Formulation of food emulsions using natural emulsifiers: Utilization of quillaja saponin and soy lecithin to fabricate liquid coffee whiteners

Accepted Manuscript Formulation of food emulsions using natural emulsifiers: Utilization of quillaja saponin and soy lecithin to fabricate liquid coffee whiteners Cheryl Chung, Alexander Sher, Philippe Rousset, Eric Andrew Decker, David Julian McClements PII:

S0260-8774(17)30151-6

DOI:

10.1016/j.jfoodeng.2017.04.011

Reference:

JFOE 8845

To appear in:

Journal of Food Engineering

Received Date: 22 February 2017 Revised Date:

30 March 2017

Accepted Date: 6 April 2017

Please cite this article as: Chung, C., Sher, A., Rousset, P., Decker, E.A., McClements, D.J., Formulation of food emulsions using natural emulsifiers: Utilization of quillaja saponin and soy lecithin to fabricate liquid coffee whiteners, Journal of Food Engineering (2017), doi: 10.1016/ j.jfoodeng.2017.04.011. 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

Formulation of Food Emulsions using Natural Emulsifiers: Utilization of

2

Quillaja Saponin and Soy Lecithin to Fabricate Liquid Coffee Whiteners

RI PT

3 4

Cheryl Chung1, Alexander Sher2, Philippe Rousset2, Eric Andrew Decker1,

5

David Julian McClements1* 1

Department of Food Science, University of Massachusetts, Amherst, MA, 01003, USA.

SC

6 7

2

Nestlé Development Center Marysville, 809 Collins Av, Marysville, OH, 43040, USA

9 10

Journal: Journal of Food Engineering

11

Submitted: February 2017

13

16

To whom correspondence should be addressed. Tel: (413) 545-1019; Fax: (413) 545-1262.

EP

15

*

E-mail: [email protected]

AC C

14

TE D

12

M AN U

8

1

ACCEPTED MANUSCRIPT

17

Abstract Rising consumer demand for food products made with natural and plant-based ingredients

19

has led to a search for natural alternatives to synthetic food ingredients. The present study

20

compared the ability of two natural small molecule surfactants – quillaja saponin (0.5 to 2.5%)

21

and soy lecithin (1 to 5%) – to stabilize 10% oil-in-water emulsions. Emulsion lightness

22

decreased with increasing emulsifier concentration in both systems, which was attributed to the

23

inherent color of the emulsifiers (increased absorption) and the decrease in droplet size

24

(decreased scattering). The mean droplet diameter decreased with increasing emulsifier

25

concentration (0.5 to 0.15 µm for quillaja saponin and 0.8 to 0.14 µm for soy lecithin) due to

26

their ability to cover more surface area. Both emulsifiers led to the formation of oil droplets with

27

a high negative charge (ζ = -45 to -70 mV), thereby generating a strong electrostatic repulsion

28

that helped protect them against aggregation. The emulsions remained physically stable when

29

added to an acidic hot coffee solution (85°C), with no visible phase separation or increase in

30

particle size. This study provides insight into the potential of two natural emulsifiers to form

31

stable emulsions suitable for application in coffee creamers.

SC

M AN U

TE D

Keywords: Natural Emulsifiers; Quillaja Saponin; Soy Lecithin; Coffee Creamer; Color

EP

33

AC C

32

RI PT

18

2

ACCEPTED MANUSCRIPT

34

1. Introduction Increasing consumer demand for food and beverage products containing natural ingredients

36

is a major trend in the food industry (FoodNavigator 2013, FoodNavigator 2016). For this

37

reason, many food manufacturers are searching for commercially viable natural ingredients that

38

have similar or better functional attributes than synthetic ones (Baines and Seal 2012). Emulsions

39

represent a major category of foods and beverages, including products such as milk, creams,

40

sauces, dressings, soups, desserts, beverages, and creamers (Dickinson 2009, McClements 2010,

41

Petrut, Danthine and Blecker 2016). Many of these products are currently stabilized by synthetic

42

emulsifiers, and so there is a need to replace them with natural alternatives (McClements and

43

Gumus 2016). The manufacture of high quality emulsion-based products depends on identifying

44

emulsifiers that possess the appropriate emulsifying and stabilizing properties required for the

45

specific food application (Dickinson and Leser 2013, McClements 2015). The emulsifiers

46

should: (a) rapidly adsorb to the oil droplet surfaces during homogenization so as to reduce the

47

interfacial tension and facilitate droplet disruption; (b) form a stable protective layer around the

48

oil droplets that prevents them from aggregating during manufacture, transport, and storage; (c)

49

be easy to use, and available in a reliable quantity and quality; and, (d) be economically viable

50

(Dickinson 2009, McClements 2015). In this study, we focus on two kinds of natural surfactant

51

that have considerable potential for replacing synthetic surfactants or animal-based emulsifiers in

52

many food products: quillaja saponin and soy lecithin (phospholipids) (McClements and Gumus

53

2016).

EP

TE D

M AN U

SC

RI PT

35

Quillaja saponins are natural extracts from the soapbark tree (Quillaja saponaria) normally

55

found in Chile, which exhibit excellent emulsifying properties due to their amphiphilic structure.

56

Saponins are relatively small molecules that consist of a hydrophobic aglycone and a hydrophilic

57

sugar moiety (Figure 1) (Augustin, Kuzina, Andersen and Bak 2011, Mitra and Dungan 1997,

58

Orczyk and Wojciechowski 2015, Tippel, Lehmann, von Klitzing and Drusch 2016). Numerous

59

studies have reported that they have strong surface activity (Golemanov, Tcholakova, Denkov,

60

Pelan and Stoyanov 2014, Stanimirova, Marinova, Tcholakova, Denkov, Stoyanov and Pelan

61

2011, Tippel, Lehmann, von Klitzing and Drusch 2016) and that they exhibit good emulsifying

62

properties (Ozturk, Argin, Ozilgen and McClements 2014, Yang, Leser, Sher and McClements

63

2013, Zhang, Bing and Reineccius 2016).

AC C

54

3

ACCEPTED MANUSCRIPT

Phospholipids consist of two non-polar fatty acids esterified to a glycerol backbone that has

65

a polar phosphate head group with a hydrophilic residue attached e.g., choline, inositol,

66

ethanolamine or serine (Figure 1) (Cui and Decker 2016, Küllenberg, Taylor, Schneider and

67

Massing 2012, Pichot, Watson and Norton 2013). The phospholipids used in food applications

68

are typically referred to as lecithins, which actually contain a complex mixture of different

69

phospholipids

70

phosphatidylinositol) and other lipids (such as triglycerides, glycolipids, or sterols) (Cabezas,

71

Diehl and Tomás 2016, Cui and Decker 2016). Some of the most common lecithins used in the

72

food industry are extracted from soybeans, eggs, milk, sunflower, rice, rapeseed, and canola seed

73

(Cui and Decker 2016, Küllenberg, Taylor, Schneider and Massing 2012, Liu, Waters, Rose, Bao

74

and King 2013, Wu and Wang 2003). Stable oil-in-water emulsions can be produced using

75

certain types of lecithin as emulsifiers, provided the level utilized is optimized (Cui and Decker

76

2016, Guiotto, Capitani, Nolasco and Tomás 2016, Komaiko, Sastrosubroto and McClements

77

2015, Pichot, Watson and Norton 2013, Zhu and Damodaran 2013).

as

phosphatidylcholine,

phosphotidylethanolamine

and

M AN U

SC

(such

RI PT

64

In the present study, we investigated the potential application of quillaja saponins and de-

79

oiled soy lecithin to form stable oil-in-water emulsions suitable for application in non-dairy

80

creamers, which are usually produced using dairy proteins as emulsifiers. The emulsions used for

81

this application must remain stable during storage at ambient temperature, and after

82

incorporation into hot coffee (which is a mildly acidic aqueous solution). The effect of emulsifier

83

type and concentration on the physical properties of the emulsions was therefore characterized,

84

and then the impact of their incorporation into hot coffee was determined.

85

2. Materials and methods

86

2.1 Materials

AC C

EP

TE D

78

87

Quillaja saponin with 14.1% surface active component (Q-Naturale 200®) was kindly

88

donated by Ingredion Incorporated (Westchester, Illinois, USA). De-oiled soy lecithin (Solec® F)

89

with a Hydrophile-Lipophile Balance (HLB) value around 7 and 97% surface active component

90

was a gift from DuPont Danisco (St. Louis, MO, USA). Medium chain triglyceride (Miglyol

91

812N) (MCT) was purchased from the Warner Graham Company (Cockeysville, MD, USA). A

92

dark roast coffee powder (Nescafe Clásico, Nestlé, Switzerland) was purchased from a local

4

ACCEPTED MANUSCRIPT

supermarket. All analytical grade chemicals used, including hydrochloric acid, sodium

94

hydroxide, sodium phosphate monobasic, sodium phosphate dibasic, calcium chloride, and

95

magnesium chloride were purchased from Fisher Scientific Company LLC (Pittsburgh,

96

Pennsylvania, USA). All concentrations reported in this study are on a percentage weight to

97

weight (w/w) basis and represented as ‘%’, unless otherwise stated.

98

2.2 Interfacial tension measurements

RI PT

93

The interfacial tension of the MCT – water interface was measured at varying surfactant

100

concentrations (based on their surface-active component) using a drop shape analysis instrument

101

fitted with a J-shaped needle (0.906 mm diameter) (DSA 100, Krüss, GmbH, Hamburg,

102

Germany). The interfacial tension measurement was performed using the instrument software

103

with the “Pendant Drop” method selected. For the quillaja saponins, oil phase consisted of MCT

104

and the water phase consisted of a phosphate buffer solution (10 mM at pH 7) containing

105

different levels of saponin. For the soy lecithins, the oil phase consisted of MCT containing

106

different levels of lecithin, while the water phase consisted of phosphate buffer solution (10 mM

107

at pH 7). All measurements were recorded for over 15 min, at 30 sec intervals, which was long

108

enough to ensure that the interfacial tension reached a steady value. The interfacial tension was

109

determined using the Young-Laplace equation using the instrument software.

110

2.3 Fabrication of emulsion systems

TE D

M AN U

SC

99

Emulsions containing 10% medium chain triglyceride (MCT) oil stabilized by 0.5 to 2.5%

112

quillaja saponin or 1 to 5% soy lecithin were fabricated using high pressure microfluidization.

113

For the quillaja saponin, different amounts of Q-Naturale 200 (to obtain final levels of 0.5 to

114

2.5%) were dispersed in 10 mM phosphate buffer (pH 7) and stirred until homogeneous at

115

ambient temperature. For the soy lecithin, different amounts of Solec F powder were dissolved in

116

the oil phase and stirred overnight at ambient temperature. Coarse emulsions were then

117

fabricated by blending the oil phase (10%) into the aqueous phase (90%) at 15, 000 rpm using a

118

high speed mixer (Bamix, Biospec Products, Bartlesville, Oklahoma, USA) for 1 min.

119

Subsequently, the coarse emulsions were passed through a high pressure homogenizer

120

(Microfluidizer M-110P, Microfluidics, Newton, MA, USA) at 20,000 psi for 3 passes to reduce

121

the droplet size. The homogenization was carried out at ambient temperature (25 – 27°C) and the

AC C

EP

111

5

ACCEPTED MANUSCRIPT

122

freshly produced emulsions had temperature around 40°C. The fresh emulsions were then

123

adjusted to pH 7.0 using 1 M sodium hydroxide (NaOH).

124

2.4 Application of emulsions in coffee system The potential application of the fabricated emulsions as non-dairy coffee creamers was

126

analyzed by measuring their physical properties and stability after being added to hot coffee.

127

White coffee was prepared by addition of 10 mL of the emulsions (10% oil), hereafter referred to

128

as “model creamers”, into 60 mL of 1% hot black coffee.

RI PT

125

The 1% hot black coffee solutions were prepared to contain 270 ppm hardness (expressed as

130

ppm of calcium carbonate, CaCO3) to mimic the highest hard water level typically found in

131

drinking water. A weighed amount of coffee, calcium chloride (1.95 mM) and magnesium

132

chloride (0.62 mM) were reconstituted in boiled distilled water to make hot black coffee

133

solutions containing 1% coffee and 270 ppm CaCO3 hardness. To make white coffee solutions,

134

10 mL of emulsions were added into 60 mL of black coffee solutions (85°C) and stirred until

135

homogeneous.

M AN U

SC

129

Model creamers with a fixed emulsifier concentration were used to establish the influence of

137

creamer level on the physical properties of white coffee: 1.5% quillaja saponin or 5% soy

138

lecithin. These emulsifier levels were selected because they produced physically stable

139

emulsions with high lightness (L*) values. In this series of experiments, 5, 10, 20 or 30 mL

140

emulsions were added to 60 mL of black coffee solutions (1% coffee with 270 ppm CaCO3

141

hardness).

142

2.5 Visual appearance and colorimetry measurements

AC C

EP

TE D

136

143

The visual appearance of emulsions and coffee solutions stored in glass test tubes was

144

recorded using a digital camera. The emulsions were stored overnight prior to analysis, while the

145

coffee solutions were cooled down prior to analysis.

146

The tristimulus color coordinates of the emulsions and coffee solutions were determined

147

using a colorimeter with a tristimulus absorption filter (ColorFlez EZ, HunterLab, Reston, VA,

148

USA). The L* (lightness), a* (red to green) and b* (blue to yellow) color coordinates of the

149

different samples were measured.

6

ACCEPTED MANUSCRIPT

150

2.6 Particle size and microscopy measurements The particle size distributions of the emulsions were determined using a laser diffraction

152

particle size analyzer (Beckman Coulter LS 12 320, Brea, CA, USA). About 0.3 to 1 mL of

153

undiluted sample were added into a measurement cell containing either 10 mM phosphate buffer

154

pH 7.0 (for emulsions) or distilled water with 270 ppm CaCO3 hardness (for coffee solutions) to

155

achieve an optimum obscuration rate of 40 to 55%. A refractive index of 1.333 was used for the

156

aqueous phase, while a value of 1.448 was used for the oil phase. The particle size distribution

157

determined by the laser diffraction instrument was based on an analysis of the measured angular

158

light scattering pattern using Mie theory. The particle size measurements are reported as either

159

surface area-weighted mean diameter, d3,2 or volume-weighted mean diameter, d4,3.

SC

RI PT

151

A fluorescence confocal laser scanning microscope with a 10× eyepiece and a 60× objective

161

lens (Nikon D-Eclipse C1 80i, Nikon, Melville, NY, USA) were used to acquire the

162

microstructure of the different samples. 1 mL of sample was mixed with 0.1 to 0.2 mL of Nile

163

red (1 mg Nile red/1 mL ethyl alcohol) to stain the oil phase. The excitation and emission

164

spectrum for the Nile red stain were 543 nm and 605 nm, respectively. Digital images were

165

acquired and then analyzed using image analysis software (EZ-CS1, Nikon, Melville, NY, USA).

166

2.7 Electrical potential measurements

TE D

M AN U

160

The zeta-potential (ζ-potential) of the emulsifier-coated oil droplets in the emulsions was

168

determined using an electrophoretic light scattering (Zetasizer Nano ZS, Malvern Instruments

169

Ltd., Worcestershire, UK). All samples were diluted 10- to 40-fold using either 10 mM

170

phosphate buffer pH 7 (for emulsion systems) or distilled water with 270 ppm CaCO3 hardness

171

(for coffee solutions), which were at the same pH as the samples. The electrical potential applied

172

across the electrodes in the measurement cell was 150 V.

173

2.8 Statistical analysis

AC C

EP

167

174

For all systems, two (replicates) samples were freshly prepared, and three (triplicates)

175

measurements were made per sample replicate. A total of 6 measurements, n = 6 was therefore

176

obtained for each system. The mean and standard deviations were then calculated from these

177

four values using Microsoft Excel 2016.

7

ACCEPTED MANUSCRIPT

178

3. Results and discussion

179

3.1 Interfacial characteristics of surfactants The interfacial activity of the two surfactants used was characterized at the MCT oil–water

181

interface by measuring the change in interfacial tension with surfactant concentration. The

182

interfacial tension decreased with increasing surfactant concentration for both surfactants

183

(Figure 2), as more surfactant was available to adsorb to the interface and reduce the interfacial

184

tension (Rosen and Kunjappu 2012). Other studies have also demonstrated the ability of saponins

185

and phospholipids to reduce the interfacial tension (Bai, Huan, Gu and McClements 2016,

186

Tippel, Lehmann, von Klitzing and Drusch 2016). Saponin was more surface active than lecithin,

187

since the decrease in interfacial tension with surfactant concentration occurred more rapidly, and

188

a lower interfacial tension was reached at high surfactant levels for saponin (6 mN/m) than

189

lecithin (10 mN/m). Another recent study reported a similar interfacial tension (6 mN/m) for

190

quillaja saponin saturated interfaces (Tippel, Lehmann, von Klitzing and Drusch 2016). The

191

different interfacial tension reductions achieved by the two surfactants may have been due to

192

differences in the packing of the surfactant molecules at the interface. Presumably, saponins

193

packed more closely together, and therefore screened unfavorable molecular interactions

194

between the oil and water phases more effectively. The fact that a lower interfacial tension was

195

obtained with the saponins, suggests that they should generate smaller droplets during

196

homogenization, since the Laplace pressure (which opposes droplet deformation and disruption)

197

is proportional to interfacial tension.

198

3.2 Effect of surfactant type and content on properties of model creamers

EP

TE D

M AN U

SC

RI PT

180

The impact of surfactant type and concentration on the physiochemical and structural

200

properties of the model creamers (10% MCT oil-in-water emulsions) was then determined. The

201

surfactant levels used in our study were based on previously reported values for emulsions

202

formed using either quillaja saponins (Yang, Leser, Sher and McClements 2013) or lecithins

203

(Komaiko, Sastrosubroto and McClements 2016, Zhu and Damodaran 2013).

204

3.2.1 Appearance and tristimulus color coordinates of model creamers

AC C

199

205

The optical properties of food emulsions make an important contribution to their overall

206

sensory attributes and consumer appeal. For this reason, the impact of surfactant type and

207

concentration on the visual appearance and tristimulus coordinates of the model creamers were

8

ACCEPTED MANUSCRIPT

measured (Figure 3). Model creamers prepared using both surfactants initially had a uniform

209

creamy appearance at all surfactant levels used (Figure 3a). At the same surfactant level, lecithin

210

produced whiter (higher L*) emulsions than saponin (Figure 3b), which may have been because

211

saponin had a dark brown color, whereas soy lecithin only had a slight yellowish color. Overall,

212

the lightness of the emulsions decreased with increasing surfactant level for both surfactants

213

(Figure 3b). For the saponin-stabilized emulsions, the lightness decreased approximately

214

linearly from around 91.2 to 86.0 when the surfactant level increased from 0.5 to 2.5% (R2 =

215

0.99). For the lecithin-stabilized emulsions, the lightness decreased approximately linearly from

216

around 92.7 to 88.4 when the surfactant concentration increased from 1 to 5% (R2 = 0.97). The

217

decrease in lightness in both systems can at least partly be attributed to selective absorption of

218

light waves by the surfactants, thus meaning that less light was reflected from their surfaces

219

(McClements 2002). In addition, the lightness of the model creamers may also have changed due

220

to differences in particle size distribution. The mean particle diameter decreased with increasing

221

surfactant concentration (see section 3.2.2), which may have reduced the intensity of back-

222

scattered light and therefore the lightness. Indeed, it is known that there is a maximum in the

223

lightness of oil-in-water emulsions when the droplet diameter is around 0.2 µm, and that L*

224

decreases when the droplet diameter either increases above or decreases below this value

225

(McClements 2002).

TE D

M AN U

SC

RI PT

208

The positive a* values (green to red coordinate) for the saponin-stabilized emulsions

227

indicated that they had a slight reddish tinge, whereas the negative a* values for the lecithin-

228

stabilized emulsions indicated that they had a slight greenish tinge (Figure 3b). The positive b*

229

values (blue to yellow coordinate) indicate that both types of emulsions had a yellowish tinge,

230

that increased with increasing surfactant level (Figure 3b). These differences in emulsion color

231

can be attributed to differences in the intrinsic colors of the surfactants they contain: quillaja

232

saponin had a dark brownish red color, whereas soy lecithin had a yellowish color. These

233

differences in color should be taken into account when formulating non-dairy creamers since

234

consumers have a preconceived expectation for the appearance of this type of product.

235

3.2.2 Particle size and microstructure of model creamers

AC C

EP

226

236

The mean particle diameter and particle size distribution of the oil droplets in the model

237

creamers were measured to determine the emulsifying efficiency and capacity of the two

9

ACCEPTED MANUSCRIPT

surfactants (Figure 4a). At the same surfactant level, saponin-stabilized emulsions had smaller

239

mean particle diameters than lecithin-stabilized ones, which indicates that the saponins were

240

more effective at producing small droplets during homogenization. There may be a number of

241

reasons for this effect. First, the saponin molecules may have adsorbed more rapidly to the

242

droplet surfaces during homogenization. This may have occurred because the saponins were

243

travelling through the low-viscosity aqueous phase, whereas the lecithins were travelling through

244

the high-viscosity oil phase. Second, the adsorption mechanism of the saponin and lecithin

245

molecules may have been different due to the differences in their molecular environments. Third,

246

the saponins reduced the interfacial tension more effectively than the lecithin (Figure 2), which

247

would decrease the Laplace pressure of the oil droplets and therefore make them easier to disrupt

248

within the homogenizer (Tcholakova, Denkov and Danner 2004). Fourth, the surface load of

249

quillaja saponin (1.0 mg m-2) has been reported to be considerably less than that of soy lecithin

250

(4.4 mg m-2), which means that a greater oil-water interfacial area can be covered at the same

251

surfactant concentration, leading to smaller droplet sizes (Ozturk, Argin, Ozilgen and

252

McClements 2014).

M AN U

SC

RI PT

238

For both systems, increasing the surfactant level led to a decrease in mean particle diameter

254

(d3,2), as well as to changes in the particle size distribution (Figure 4a). The particle size

255

distributions of the quillaja saponin systems changed from monomodal (large droplets) to multi-

256

modal (small and large droplets) and back to monomodal (small droplets) as the surfactant

257

concentration increased, whereas those of the lecithin systems changed from multi-modal (large

258

droplets) to monomodal (small droplets). For creamers, it is usually advantageous to have a

259

narrow particle size distribution so that there is not a population of large droplets that will cream

260

rapidly, and alter the visible appearance of the product. The mean particle diameter decreased

261

from around 0.36 to 0.15 µm when the saponin content increased from 0.5 to 2.5%, and from

262

around 0.37 to 0.17 µm when the lecithin content increased from 1 to 5%. This decrease in mean

263

particle diameter with increasing surfactant level can mainly be attributed to the fact that a

264

greater oil-water interfacial area can be covered, thereby leading to smaller droplet sizes

265

(Dickinson 2010, McClements 2015, Pichot, Spyropoulos and Norton 2010, Tcholakova,

266

Denkov and Danner 2004). In addition, the oil droplet surfaces are covered by surfactant

267

molecules more rapidly at high surfactant levels, thereby facilitating droplet disruption (lower

268

interfacial tension) and inhibiting droplet coalescence (stronger repulsive forces). Other studies

AC C

EP

TE D

253

10

ACCEPTED MANUSCRIPT

have also reported a decrease in droplet size with increasing surfactant concentration for oil-in-

270

water emulsions produced using quillaja saponins (Ozturk, Argin, Ozilgen and McClements

271

2014, Yang, Leser, Sher and McClements 2013, Zhang and Reineccius 2016) and soy lecithins

272

(Bai, Huan, Gu and McClements 2016, Ozturk, Argin, Ozilgen and McClements 2014, Zhu and

273

Damodaran 2013). The droplet diameter versus surfactant concentration profiles obtained in this

274

study (Figure 4a) exhibit the two-regimes reported for high-pressure homogenization: (i) a

275

surfactant-poor regime where the particle size is limited by the amount of surfactant present, and

276

therefore decreases with increasing surfactant level; (ii) a surfactant-rich regime where the

277

particle size is limited by the maximum disruptive energy that can be generated by the

278

homogenizer, and therefore the particle size remains fairly constant with increasing surfactant

279

level (Maindarkar, Hoogland and Henson 2015, Tcholakova, Denkov and Danner 2004).

M AN U

SC

RI PT

269

The fluorescence confocal scanning laser microscopy images of the emulsions supported the

281

light scattering measurements, showing that smaller oil droplets (stained red) were formed in

282

emulsions containing higher amounts of surfactant (Figure 4b). In particular, a number of large

283

individual oil droplets were observed in the emulsions at the lowest surfactant levels used (i.e.,

284

0.5% quillaja saponin and 1% soy lecithin), which indicated that there was insufficient surfactant

285

present to stabilize all the small droplets produced during homogenization and leading to some

286

coalescence inside the homogenizer.

287

3.2.3 Electrical characteristics of oil droplets in model creamers

TE D

280

The electrical characteristics of oil droplets are important because they influence the

289

magnitude of the electrostatic interactions between droplets, and therefore their stability and

290

tendency to aggregate with each other (McClements 2015, Zalazar, Gliemmo and Campos 2016).

291

The electrical properties of oil droplets are also important because they influence their

292

interactions with other charged substances that may be present in an emulsion, such as ionic

293

biopolymers, multivalent ions, or charged surfaces, which may impact the physical and chemical

294

stability of the emulsions. The model creamers fabricated using both types of natural surfactant

295

contained oil droplets had a relatively high negative charge, with the soy lecithins giving an

296

appreciably higher magnitude of the surface potential than the quillaja saponins (Figure 4c). The

297

negative charge on the quillaja saponins can be attributed to the presence of anionic carboxyl

298

groups associated with their polar regions (Gilabert-Oriol, Weng, von Mallinckrodt, Stoshel,

AC C

EP

288

11

ACCEPTED MANUSCRIPT

Nissi, Melzig, Fuchs and Thakur 2015), whereas the negative charge on the soy lecithins can be

300

attributed to the presence of phosphate and carboxyl groups associated with their polar head

301

groups (Wu and Wang 2003). The ζ-potential of the saponin-coated droplets changed from -44

302

to -53 mV as the surfactant level was increased from 0.5 to 2.5%, whereas that of the lecithin-

303

coated droplets changed from -72 to -66 mV when the surfactant level was increased from 1 to

304

5% (Figure 4c). These results suggest that there may have been some change in the

305

concentration or packing of the surfactant molecules at the droplet surfaces at varying surfactant

306

levels. In the case of the lecithin, there may have been some ionic impurities in the ingredient

307

that led to electrostatic screening effects at high emulsifier levels, thereby reducing the

308

magnitude of the ζ-potential. A recent study has also shown the influence of oil droplet size on

309

the ζ-potential of dairy creamer (Hussain, Truong, Bansal and Bhandari 2016). A number of

310

other studies have also reported a high negative charge on soy lecithin-coated and quillaja

311

saponin-coated oil droplets at neutral pH (Bai, Huan, Gu and McClements 2016, Maier, Zeeb

312

and Weiss 2014, Ozturk, Argin, Ozilgen and McClements 2014).

313

3.3 Application of model creamers in coffee

M AN U

SC

RI PT

299

Non-dairy creamers are added to coffee drinks for a number of reasons: to whiten them

315

(Oldfield, Teehan and Kelly 2000), to partially neutralize their acidity (Golde and Schmidt

316

2005), and/or to create a desirable flavor profile or mouthfeel (Khatkar and Gupta 2014). Under

317

certain circumstances, the addition of non-dairy creamers to hot coffee solutions can lead to

318

undesirable quality attributes such as “oiling-off”, “feathering” or “sedimentation”. Oiling-off

319

(oil layer at the top), feathering (white flakes at the top) or sedimentation (precipitates at the

320

bottom) have been attributed to aggregation of oil droplets and/or protein molecules as a result of

321

changes in temperature, ionic strength, or pH when the creamer is added to hot acidic coffee

322

solutions (Burgwald 1923, Golde and Schmidt 2005, Oldfield, Teehan and Kelly 2000). The

323

behavior of the model creamers was therefore examined after they were added to model hot

324

coffee solutions.

AC C

EP

TE D

314

325

An aliquot (10 mL) of creamer (pH 7, 25°C) was added to a fixed quantity (60 g) of hot

326

black coffee (pH 4.9, 85 °C). The black coffee solution was prepared by adding 1% coffee

327

powder to boiled distilled water containing 270 ppm calcium carbonate hardness to mimic the

12

ACCEPTED MANUSCRIPT

328

hardness of water found in some local water sources (Oldfield, Teehan and Kelly 2000). The

329

mixture resulting from the addition of creamer to black coffee was referred to as white coffee.

330

3.3.1 Effect of model creamer addition on appearance Visual observation of the white coffees indicated that there was no visible oiling-off,

332

feathering, or sedimentation, with the exception of the model creamers with the lowest surfactant

333

concentrations (Figure 5a). A layer of free oil was observed on the surface of the white coffee

334

prepared using the creamer with 0.5% quillaja saponins, which could be due to the fact that it

335

contained relatively large droplets that were therefore highly prone to coalescence. In addition,

336

there was a distinct whitish layer on top of the white coffee prepared using the creamer with 1%

337

soy lecithin, which was probably due to creaming of the relatively large oil droplets in this

338

creamer.

339

indicating that the oil droplets remained stable to aggregation and gravitational separation when

340

added to the hot coffee. These results suggest that using greater than 0.5% quillaja saponin or 1%

341

soy lecithin is required to form stable 10% oil-in-water emulsions suitable for application as non-

342

dairy creamers for application in hot coffee.

SC

RI PT

331

M AN U

In all the other systems, no free oil, feathering or sedimentation was observed,

The tristimulus color coordinates of the white coffees were also measured to provide more

344

quantitative information about the influence of the creamers on their appearance. As expected,

345

there was a large increase in the lightness of the systems when model creamers were added to the

346

hot black coffee solutions, with the L* value increasing from around 5.4 for black coffee to

347

around 50 for white coffee (Figure 5b). The lightness of the white coffees did not depend

348

strongly on the amount of surfactant that was present in the model creamers, being around 51 to

349

52 for quillaja saponin and increasing from around 50 to 53 for soy lecithin. The higher lightness

350

of the white coffee compared to the black coffee is due to light scattering by the emulsion

351

droplets (McClements 2002). The lightness of a white coffee prepared by adding a commercial

352

coffee creamer to the same black coffee was around 52 (data not shown), which demonstrated

353

that the model creamers prepared in this study had a similar whitening power as their

354

commercial counterparts.

AC C

EP

TE D

343

355

The a* values of the samples increased from around +3 for black coffee to +8 for white

356

coffee containing creamers prepared with the lowest surfactant content, which suggested that the

357

white coffees had a slight reddish tinge (Figure 5b). The redness of the white coffees decreased 13

ACCEPTED MANUSCRIPT

as the surfactant content in the model creamers increased for both types of natural surfactant.

359

This effect cannot be attributed to the intrinsic color of the surfactants themselves because one

360

had a reddish color and the other a greenish color (compare Figure 3b and Figure 5b), which is

361

probably because the final concentration of surfactant in the white coffee was relatively low.

362

Instead, this effect is probably due to the change in light scattering efficiency associated with the

363

decrease in oil droplet size with increasing surfactant concentration (Section 3.2.1). Similar

364

trends were also obtained for the yellowish color of the coffee, where the b* value increased

365

from +3 for black coffee to +27 after addition of both model creamers stabilized with the lowest

366

surfactant content (Figure 5b). The b* value also decreased with increasing surfactant content in

367

the model creamer, which may again be due to changes in oil droplet size.

368

3.3.2 Particle size and microstructure of white coffee

M AN U

SC

RI PT

358

For most of the systems, similar mean particle size diameters (Figure 6a) and particle size

370

distributions (data not shown) were measured for the white coffee and corresponding model

371

creamers (Section 3.1). This result indicates that the surfactant-coated droplets in the creamers

372

were stable after addition to acidic hot coffee (pH 4.9, 85°C). The only exceptions were the

373

model creamers containing the lowest surfactant concentrations: a thin oil layer was observed on

374

top of the white coffee formed from 0.5% quillaja saponin creamer (indicative of coalescence),

375

whereas a thin white layer was observed at the top of the white coffee formed from 1% soy

376

lecithin creamer (indicative of creaming). Nevertheless, this visible instability was not reflected

377

in the particle sizes determined by light scattering, which suggests that only a small fraction of

378

the droplets in the creamers were prone to aggregation and gravitational separation. Similar

379

microstructures were also observed in the white coffees (Figure 6b) as in the corresponding

380

model creamers (Figure 4b), which further suggests that the oil droplets remained stable after

381

addition to the hot coffee solutions.

382

3.3.3 Electrical characteristics of oil droplets in white coffee

AC C

EP

TE D

369

383

The electrical characteristics (ζ-potential) of the particles in the coffee systems was

384

measured using electrophoresis (Figure 6c). The black coffee had a ζ-potential of around -7.5

385

mV, which suggested that it contained some polymers or particles that had a slight negative

386

charge. The surface potential of the particles in the white coffee was more negative than that in

387

the black coffee, which can be attributed to the fact that the surfactant-coated droplets in the

14

ACCEPTED MANUSCRIPT

model creamers were highly negatively charged (Section 3.2.3). However, the electrical potential

389

of the particles in the white coffee was much less negative than that in the corresponding

390

creamers (Figure 4c). This change in ζ-potential can be attributed to various effects: (i) the

391

negative charge on carboxylic acid groups usually decreases with decreasing pH, and may

392

therefore have been reduced in acidic coffee solutions; (ii) the negative charge on the anionic oil

393

droplets may have been reduced due to adsorption of cationic calcium ions (Ca2+) arising from

394

the calcium ions used to simulate the hardness of typical tap water; (iii) there may have been

395

other charged components within the coffee solution that contributed to the electrical signal used

396

to calculate the ζ-potential.

SC

RI PT

388

The magnitude of the ζ-potential on the oil droplets in the white coffee depended on the type

398

and amount of surfactant used to stabilize them (Figure 6c), and followed a similar trend to that

399

observed in the creamers themselves (Figure 4c). In particular, the lecithin-coated droplets were

400

more negative than the saponin-coated ones, and the ζ-potential became more negative with

401

increasing surfactant level for the saponins, but less negative for the lecithin. Again, these

402

results suggest that the composition, structure of the surfactant molecules at the oil droplet

403

surfaces or the droplet size may have changed with increasing surfactant concentration.

404

3.3.4 Effect of model creamer content on white coffee properties

TE D

M AN U

397

Consumers have different preferences for the amount of creamer they add to their coffee,

406

and so the impact of creamer level on white coffee properties was studied. In these experiments,

407

only model creamers containing either 1.5% quillaja saponin or 5% soy lecithin were studied

408

because these levels of surfactant formed stable emulsions with high lightness when added in

409

black coffee (Figure 5b). Different amounts of the model creamers (5, 10, 20 or 30 mL) were

410

then added to the hot black coffee solution (60 g), and the change in white coffee properties was

411

measured.

AC C

EP

405

412

As expected, increasing the total amount of model creamer added produced whiter coffee,

413

which was observed visually (Figure 7a) and indicated by the instrumental lightness values

414

(Figure 7b). This effect can be attributed to the increase in the fraction of light back-scattered

415

from the emulsions at higher droplet concentrations (McClements 2002). No feathering,

416

sedimentation or oiling off was observed in any of these white coffee drinks, indicating that the

15

ACCEPTED MANUSCRIPT

model creamers were stable. Increasing the amount of model creamer present decreased the

418

redness and yellowness of the white coffee drinks (Figure 7b), which can be attributed to color

419

fading due to greater light scattering by the higher level of oil droplets present (McClements

420

2002).

RI PT

417

The mean particle diameter, particle size distribution, and microstructure of the white

422

coffees did not depend on creamer level (data not shown), and were similar to those of the

423

corresponding creamer systems. This further suggests that the model creamers were stable when

424

added to coffee, regardless of the level used.

SC

421

In addition to their ability to increase the desirable creamy appearance of coffee, non-dairy

426

creamers are also used to neutralize the acidity of coffee drinks. For this reason, the impact of

427

creamer level on the pH of the white coffee drinks was measured. The pH of the coffee

428

increased steadily from around pH 4.9 in black coffee to around pH 6.0 in coffee containing 30

429

mL model creamer (Figure 8). This result highlights the ability of the model creamers to

430

neutralize acidity of the coffee drinks. The creamer containing lecithin-coated droplets was

431

slightly more effective at neutralizing the acidity of the coffee than the one containing saponin-

432

coated droplets.

433

characteristics of the particles in the coffee drinks, with the ζ-potential increasing from around -

434

7.5 mV for black coffee to around -20 mV with 30 mL model creamer (Figure 8). (At similar

435

emulsifier and creamer compositions, the results are in agreement with those shown in Figure

436

6c). The increasing ζ-potential can be attributed to the fact that the pH increased and the calcium

437

ion concentration decreased as the creamer level was increased, which would lead to more

438

negative droplets (Section 3.2.3).

439

4. Conclusion

M AN U

425

AC C

EP

TE D

Increasing the amount of model creamer also altered the electrical

440

This study demonstrated the prospective use of quillaja saponin and soy lecithin as

441

surfactants for formulating liquid coffee whiteners. Both types of surfactant were able to reduce

442

the interfacial tension at the oil–water interface and form stable oil-in-water emulsions with

443

small droplet sizes and high whitening capacity. Increasing the surfactant content decreased the

444

whiteness of the emulsions, which was attributed to the intrinsic color of the surfactants used

445

(absorption effects) and a decrease in droplet size (scattering effects). Quillaja saponin formed 16

ACCEPTED MANUSCRIPT

dark brown solutions in water, whereas soy lecithin formed yellow solutions in oil. Increasing

447

the surfactant content decreased the droplet size because more surfactant was available to cover

448

the droplet surfaces formed during homogenization. Both types of natural surfactant had anionic

449

functional groups, which led to oil droplets with high negative charges at neutral pH. The model

450

creamers were added to hot coffee prepared with hard water to determine their stability in a hot

451

acidic and high mineral content environment. The oil droplets remained stable to aggregation

452

after they were added to the coffee solutions, with no visible evidence of oiling-off, feathering,

453

or sedimentation. Increasing the level of creamer added to black coffee led to an increase in the

454

whiteness and decrease in the acidity of the resulting white coffee. Overall, this study

455

demonstrated the potential of two natural surfactants (saponins and phospholipids) to form

456

emulsions that are suitable for applications in liquid coffee whiteners that are stable as is and in

457

hot acidic, high mineral content environments.

458

Acknowledgements

SC

M AN U

459

RI PT

446

The authors would like to thank Nestlé Development Center Marysville (OH, USA) for the funding provided to support this study. There is no conflict of interest with any funding agencies

461

during the course of study.

AC C

EP

TE D

460

17

ACCEPTED MANUSCRIPT

References

463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506

Augustin JM, Kuzina V, Andersen SB, Bak S. 2011. Molecular activities, biosynthesis and evolution of triterpenoid saponins. Phytochemistry 72: 435-57 Bai L, Huan S, Gu J, McClements DJ. 2016. Fabrication of oil-in-water nanoemulsions by dualchannel microfluidization using natural emulsifiers: Saponins, phospholipids, proteins, and polysaccharides. Food Hydrocolloids 61: 703-11 Bai L, Huan SQ, Gu JY, McClements DJ. 2016. Fabrication of oil-in-water nanoemulsions by dual-channel microfluidization using natural emulsifiers: Saponins, phospholipids, proteins, and polysaccharides. Food Hydrocolloids 61: 703-11 Baines D, Seal R. 2012. Natural food additives, ingredients and flavourings Introduction. In Natural Food Additives, Ingredients and Flavourings, ed. D Baines, R Seal, pp. XXIIIXXVI Burgwald L. 1923. Some factors which influence the feathering of cream in coffee. Jour. Agr. Res 26 Cabezas D, Diehl BW, Tomás MC. 2016. Emulsifying properties of hydrolysed and low HLB sunflower lecithin mixtures. European Journal of Lipid Science and Technology 118: 975 - 83 Cabezas D, Madoery R, Diehl B, Tomás M. 2012. Emulsifying properties of different modified sunflower lecithins. Journal of the American Oil Chemists' Society 89: 355-61 Cui L, Decker EA. 2016. Phospholipids in foods: prooxidants or antioxidants? Journal of the science of food and agriculture 96: 18-31 Dickinson E. 2009. Hydrocolloids as emulsifiers and emulsion stabilizers. Food Hydrocolloids 23: 1473-82 Dickinson E. 2010. Food emulsions and foams: stabilization by particles. Current Opinion in Colloid & Interface Science 15: 40-49 Dickinson E, Leser ME. 2013. Food colloids today... understanding structural change during processing, storage, eating and digestion. Current opinion in colloid & interface science 18: 245-48 FoodNavigator. 2013. What do 'natural' and 'clean label' mean anyway? , ed. C Scott-Thomas. http://www.foodnavigator.com/Market-Trends/What-do-natural-and-clean-label-meananyway?utm_source=copyright&utm_medium=OnSite&utm_campaign=copyright FoodNavigator. 2016. Clean-label claims sought by one in three shoppers, according to survey. ed. J Cornall. http://www.foodnavigator.com/Trends/Natural-and-clean-label/Cleanlabel-claims-sought-by-one-in-three-shoppers-survey-finds Gilabert-Oriol R, Weng A, von Mallinckrodt B, Stoshel A, Nissi L, et al. 2015. Electrophoretic mobility as a tool to separate immune adjuvant saponins from Quillaja saponaria Molina. International Journal of Pharmaceutics 487: 39-48 Golde A, Schmidt K. 2005. Quality of coffee creamers as a function of protein source. Journal of food quality 28: 46-61 Golemanov K, Tcholakova S, Denkov N, Pelan E, Stoyanov SD. 2014. The role of the hydrophobic phase in the unique rheological properties of saponin adsorption layers. Soft matter 10: 7034-44 Guiotto EN, Cabezas DM, Diehl BW, Tomás MC. 2013. Characterization and emulsifying properties of different sunflower phosphatidylcholine enriched fractions. European Journal of Lipid Science and Technology 115: 865-73

AC C

EP

TE D

M AN U

SC

RI PT

462

18

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

Guiotto EN, Capitani MI, Nolasco SM, Tomás MC. 2016. Stability of Oil-in-Water Emulsions with Sunflower (Helianthus annuus L.) and Chia (Salvia hispanica L.) By-Products. Journal of the American Oil Chemists' Society 93: 133-43 Hussain H, Truong T, Bansal N, Bhandari B. 2016. The Effect of Manipulating Fat Globule Size on the Stability and Rheological Properties of Dairy Creams. Food Biophysics: 1-10 Khatkar SK, Gupta VK. 2014. Physicochemical and Functional Quality Attributes of Dairy Whitener Prepared from Ultrafiltration Process. Journal of Food Processing and Preservation 38: 1145-54 Komaiko J, Sastrosubroto A, McClements DJ. 2015. Formation of oil-in-water emulsions from natural emulsifiers using spontaneous emulsification: Sunflower phospholipids. Journal of agricultural and food chemistry 63: 10078-88 Komaiko J, Sastrosubroto A, McClements DJ. 2016. Encapsulation of ω-3 fatty acids in nanoemulsion-based delivery systems fabricated from natural emulsifiers: Sunflower phospholipids. Food chemistry 203: 331-39 Küllenberg D, Taylor LA, Schneider M, Massing U. 2012. Health effects of dietary phospholipids. Lipids in health and disease 11: 1-16 Liu L, Waters DL, Rose TJ, Bao J, King GJ. 2013. Phospholipids in rice: significance in grain quality and health benefits: a review. Food chemistry 139: 1133-45 Maier C, Zeeb B, Weiss J. 2014. Investigations into aggregate formation with oppositely charged oil-in-water emulsions at different pH values. Colloids and Surfaces B: Biointerfaces 117: 368-75 Maindarkar SN, Hoogland H, Henson MA. 2015. Predicting the combined effects of oil and surfactant concentrations on the drop size distributions of homogenized emulsions. Colloids and Surfaces A: Physicochemical and Engineering Aspects 467: 18-30 McClements DJ. 2002. Theoretical prediction of emulsion color. Advances in Colloid and Interface Science 97: 63-89 McClements DJ. 2010. Emulsion design to improve the delivery of functional lipophilic components. Annual review of food science and technology 1: 241-69 McClements DJ. 2015. Food emulsions: principles, practices, and techniques: CRC press McClements DJ, Gumus CE. 2016. Natural emulsifiers - Biosurfactants, phospholipids, biopolymers, and colloidal particles: Molecular and physicochemical basis of functional performance. Advances in Colloid and Interface Science 234: 3-26 Mitra S, Dungan SR. 1997. Micellar properties of Quillaja saponin. 1. Effects of temperature, salt, and pH on solution properties. Journal of Agricultural and Food Chemistry 45: 1587-95 Oldfield D, Teehan C, Kelly P. 2000. The effect of preheat treatment and other process parameters on the coffee stability of instant whole milk powder. International dairy journal 10: 659-67 Orczyk M, Wojciechowski K. 2015. Comparison of the effect of two Quillaja bark saponin extracts on DPPC and DPPC/cholesterol Langmuir monolayers. Colloids and Surfaces B: Biointerfaces 136: 291-99 Ozturk B, Argin S, Ozilgen M, McClements DJ. 2014. Formation and stabilization of nanoemulsion-based vitamin E delivery systems using natural surfactants: Quillaja saponin and lecithin. Journal of Food Engineering 142: 57-63 Petrut RF, Danthine S, Blecker C. 2016. Assessment of partial coalescence in whippable oil-inwater food emulsions. Advances in colloid and interface science 229: 25-33

AC C

507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552

19

ACCEPTED MANUSCRIPT

RI PT

SC

M AN U

TE D

EP

581

Pichot R, Spyropoulos F, Norton I. 2010. O/W emulsions stabilised by both low molecular weight surfactants and colloidal particles: The effect of surfactant type and concentration. Journal of colloid and interface science 352: 128-35 Pichot R, Watson RL, Norton IT. 2013. Phospholipids at the interface: current trends and challenges. International journal of molecular sciences 14: 11767-94 Rosen MJ, Kunjappu JT. 2012. Surfactants and interfacial phenomena: John Wiley & Sons Stanimirova R, Marinova K, Tcholakova S, Denkov N, Stoyanov S, Pelan E. 2011. Surface rheology of saponin adsorption layers. Langmuir 27: 12486-98 Tcholakova S, Denkov ND, Danner T. 2004. Role of surfactant type and concentration for the mean drop size during emulsification in turbulent flow. Langmuir 20: 7444-58 Tippel J, Lehmann M, von Klitzing R, Drusch S. 2016. Interfacial properties of Quillaja saponins and its use for micellisation of lutein esters. Food Chemistry 212: 35-42 Wu Y, Wang T. 2003. Soybean lecithin fractionation and functionality. Journal of the American Oil Chemists' Society 80: 319-26 Yang Y, Leser ME, Sher AA, McClements DJ. 2013. Formation and stability of emulsions using a natural small molecule surfactant: Quillaja saponin (Q-Naturale®). Food Hydrocolloids 30: 589-96 Zalazar AL, Gliemmo MF, Campos CA. 2016. Effect of stabilizers, oil level and structure on the growth of Zygosaccharomyces bailii and on physical stability of model systems simulating acid sauces. Food Research International 85: 200-08 Zhang J, Bing L, Reineccius GA. 2016. Comparison of modified starch and Quillaja saponins in the formation and stabilization of flavor nanoemulsions. Food chemistry 192: 53-59 Zhang J, Reineccius GA. 2016. Factors controlling the turbidity of submicron emulsions stabilized by food biopolymers and natural surfactant. LWT-Food Science and Technology 71: 162-68 Zhu D, Damodaran S. 2013. Dairy Lecithin from Cheese Whey Fat Globule Membrane: Its Extraction, Composition, Oxidative Stability, and Emulsifying Properties. Journal of the American Oil Chemists' Society 90: 217-24

AC C

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

20

RI PT

ACCEPTED MANUSCRIPT

Lecithin

AC C

EP

TE D

M AN U

SC

Quillaja Saponin

N+ OO

P

O

O O O O O

e 1: Detailed structural representation of quillaja saponin and general structural representation of lecithin. The quillaja saponin structure is based on that presented by Mitra and an (1997) and the lecithin (phosphatidylcholine) was a general structure drawn using a software (ChemDraw®).

ACCEPTED MANUSCRIPT

RI PT

25

Soy Lecithin

SC

Quillaja Saponin

M AN U

15

TE D

10

EP

5

0 0

AC C

Interfacial Tension mN/m

20

0.02

0.04 0.06 0.08 0.1 Surfactant Content (%)

0.12

Figure 2: Effect of surfactant content on the interfacial tension at MCT– aqueous interface. The surfactant content was calculated based on their active components.

a) 1%

Quillaja Saponin 1.5% 2%

2.5%

1%

2%

Soy Lecithin 3% 4%

5%

AC C

EP

TE D

M AN U

SC

0.5%

RI PT

ACCEPTED MANUSCRIPT

Figure 3a: Effect of emulsifier content (Quillaja saponin: 0.5 to 2.5%; soy lecithin 1 to 5%) on the visual appearance of emulsions (10% medium chain triglyceride) after 36 hours storage.

ACCEPTED MANUSCRIPT

1.5

Quillaja Saponin

Quillaja Saponin Soy Lecithin

M AN U

0.5 a*

0 -0.5

-2 4

Emulsifier Content (%)

5

AC C

3

0

1

2 3 4 Emulsifier Content (%)

7

Quillaja Saponin

5

3 2 1 0

-2.5 2

Soy Lecithin

4

EP

-1.5

8

6

TE D

-1

1

9

SC

1

10

b*

Soy Lecithin

RI PT

2

5

0

1

2

3

4

Emulsifier Content (%

Figure 3b: Effect of emulsifier content (Quillaja saponin: 0.5 to 2.5%; soy lecithin 1 to 5%) on the tristimulus color coordinate of emulsions (10% medium chain triglyceride)

ACCEPTED MANUSCRIPT

RI PT

Quillaja Saponin Emulsions

70

Quillaja Saponin

2.5% 2.0% 1.5% 1.0% 0.5%

M AN U

40 30

TE D

Volume (%)

50

1 2 3 4 Emulsifier Content (%)

5

AC C

0

0 0.03

0.3 Particle Size (µ µm)

Soy Lecithin Emulsions

5% Solec 4% Solec 3% Solec 2% Solec 1% Solec

40 30 20 10

EP

20 10

50

SC

60

Volume (%)

Soy Lecithin

60

3

0 0.03

0.3 3 Particle Size (µ µm)

re 4a: Effect of emulsifier content (Quillaja saponin: 0.5 to 2.5%; soy lecithin 1 to 5%) on the particle size diameter and distributions of emulsions (10% medium chain triglyce

1.5% Quillaja Saponin

2.5% Quillaja Sapo

M AN U

SC

RI PT

2% Quillaja Saponin

3% Soy Lecithin

TE D

2% Soy Lecithin

4% Soy Lecithin

5% Soy Lecithin

EP

1% Soy Lecithin

1% Quillaja Saponin

AC C

% Quillaja Saponin

ACCEPTED MANUSCRIPT

e 4b: Effect of emulsifier content (Quillaja saponin: 0.5 to 2.5%; soy lecithin 1 to 5%) on the microstructure of emulsions (10% medium chain triglyceride). Confocal microgra obtained at 60x magnification with scale bar of 20 µm shown on bottom right. Oil droplets were stained with Nile Red (0.1%).

ACCEPTED MANUSCRIPT

RI PT

-40

c)

Quillaja Saponin

SC

Soy Lecithin

M AN U

-50 -55 -60 -65

TE D

Zeta Potential (mV)

-45

EP

-70

AC C

-75 0

2

4

Emulsifier Content (%)

Figure 4c: Effect of emulsifier content (Quillaja saponin: 0.5 to 2.5%; soy lecithin 1 to 5%) on the zeta potential of emulsions (10% medium chain triglyceride).

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

5a: Effect of emulsifier type and content (Quillaja saponin: 0.5 to 2.5% or soy lecithin 1 to 5% with 10% medium chain triglyceride) on the visual appearance of white coffee. coffee was made with 1% black coffee reconstituted in 270 ppm CaCO3 hard water (60 mL) followed by addition of 10 mL of model creamer at 85ºC.

ACCEPTED MANUSCRIPT

10

RI PT

30

9 8

SC

7

2 1 0

2

3

Emulsifier Content (%)

4

5

0

1

AC C

1

EP

Soy Lecithin

TE D

3

b*

5 4

Quillaja Saponin

20

M AN U

a*

6

25

15 10

Quillaja Saponi

Quillaja Saponin

Soy Lecithin

5 Soy Lecithin

0 2

3

4

Emulsifier Content (%)

5

0

1

2 3 4 Emulsifier Content (%)

5b: Effect of emulsifier type and content (Quillaja saponin: 0.5 to 2.5% or soy lecithin 1 to 5% with 10% medium chain triglyceride) on the tristimulus color coordinates of wh White coffee was made with 1% black coffee reconstituted in 270 ppm CaCO3 hard water (60 mL) followed by addition of 10 mL of emulsions at 85ºC. Note: 0 emulsifier con presents black coffee without model creamer addition.

RI PT

ACCEPTED MANUSCRIPT

0.4

SC

Soy Lecithin

Quillaja Saponin

M AN U

0.2

TE D

D3,2 (µm)

0.3

AC C

EP

0.1

0

0

1

2 3 4 Emulsifier Content (%)

5

6a: Effect of emulsifier type and content (Quillaja saponin: 0.5 to 2.5% or soy lecithin 1 to 5% with 10% medium chain triglyceride) on the averaged particle diameter of whit White coffee was made with 1% black coffee reconstituted in 270 ppm CaCO3 hard water (60 mL) followed by addition of 10 mL of emulsions at 85ºC.

2% Soy Lecithin

3% Soy Lecithin

2% Quillaja Saponin

2.5% Quillaja Sapo

4% Soy Lecithin

5% Soy Lecithin

TE D

M AN U

SC

RI PT

1.5% Quillaja Saponin

EP

1% Soy Lecithin

1% Quillaja Saponin

AC C

% Quillaja Saponin

ACCEPTED MANUSCRIPT

e 6b: Effect of emulsifier type and content (Quillaja saponin: 0.5 to 2.5% or soy lecithin 1 to 5% with 10% medium chain triglyceride) on the microstructure of white coffee. W was made with 1% black coffee reconstituted in 270 ppm CaCO3 hard water (60 mL) followed by addition of 10 mL of emulsions at 85ºC. Confocal micrographs were obtaine magnification with scale bar of 20 µm shown on bottom right. Oil droplets were stained with Nile Red (0.1%).

c)

RI PT

ACCEPTED MANUSCRIPT

0

Quillaja Saponin

M AN U

-10

TE D

-15

EP

-20

AC C

Zeta Potential (mV)

Black Coffee = -7.5

SC

Soy Lecithin

-5

-25

0

1

2 3 4 Emulsifier Content (%)

5

6c: Effect of emulsifier type and content (Quillaja saponin: 0.5 to 2.5% or soy lecithin 1 to 5% with 10% medium chain triglyceride) on the zeta potential of white coffee. Whi was made with 1% black coffee reconstituted in 270 ppm CaCO3 hard water (60 mL) followed by addition of 10 mL of emulsions at 85ºC. Note: 0 emulsifier content (%) ents black coffee without model creamer addition.

ACCEPTED MANUSCRIPT

a)

AC C

EP

TE D

M AN U

SC

Coffee

Soy Lecithin Model Creamer 5 mL 10 mL 20 mL 30 mL

RI PT

Quillaja Saponin Model Creamer Coffee 5 mL 10 mL 20 mL 30 mL

re 7a: Effect of model creamer content (0, 5, 10, 20 or 30 mL) on the visual appearance of white coffee. White coffee was made with 1% black coffee reconstituted in 270 ppm O3 hard water (60 mL) followed by addition of varying amount of model creamer at 85ºC.

RI PT

ACCEPTED MANUSCRIPT

10 9

SC

8

5 4

Quillaja Saponin

3

Soy Lecithin

15

TE D

Quillaja Saponin

b*

a*

6

Soy Lecithin 5

EP

2

AC C 0

30

Quillaja Saponin

10

Soy Lecithin

1 10 20 Creamer Content (mL)

25

20

M AN U

7

30

0

0 10 20 Creamer Content (mL)

30

0

10 20 Creamer Content (mL)

7b: Effect of model creamer content (0, 5, 10, 20 or 30 mL) on the tristimulus color coordinate of white coffee. White coffee was made with 1% black coffee reconstituted in 2 hard water (60 mL) followed by addition of varying amount of model creamer at 85ºC. Note: 0 creamer content (mL) represents black coffee without model creamer addition.

ACCEPTED MANUSCRIPT

6.5

0 Soy Lecithin: pH Quillaja Saponin: pH Soy Lecithin: Zeta Potential Quillaja Saponin: Zeta Potential

RI PT

6.3

SC

5.9

-10

M AN U

5.7 5.5 5.3

-15

TE D

5.1 4.9

-20

EP

4.7 4.5

AC C

pH

-5

0

10 20 Creamer Content (mL)

Zeta Potential (mV)

6.1

-25 30

re 8: Effect of model creamer content (0, 5, 10, 20 or 30 mL) on the pH and zeta potential of white coffee. White coffee was made with 1% black coffee reconstituted in 270 pp O3 hard water (60 mL) followed by addition of varying amounts of model creamer at 85ºC. Model creamers containing 1.5% quillaja saponin or 5% soy lecithin were used. Not mer content (mL) represents black coffee without model creamer addition.

ACCEPTED MANUSCRIPT

Highlights “Formulation of Food Emulsions using Natural Emulsifiers: Utilization of Quillaja Saponin and Soy Lecithin to Fabricate Liquid Coffee Whiteners” by Chung et al.

•

RI PT

Journal of Food Engineering

The impact of two natural emulsifiers on the formation and properties of coffee creamers was compared

Model creamers were prepared using quillaja saponin and soy lecithin as emulsifiers

•

Saponin- and lecithin-stabilized creamers had comparable performance as commercial

SC

•

EP

TE D

The creamers were stable to aggregation and separation when added to acidic hot coffee.

AC C

•

M AN U

creamers.