter k and it can predict over

AC C

EP

TE D

M AN U

SC

417

18

ACCEPTED MANUSCRIPT range of water activity (0.05 ~ 0.9) than BET model fails over water activity 0.5 (Jonquières

435

and Fane, 1998; Goula et al., 2008).

436

Table 5. Equilibrium moisture content (EMC) and Glass transition temperature (Tg) of

437

VSD orange juice powder

438

The monolayer moisture content Xm is the safest moisture content that strongly absorbs to

439

specific sites on the food surface at a given temperature. The Xm values of orange

440

juice/maltodextrin powder range from 3.60 to 4.78% using BET models, whereas GAB

441

models resulted in Xm values of 3.54-4.03%. These results agree with Tonon et al. (2009) and

442

Caparino et al. (2013) for spray-dried Acai juice powder and freeze-dried mango powder,

443

respectively. COJ/MD 50:50 powder exhibited the highest water adsorption (Fig 4) because

444

of the high hygroscopic nature of powder rich in orange juice solids. Conversely, powder

445

with higher maltodextrin concentrations are less hydrolyzed and have fewer hydrophilic

446

groups, causing them to absorb less moisture and which explains its lower hygroscopicity

447

(Cai and Corke, 2000).

448 449

Fig 4. Water sorption isotherm data for VSD orange juice/maltodextrin powder at 25°C with curves fitted using the (a) BET and (b) GAB models

450

The physical characteristics of orange juice powder were investigated during storage at 25°C

451

with different water activities. The powders were changed after reaching equilibrium with

452

saturated salt solutions of specific water activities. Free-flowing powder had water activities

453

below 0.53, and stickiness or agglomeration was observed above an aw of 0.53. At a relative

454

humidity above 75%, collapse and liquefaction were observed in the powder.

455

Table 6. Estimated BET and GAB model parameters for Vacuum spray dried orange

456

juice powder

AC C

EP

TE D

M AN U

SC

RI PT

434

19

ACCEPTED MANUSCRIPT 457

3.6. Glass Transition Phenomenon of VSD Orange Juice/Maltodextrin Powder

458

The glass transition temperature (Tg) of orange juice powder was investigated by DSC.

459

Based on the transitions occur at Tgonset, but the present study considered midpoint

460

temperature of

461

temperature of the orange juice powder, furthermore it was used to establish the relationship

462

between Tg, water activity and water content of the orange juice powder. The glass transition

463

temperature (Tg) of orange juice powder was determined at different water activity levels as

464

shown in Table 5 and it showed an inverse relationship between Tg and water activity. With

465

increasing the water activity the Tg of orange juice powder decreases. Similar trend

466

determined by Tonon et al. (2009) and (Goula and Adamooulos, 2010) for spray-dried Acai

467

juice and concentrated orange juice powder.

468

Fig. 5. Relationship between the glass transition temperature and water content of VSD

469

orange juice powder with curves fitted to the Gordon-Taylor model

470

During storage and also in the processing orange juice powder become agglomerated and

471

caking occur due to the plasticizing effects of water on the particle surface. These physical

472

phenomena were successfully explained and predicted by the glass transition concept (Chuy

473

and Labuza, 1994). The experimental data were fitted to the Gordon-Taylor model to predict

474

the Tg of the binary solid mixture using Tgw = -135°C for pure water. The predicted values

475

are given in Table 7, and the fitted curve is shown in Fig 5; satisfactory values for R2 and Me

476

were obtained. Due to the plasticizing effects of water, the Tg of the powder decreased with

477

increasing water content. Similar results were obtained by Wang et al. (2008) and Goula and

478

Adamopoulos (2010) in the spray drying of gooseberry and concentrated orange juice powder

479

respectively. Table 7 shows that the Tgs values changed from 71.89 to 103.82°C with an

480

increasing maltodextrin solid content from 50 to 70%, and the k values predicted by the

AC C

EP

TE D

M AN U

SC

RI PT

the change in heat capacity was taken as the safest the glass transition

20

ACCEPTED MANUSCRIPT Gordon-Taylor model were between 2.50 and 3.35, which were higher than the values (0.83-

482

0.85) obtained by Goula and Adamopoulos (2010) for spray-dried concentrated orange juice

483

with maltodextrin. These differences in the Gordon-Taylor parameter may result from the

484

different feed compositions and drying conditions of the orange juice/maltodextrin powder.

485 486 487

Table 7. Estimated Gordon Taylor Parameters for VSD concentrated orange juice powder

488

4. Conclusions

489

Vacuum spray drying of concentrated orange juice with maltodextrin using superheated

490

steam as the drying medium; the effects of VSD on the physicochemical properties of

491

resulting powder were also evaluated. The produced VSD orange powder were stable due to

492

their low moisture content (2.29-3.49%) , low water activity (0.15-0.25) and higher glass

493

transition temperature. VSD orange juice powder retained a maximum amount of 71.01%

494

ascorbic acid. The maltodextrin concentration in relation to the glass transition temperature

495

was also optimized. Cyclone recovery increased to 63.3% with increasing as Tg and

496

maltodextrin. The maximum color retained by vacuum spray drying of orange juice powder.

497

The particle sizes of the VSD orange powder were smaller, smooth and spherical in

498

morphology. Overall, the sorption behavior of the orange juice powder exhibited a type III

499

sigmoid curve, and the highest and lowest water adsorption occurred at aw values above and

500

below 0.53, respectively. The experimental water adsorption data were satisfactory correlated

501

by both the BET and GAB models. Based on the stability and product recovery the present

502

study concluded that COJ/MD 30:70 by weight can be used in industrially to produce orange

503

juice powder.

AC C

EP

TE D

M AN U

SC

RI PT

481

504

21

ACCEPTED MANUSCRIPT Acknowledgements

506

The author thanks the Tanabe Engineering Co. Ltd., Himeji Technology Center, Japan for the

507

technical support to successfully carry out this research.

508

References

509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547

A. ArabHosseini, W.H., 2005. Modeling of the Equilibrium Moisture Content (EMC) of Tarragon (Artemisia Dracunculus L.). Int. J. Food Eng. 1 2005 5. Adhikari, B., Howes, T., Bhandari, B.R., Truong, V., 2003. Characterization of the Surface Stickiness of Fructose–Maltodextrin Solutions During Drying. Dry. Technol. 21, 17– 34. doi:10.1081/DRT-120017281 Angel, R.C.M., Espinosa-Muñoz, L.C., Aviles-Aviles, C., González-García, R., MoscosaSantillán, M., Grajales-Lagunes, A., Abud-Archila, M., 2009. Spray-drying of passion fruit juice using lactose-maltodextrin blends as the support material. Braz. Arch. Biol. Technol. 52, 1011–1018. doi:10.1590/S1516-89132009000400026 AOAC, 1990. Official Methods of Analysis, 14th ed. Association of Official Analytical Chemists, Washington, DC Bhandari, B., Howes, T., 2005. Relating the Stickiness Property of Foods Undergoing Drying and Dried Products to their Surface Energetics. Dry. Technol. 23, 781–797. doi:10.1081/DRT-200054194 Bizot, H., 1983. Using the GAB model to construct sorption isotherms. In Physical Properties of Foods. (Jowitt, R., Escher, F., Hallstrom, B.M., Spiess, W.E., Vos, G., Eds., PP. 42–54) Applied Science, London. Brennan, J.G., Herrera, J., Jowitt, R., 1971. A study of some of the factors affecting the spray drying of concentrated orange juice, on a laboratory scale. Int. J. Food Sci. Technol. 6, 295–307. doi:10.1111/j.1365-2621.1971.tb01618.x Cai, Y. z., Corke, H., 2000. Production and Properties of Spray-dried Amaranthus Betacyanin Pigments. J. Food Sci. 65, 1248–1252. doi:10.1111/j.1365-2621.2000.tb10273.x Cano-Chauca, M., Stringheta, P.C., Ramos, A.M., Cal-Vidal, J., 2005. Effect of the carriers on the microstructure of mango powder obtained by spray drying and its functional characterization. Innov. Food Sci. Emerg. Technol. 6, 420–428. doi:10.1016/j.ifset.2005.05.003 Caparino, O.A., Sablani, S.S., Tang, J., Syamaladevi, R.M., Nindo, C.I., 2013. Water Sorption, Glass Transition, and Microstructures of Refractance Window– and FreezeDried Mango (Philippine “Carabao” Var.) Powder. Dry. Technol. 31, 1969–1978. doi:10.1080/07373937.2013.805143 Carolina, B.C., Carolina, S., Zamora, M.C., Jorge, C., 2007. Glass transition temperatures and some physical and sensory changes in stored spray-dried encapsulated flavors. LWT - Food Sci. Technol. 40, 1792–1797. doi:10.1016/j.lwt.2006.10.009 Chegini, G.R., Ghobadian, B., 2005. Effect of Spray-Drying Conditions on Physical Properties of Orange Juice Powder. Dry. Technol. 23, 657–668. doi:10.1081/DRT200054161 Chegini, G.R., Khazaei, J., Ghobadian, B., Goudarzi, A.M., 2008. Prediction of process and product parameters in an orange juice spray dryer using artificial neural networks. J. Food Eng. 84, 534–543. doi:10.1016/j.jfoodeng.2007.06.007

AC C

EP

TE D

M AN U

SC

RI PT

505

22

ACCEPTED MANUSCRIPT Chuy, L.E., Labuza, T.P., 1994. Caking and Stickiness of Dairy-Based Food Powder as Related to Glass Transition. J. Food Sci. 59, 43–46. doi:10.1111/j.13652621.1994.tb06893.x Ferrari, C.C., Germer, S.P.M., Aguirre, J.M. de, 2012. Effects of Spray-Drying Conditions on the Physicochemical Properties of Blackberry Powder. Dry. Technol. 30, 154–163. doi:10.1080/07373937.2011.628429 Gabas, A.L., Telis, V.R.N., Sobral, P.J.A., Telis-Romero, J., 2007. Effect of maltodextrin and arabic gum in water vapor sorption thermodynamic properties of vacuum dried pineapple pulp powder. J. Food Eng. 82, 246–252. doi:10.1016/j.jfoodeng.2007.02.029 Gordon, M., Taylor, J.S., 1952. Ideal copolymers and the second-order transitions of synthetic rubbers. i. non-crystalline copolymers. J. Appl. Chem. 2, 493–500. doi:10.1002/jctb.5010020901 Goula, A.M., Adamopoulos, K.G., 2010. A new technique for spray drying orange juice concentrate. Innov. Food Sci. Emerg. Technol. 11, 342–351. doi:10.1016/j.ifset.2009.12.001 Goula, A.M., Adamopoulos, K.G., 2006. Retention of Ascorbic Acid during Drying of Tomato Halves and Tomato Pulp. Dry. Technol. 24, 57–64. doi:10.1080/07373930500538709 Goula, A.M., Adamopoulos, K.G., Kazakis, N.A., 2004. Influence of Spray Drying Conditions on Tomato Powder Properties. Dry. Technol. 22, 1129–1151. doi:10.1081/DRT-120038584 Goula, A.M., Karapantsios, T.D., Achilias, D.S., Adamopoulos, K.G., 2008. Water sorption isotherms and glass transition temperature of spray dried tomato pulp. J. Food Eng. 85, 73–83. doi:10.1016/j.jfoodeng.2007.07.015 Grabowski, J.A., Truong, V.-D., Daubert, C.R., 2006. Spray-Drying of Amylase Hydrolyzed Sweetpotato Puree and Physicochemical Properties of Powder. J. Food Sci. 71, E209– E217. doi:10.1111/j.1750-3841.2006.00036.x Greenspan, L., 1977. Humidity fixed points of binary saturated aqueous solutions. J. Res. Nat. Bur. Stand. Sec. A: Phys. Chem. 81, 89–96. Jonquières, A., Fane, A., 1998. Modified BET models for modeling water vapor sorption in hydrophilic glassy polymers and systems deviating strongly from ideality. J. Appl. Polym. Sci. 67, 1415–1430. doi:10.1002/(SICI)10974628(19980222)67:8<1415::AID-APP7>3.0.CO;2-H Kaya, A., Aydın, O., Kolaylı, S., 2010. Effect of different drying conditions on the vitamin C (ascorbic acid) content of Hayward kiwifruits (Actinidia deliciosa Planch). Food Bioprod. Process. 88, 165–173. doi:10.1016/j.fbp.2008.12.001 Kitamura, Y., Yanase, Y., 2011. Development of vacuum spray drying system for probiotics powder. Comeau. M. A (Eds.), New Topics in Food Engineering (pp 171-231). Nova Science Publishers, Inc. New York.

588 589 590 591 592 593 594 595

Kitamura, Y., Itoh, H., Echizen, H., Satake, T., 2009. Experimental Vacuum Spray Drying of Probiotic Foods Included with Lactic Acid Bacteria. J. Food Process. Preserv. 33, 714–726. doi:10.1111/j.1745-4549.2008.00299.x Koyama, M., Kitamura, Y., 2014. Development of a new rice beverage by improving the physical stability of rice slurry. J. Food Eng. 131, 89–95. doi:10.1016/j.jfoodeng.2014.01.030 Lee, K.-H., Wu, T.-Y., Siow, L.-F., 2013. Spray drying of red (Hylocereus polyrhizus) and white (Hylocereus undatus) dragon fruit juices: physicochemical and antioxidant

AC C

EP

TE D

M AN U

SC

RI PT

548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587

23

ACCEPTED MANUSCRIPT properties of the powder. Int. J. Food Sci. Technol. 48, 2391–2399. doi:10.1111/ijfs.12230 Levine, H., Slade, L., 1990. Cryostabilization Technology: Thermoanalytical Evaluation of Food Ingredients and Systems in Thermal Analysis of Foods (Ma, C.Y.; Hawalkar, V.R. Eds., pp. 221-305) ElsevierApplied Science, London.

601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642

Lloyd, R.J., Dong Chen, X., Hargreaves, J.B., 1996. Glass transition and caking of spraydried lactose. Int. J. Food Sci. Technol. 31, 305–311. doi:10.1046/j.13652621.1996.00352.x Mishra, P., Mishra, S., Mahanta, C.L., 2014. Effect of maltodextrin concentration and inlet temperature during spray drying on physicochemical and antioxidant properties of amla (Emblica officinalis) juice powder. Food Bioprod. Process. 92, 252–258. doi:10.1016/j.fbp.2013.08.003 Mrad, N.D., Bonazzi, C., Boudhrioua, N., Kechaou, N., Courtois, F., 2012. Moisture Sorption Isotherms, Thermodynamic Properties, and Glass Transition of Pears and Apples. Dry. Technol. 30, 1397–1406. doi:10.1080/07373937.2012.683843 Papadakis, S.E., Bahu, R.E., 1992. The Sticky Issues of Drying. Dry. Technol. 10, 817–837. doi:10.1080/07373939208916484 Quek, S.Y., Chok, N.K., Swedlund, P., 2007a. The physicochemical properties of spray-dried watermelon powder. Chem. Eng. Process. Process Intensif. 46, 386–392. doi:10.1016/j.cep.2006.06.020 Quek, S.Y., Chok, N.K., Swedlund, P., 2007b. The physicochemical properties of spray-dried watermelon powder. Chem. Eng. Process. Process Intensif., Advances in the Application of Chemical Engineering Principles in Food Industry 46, 386–392. doi:10.1016/j.cep.2006.06.020 Rockland, 1987. Water Activity: Theory and Applications to Food. CRC Press. Rodríguez-Hernández, G.R., González-García, R., Grajales-Lagunes, A., Ruiz-Cabrera*, M.A., Abud-Archila, M., 2005. Spray-Drying of Cactus Pear Juice (Opuntia streptacantha): Effect on the Physicochemical Properties of Powder and Reconstituted Product. Dry. Technol. 23, 955–973. doi:10.1080/DRT-200054251 Rodriguez, M.A.R., Oderiz, M.L.V., Hernandez, J.L., Lozano, J.S., 1992. Determination of Vitamin C and Organic Acids in Various Fruits by HPLC. J. Chromatogr. Sci. 30, 433–437. doi:10.1093/chromsci/30.11.433 Roos, Y. H., Karel, M., Kokini, J. L., 1996. Glass transitions in low moisture and frozen foods: Effects on shelf life and quality. Scientific status summary. Food Technology, 50, 95–108 Saénz, C., Tapia, S., Chávez, J., Robert, P., 2009. Microencapsulation by spray drying of bioactive compounds from cactus pear (Opuntia ficus-indica). Food Chem. 114, 616– 622. doi:10.1016/j.foodchem.2008.09.095 Saikia, S., Mahnot, N.K., Mahanta, C.L., 2014. Effect of Spray Drying of Four Fruit Juices on Physicochemical, Phytochemical and Antioxidant Properties. J. Food Process. Preserv. n/a–n/a. doi:10.1111/jfpp.12395 Santhalakshmy, S., Don Bosco, S.J., Francis, S., Sabeena, M., 2015. Effect of inlet temperature on physicochemical properties of spray-dried jamun fruit juice powder. Powder Technol. 274, 37–43. doi:10.1016/j.powtec.2015.01.016 Semyonov, D., Ramon, O., Shimoni, E., 2011. Using ultrasonic vacuum spray dryer to produce highly viable dry probiotics. LWT - Food Sci. Technol. 44, 1844–1852. doi:10.1016/j.lwt.2011.03.021

AC C

EP

TE D

M AN U

SC

RI PT

596 597 598 599 600

24

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

Shrestha, A.K., Ua-arak, T., Adhikari, B.P., Howes, T., Bhandari, B.R., 2007. Glass Transition Behavior of Spray Dried Orange Juice Powder Measured by Differential Scanning Calorimetry (DSC) and Thermal Mechanical Compression Test (TMCT). Int. J. Food Prop. 10, 661–673. doi:10.1080/10942910601109218 Spiess, W.E.L. Wolf, W., 1987. Critical evaluation of methods to determine moisture sorption isotherms. InWater Activity: Theory and Applications to Foods; (Rockland, L.B., Beuchat, L.R., Eds., pp.215-234) MarcelDekker, New York. Thankitsunthorn, S.,Thawornphiphatdit, C., Laohaprasit, N., Srzednicki, G., 2009. Effects of drying temperature on quality of dried Indian Gooseberry powder. International Food Research Journal 16, 355-361 Tonon, R.V., Baroni, A.F., Brabet, C., Gibert, O., Pallet, D., Hubinger, M.D., 2009. Water sorption and glass transition temperature of spray dried açai (Euterpe oleracea Mart.) juice. J. Food Eng. 94, 215–221. doi:10.1016/j.jfoodeng.2009.03.009 Tonon, R.V., Brabet, C., Hubinger, M.D., 2008. Influence of process conditions on the physicochemical properties of açai (Euterpe oleraceae Mart.) powder produced by spray drying. J. Food Eng. 88, 411–418. doi:10.1016/j.jfoodeng.2008.02.029 Tonon, R.V., Freitas, S.S., Hubinger, M.D., 2011. Spray Drying of Açai (euterpe Oleraceae Mart.) Juice: Effect of Inlet Air Temperature and Type of Carrier Agent. J. Food Process. Preserv. 35, 691–700. doi:10.1111/j.1745-4549.2011.00518.x Tsourouflis, S., Flink, J.M., Karel, M., 1976. Loss of structure in freeze-dried carbohydrates solutions: Effect of temperature, moisture content and composition. J. Sci. Food Agric. 27, 509–519. doi:10.1002/jsfa.2740270604 Wang, H., Zhang, S., Chen, G., 2008. Glass transition and state diagram for fresh and freezedried Chinese gooseberry. J. Food Eng. 84, 307–312. doi:10.1016/j.jfoodeng.2007.05.024 Young, F.E., 1967. Requirement of glucosylated teichoic acid for adsorption of phage in Bacillus subtilis 168. Proc. Natl. Acad. Sci. U. S. A. 58, 2377–2384.

TE D

643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671

Nomenclature

673

aw = Water activity,

674

Xe = Equilibrium Moisture Content g water/g of solids,

675

Xm = Monolayer Moisture Content,

676

CBET = BET Model constant,

677

CGAB = GAB model constant,

678

KGAB = GAB model constant,

679

Me = Percent of mean deviation modulus,

680

N = Population of experimental data,

681

Ve = Experimental value,

682

Vp = Predicted Value

683

Tg = Glass transition temperature of the mixture

AC C

EP

672

25

ACCEPTED MANUSCRIPT 684

Tg = Glass transition temperature of solids

685

Tgw = Glass transition temperature water

686

Xw = Mass fraction of water

687

k = Gordon-Taylor parameter

688

TSS = Total Soluble Solids

AC C

EP

TE D

M AN U

SC

RI PT

689

26

ACCEPTED MANUSCRIPT Table 1. Effects of vacuum spray drying on the physicochemical properties of an orange juice powder Types of Powder

Moisture %

Water activity

COJ/MD 50:50

3.35±0.28a

0.25±0.00a

Glass Transition Temperature Tg °C TgOneset Tgmidpoint c 56.42±1.35 61.78±0.97c

COJ/MD 40:60

2.57±0.12b

0.17±0.02b

68.01±1.52b

75.94±0.92b

60.5

COJ/MD 30:70 2.29±0.16bc

0.15±0.01b

79.35±1.28a

86.63±0.97a

63.3

Cyclone recovery, %

RI PT

53

SC

The values are mean ± S.D of three independent determinations. The means with different superscripts in a column differs significantly (p≤ 0.05)

M AN U

Table 2. Physical properties of vacuum spray - dried orange juice powder Rehydration,

Bulk density

gH₂O/g

Sec

g/mL

D50

D75

COJ/MD 50:50

0.195±0.02a

253.65±1.05a

0.70±0.03a

7.75±0.25a

12.84±.62a

COJ/MD 40:60

0.188±0.03b

237.40±1.25b

0.72±0.02a

6.36±0.45b

8.69±0.21b

COJ/MD 30:70

0.143±0.01c

122.34±1.56c

0.73±0.02a

6.02±0.16c

7.68±0.26b

TE D

Hygroscopicity,

Types of Powder

Particle size, µm

AC C

EP

The values are mean ± S.D of three independent determinations. The means with different superscripts in a column differs significantly (p≤ 0.05).

Table 3. Effects of vacuum spray drying on Ascorbic acid contents of orange juice powder Vitamin C before Vitamin C after drying Retention of Types of Powder

drying mg/100g

mg/100g

Vitamin-C, %

COJ/MD 50:50

178±1.31a

126.4±2.17a

71.01

COJ/MD 40:60

120.6±2.16b

81.71±1.51b

67.75

COJ/MD 30:70

115.5±3.27b

74.94±2.12b

64.88

The values are mean ± S.D of three independent determinations. The means with different superscripts in a column differs significantly (p≤ 0.05)

ACCEPTED MANUSCRIPT Table 4. Color characteristics of VSD concentrated orange juice powder Color parameters Types of Powder L a b Chroma b b a COJ/MD 50:50 88.8±0.57 -2.84±0.40 46.52±1.02 46.61±0.5a

Hue angle Ho 93.43±0.60b

91.78±1.12a -3.99±0.21a

37.39±0.78b

37.60±0.21b 96.03±0.35a

COJ/MD 30:70

92.86±0.51a -3.53±0.13ab

31.22±1.46c

31.41±0.13c 96.40±0.53a

RI PT

COJ/MD 40:60

The values are mean ± S.D of three independent determinations. The means with different superscripts in a column differ significantly (p≤ 0.05)

COJ/MD 50:50

SC

Table 5. Equilibrium moisture content (EMC) and Glass transition temperature (Tg) of VSD orange juice powder COJ/MD 40:60

Water activity

EMC, Xe

Tg, °C

0.11

0.0236

61.78±1.23

0.22

0.0389

53.12±2.11

0.33

0.0575

45.16±0.05

0.53

0.0786

32.23±1.13

0.71

0.1325

0.75 0.84

Tg, °C

EMC, Xe

Tg, °C

0.0251

77.94±2.15

0.0195

88.63±2.12

0.0393

67.76±1.19

0.0378

74.85±1.17

0.0421

63.23±1.13

0.0398

68.13±1.12

0.0773

45.67±1.23

0.0685

55.13±2.13

15.15±1.19

0.1256

17.21±1.11

0.1194

30.12±1.13

0.178

-2.32±0.03

0.1585

4.14±0.92

0.1507

15.19±1.11

0.290

-30.25±0.95

0.2388

-23.96±1.21

0.2104

-12.67±1.23

TE D

M AN U

EMC, Xe

COJ/MD 30:70

EP

The values are mean ± S.D of three independent determinations.

AC C

Table 6. Estimated BET and GAB model parameters for Vacuum spray dried orange juice powders Model

BET

GAB

Model Parameters Xm CBET 2

R % Me Xm CGAB KGAB R2 % Me

COJ/ MD 50:50

COJ/MD 40:60

COJ/MD 30:70

0.0478 4.31 0.99 9.85 0.0403 7.73 1.02 0.98 6.63

0.0402 5.50 0.99 9.28 0.0395 8.24 0.99 0.99 6.92

0.0360 7.56 0.98 5.39 0.0354 8.96 0.98 0.99 8.41

ACCEPTED MANUSCRIPT

Table 7. Estimated Gordon Tailor Parameters for VSD concentrated orange powder COJ/MD 50:50 71.89 2.498 0.99 6.70

95.18 3.194 0.98 3.84

COJ/MD 30:70

AC C

EP

TE D

M AN U

SC

Tgs (°C) K R2 E (%)

COJ/MD 40:60

103.82 3.52 0.99 3.47

RI PT

Parameter

1 2

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Fig 1. Schematic flow diagram of the vacuum spray drying process with superheated steam

3

10

TE D

Orange Juice:Maltodextrin 30:70

EP

6

4

AC C

Frequency %

8

Orange Juice:Maltodextrin 40:60 Orange Juice:Maltodextrin 50:50

2

0

0.1

1

10

100

Particle size µm 4 5

Fig 2. Particle size distributions of vacuum spray- dried orange juice powders

1000

ACCEPTED MANUSCRIPT

(a)

SC

RI PT

6

EP

AC C

(c)

TE D

M AN U

(b)

7

Fig. 3. Micrographs of VSD orange juice powder particles at different maltodextrin

ACCEPTED MANUSCRIPT

concentration obtained by FE-SEM (a) COJ/MD 50:50, (b) COJ/MD 40:60 and (c) COJ/MD 30:70

0.30

(a)

Juice/MD 30:70 BET Model 50:50

Juice/MD 50:50 BET model 50:50

Juice/MD 40:60 BET model 40:60

RI PT

8 9

0.20

SC

0.15 0.10 0.05 0.00 0.1

0.2

0.3

(b)

0.30

Juice/ MD 50:50 GAB model 50:50

0.7

0.8

Juice/MD 40:60 GAB Model 40:60

Juice/MD 30:70 GAB mdel 3070

0.4

0.7

0.9

EP

0.25 0.20 0.15

AC C

Equilibrium Moisture Content g water/g solid

0.4 0.5 0.6 Water Activity

TE D

10

M AN U

g water/g solid

Equilibrium Moisture Content

0.25

0.10 0.05 0.00

0.1

0.2

0.3

0.5

Water Activity

11

0.6

0.8

0.9

ACCEPTED MANUSCRIPT

12 13

Fig 4. Water sorption isotherm data for VSD orange juice powders at 25°C with curves fitted using the (a) BET and (b) GAB models 100

Experimental parameter, J/MD 30:70 Gordon Taylor Parameter

RI PT

80 60 40

SC

20 0

-40 0.00

0.05

0.10

M AN U

-20

0.15

0.20

0.25

0.30

Water content (g/g of solid) COJ/MD 30:70

14

75 50

Experimental data for J/MD 40:60 Gordon Taylor Parameter

EP

25

TE D

o

Glass Transition Temperature, Tg C

100

0

AC C

-25

-50 0.00

0.05

0.10

0.15

0.20

Water conter (g/g of solid)

15

COJ/MD 40:60

0.25

0.30

ACCEPTED MANUSCRIPT

Experimental parameter, J/MD 50:50 Gordon Taylor Parameter

60

RI PT

40

20

0

-20

-40 0.00

0.05

0.10

0.15

0.20

Water content (g/g of solid)

16 17 18

0.25

0.30

M AN U

COJ/MD 50:50

SC

o

Glass Transition Temperature, Tg C

80

0.35

Fig. 5. Relationship between the glass transition temperature and water content of VSD orange juice powders with curves fitted to the Gordon-Taylor model

AC C

EP

TE D

19

ACCEPTED MANUSCRIPT

RI PT

SC

•

M AN U

•

The vacuum spray dryer (VSD) was used to turn concentrate Orange Juice into a powder at a low temperature (40-60oC) drying using maltodextrin as carrier and superheated steam as the heat source. VSD Orange Juice Powders were stable due to their low monolayer moisture content and low water activity. VSD Orange Juice Powders retained a maximum amount of thermo-sensitive functional ingredients.

TE D

•

EP

2 3 4 5 6 7 8

Highlights

AC C

1

1