Spray dried flaxseed oil powdered microcapsules obtained using milk whey proteins-alginate double layer emulsions

Accepted Manuscript Spray dried flaxseed oil powdered microcapsules obtained using milk whey proteins-alginate double layer emulsions

Silvana A. Fioramonti, Evelyn M. Stepanic, Ariadna M. Tibaldo, Yanina L. Pavón, Liliana G. Santiago PII: DOI: Reference:

S0963-9969(18)30872-X doi:10.1016/j.foodres.2018.10.079 FRIN 8049

To appear in:

Food Research International

Received date: Revised date: Accepted date:

25 April 2018 6 September 2018 28 October 2018

Please cite this article as: Silvana A. Fioramonti, Evelyn M. Stepanic, Ariadna M. Tibaldo, Yanina L. Pavón, Liliana G. Santiago , Spray dried flaxseed oil powdered microcapsules obtained using milk whey proteins-alginate double layer emulsions. Frin (2018), doi:10.1016/j.foodres.2018.10.079

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

Spray dried flaxseed oil powdered microcapsules obtained using milk whey

CR

IP

T

proteins-alginate double layer emulsions

Silvana A. Fioramonti, Evelyn M. Stepanic, Ariadna M. Tibaldo, Yanina L. Pavón,

AN

US

Liliana G. Santiago

Instituto de Tecnología de Alimentos, Facultad de Ingeniería Química, Universidad Nacional del

ED

M

Litoral, 1 de Mayo 3250, Santa Fe, Argentina.

PT

*Corresponding author. Tel.: +54 342 4571252 ext. 2588

AC

CE

E-mail: [email protected]

ACCEPTED MANUSCRIPT

ABSTRACT

Spray-dried flaxseed oil microcapsules were produced by designing O/W double-layer emulsions (WPC) and sodium alginate (SA). The influence of

T

using a whey protein concentrate

IP

homogenization pressure (5-15 MPa), pH (4-7) and maltodextrin concentration (0.8-7 wt%) on

CR

stability of primary and secondary emulsions was investigated, through droplet size and zeta potential measurements. Powders obtained after spray drying were characterized through scanning

US

electron microscopy, encapsulation efficiency and water activity determinations. Flaxseed oil

AN

oxidative stability was assessed by measuring peroxide values (PV) and thiobarbituric reactive substances (TBARS) at different microencapsulation processing steps and during powders storage.

M

Droplet sizes of primary emulsions were reduced when increasing homogenization pressures (up to

ED

10 MPa). Zeta potential measurements evidenced double WPC-SA layer formation around oil droplets at pH 5. Encapsulation efficiencies up to 84% were obtained in powdered microcapsules

PT

with the highest MD content. Microencapsulation process produced a gradual increment on PV,

CE

whereas TBARs slightly increased. Nevertheless, these values were maintained relatively constant after powders storage at -18 and 4°C for 6 months, and at 20°C up to 6 weeks and PV did not

AC

exceed the maximum allowed for cold pressed oils.

KEYWORDS: Microencapsulation - Flaxseed Oil – Spray Drying – Whey Protein Concentrate – Alginate – Oxidative Stability

ACCEPTED MANUSCRIPT

1. INTRODUCTION

Changing lifestyles have led to a shift in dietary patterns that has notably had an impact on public

T

health (Kearney, 2010). Modern western diets contain excessive levels of -6 polyunsaturated fatty

IP

acids (PUFAs) and very low levels of -3 PUFAs, thus leading to an unbalanced highly

CR

proinflammatory -6/-3 ratio which could contribute to the prevalence of heart disease, arthritis,

US

cancer and autoimmune disorders (Simopoulos, 2016). Hence, there is a growing interest in the food industry to enrich products with -3 PUFAs. Flaxseed oil is a good vegetable -3 source in

AN

nature, since it contains high levels of α-linolenic acid (50–60% of total fatty acids) (Fioramonti et

M

al., 2015a). However, lipophilic nature and high susceptibility to oxidation rates make this nutraceutical difficult to be incorporated into food matrices. These disadvantages could be

ED

overcome using microencapsulation techniques, such as emulsification combining different wall

PT

materials to protect the oil inside the core (McClements et al., 2007). Milk whey proteins are very versatile emulsifiers because of their amphipathic properties, and they also have high nutritional

CE

values. Whey protein concentrates (WPC) contain up to 80% protein, which primarily consists of β-

AC

lactoglobulin (pI=4.7-5.2) and α-lactalbumin (pI=4.2-4.5), whereas whey protein isolates (WPI) have more than 90% protein (El-Salam et al., 2009). As proteins behave as ampholytes, their surface charge can be modified with pH variation of the aqueous medium. Alginates are linear binary polysaccharides comprised of β-D-mannuronic and α-L-guluronic acid monomers with dissociation constants of 3.38 and 3.65, respectively, so they behave as polyelectrolytes with strong negative charges in a wide range of pH due to dissociation of carboxyl groups (Zhong et al., 2010). Maltodextrins (MD) are neutral polysaccharides also widely used to stabilize emulsions, which

ACCEPTED MANUSCRIPT consist in partial hydrolysis products of starch of variable length. Although they do not have emulsifying properties, MD are often added to emulsified systems to increase wall solution solids concentration to promote better crust formation around the drying droplets during spray-drying (Gharsallaoui et al., 2007).

Regarding flaxseed oil encapsulation, a

IP

fatty acids and incorporating them into food products.

T

Conventional single-layer emulsions have been used as a delivery system for encapsulating -3

CR

number of studies attempted to formulate conventional single-layer emulsions with different

US

emulsifiers and wall materials (gum arabic, starch, maltodextrin, WPC), using high shear mixers (e.g. ultra-turrax, Silverson) and then dehydrating emulsions by spray-drying to obtain powdered

AN

microcapsules (Tonon et al., 2011, Tonon, et al., 2012, Carneiro et al., 2013, Gallardo et al., 2013, Karaca et al., 2013). Oil-to-wall material ratio usually ranges from 1:1 to 1:10 and when this ratio

M

increases so does the oil encapsulation efficiency within powdered microcapsules (Ramakrishnan et

ED

al., 2013), although microcapsules with high oil content are often demanded. During spray-drying, inlet air temperatures lower than 140°C should be avoided as this would lead to poor encapsulation

PT

efficiency and caking effects between the powdered microcapsules because of insufficient droplet

CE

drying. And for flaxseed oil, lipid oxidation is minimized at inlet air temperatures varying between 140 and 170°C (Tonon et al., 2011).

AC

However, Tonon et al. (2011, 2012) did not determine oxidation stability of the encapsulated oil during powders storage, whereas Carneiro et al. (2013) and Gallardo et al. (2013) evaluated the stability at 45°C, which would represent a temperature condition for an accelerated oxidation test instead of ranges often used for food storage (freezer, refrigerator, ambient temperatures). Karaca et al. (2013) formulated chickpea/lentil protein single layered emulsions with addition of MD and reported a protective effect against oil oxidation over 25 days storage at room temperature (25°C).

ACCEPTED MANUSCRIPT Nevertheless, coarse emulsions formulated in all the above-mentioned studies were prepared using rotor-stator mixers that would have produced larger droplet sizes than high pressure valve homogenizers. From a practical point of view, smaller droplet sizes might contribute to an enhanced physical stability of homogenized emulsions (Hebishy et al., 2015) and thereby of spray-dried

T

microcapsules. Kuhn & Cunha (2012) have studied the effect of high homogenization pressures

IP

(20-100 MPa) on the oxidative stability of flaxseed oil single-layer emulsions and found that

CR

formation of primary oxidation products increased when increasing homogenization pressure, suggesting 20 MPa as the best processing condition. Domian et al. (2018) also used high pressure

US

homogenization (75 MPa) to obtain spray-dried powders of single-layer flaxseed O/W emulsions,

AN

and it seemed like one of the powder formulations packed with non-modified atmosphere (air) did

was informed for these measurements).

M

not exceed PV allowed for foods after 3 months storage at 25°C (although no standard deviation

ED

According to literature, conventional emulsions are often prone to physical instability when exposed to environmental stresses such as heating, drying or during storage (McClements et al.,

PT

2007). Thereby, multilayer emulsions might constitute better emulsion-based delivery systems that

CE

can be designed using an electrostatic layer-by-layer deposition technique. The first step consists in obtaining a primary emulsion using an ionic emulsifier that rapidly adsorbs to the O/W interface

AC

during homogenization. Then, an oppositely charged biopolymer is added to the system to form a secondary emulsion with a two-layer interfacial membrane, thereby allowing successive deposition of layers around oil droplets stabilized by electrostatic charges (Aoki et al., 2015; Jiménez-Martínez et al., 2015). Several studies have reported that double-layer emulsions presented better stability than conventional single-layer emulsions regarding droplet aggregation, thermal processing, as well as lipid oxidation (Ogawa et al., 2003; Lim & Roos, 2017; Klinkesorn et al., 2005; Gharsallaoui et al., 2017). That is why in previous work, we attempted to microencapsulate flaxseed oil using

ACCEPTED MANUSCRIPT ultrasonically generated WPI-alginate double-layer emulsions (Fioramonti et al., 2017), but continuous

cavitation

might

have

accelerated

chemical

reactivity

of

PUFAs

as

high

hydroxyperoxide concentrations were observed. Hence, in this contribution we pretend to maintain low levels of oil oxidation and to start scaling-up the encapsulation process to pilot-scale using (i)

T

WPC, as it constitutes a cheaper reagent than WPI, and (ii) a two-stage conventional homogenizer

IP

during emulsification, where droplets are broken up only in the first orifice and the second orifice

CR

just delivers back-pressure to decrease cavitation phenomena and extreme local temperature peaks (Schlender et al. 2015, Finke et al. 2014).

US

To the knowledge of the authors, there is scarce information about microencapsulation of flaxseed

AN

oil using double-layered emulsions at conventional pressure homogenization (less than 30 MPa) with subsequent spray drying, which could probably provide better protection for the oil during

M

storage. In this context, the objective of this work was to obtain flaxseed oil spray-dried

ED

microcapsules producing double-layer emulsions at low pressure homogenization to study (i) the effect of homogenization pressure, pH and maltodextrin concentration on stability of primary and

PT

secondary emulsions and (ii) the influence of microencapsulation processing and powders storage

CE

on the oxidative stability of the oil.

AC

2. MATERIALS AND METHODS

2.1. Materials

Milk whey protein concentrate (WPC) and maltodextrin (DE 15) were kindly donated by Arla Foods (Buenos Aires, Argentina) and Productos de Maíz SA (Argentina), respectively. WPC composition on dry basis was: 80.8% proteins, 7.4% fat, 3.1% ash, 8.1% others. Commercial

ACCEPTED MANUSCRIPT sample of low density sodium alginate (SA) was provided by Cargill (Argentina) (MW 135 kDa). As stated by the manufacturer, composition of this alginate was: carbohydrate 63%, moisture 14%, ash 23% (Na+ 9300mg/100 g and K + 800mg/100 g). Cold pressed flaxseed oil was provided by

T

CIDA (Nogoyá, Argentina) and it was used without further purification.

CR

IP

2.2. Double layer emulsions preparation

Procedure 1

US

A primary emulsion was obtained by blending 18% (w/w) oil phase with 82% (w/w) aqueous

AN

phase of 4% WPC at pH 7, using a high-speed blender (Waring Blender, USA) for 30 s at the highest speed (24000 rpm). This pre-emulsion was then subjected to two passes at low pressure

M

homogenization and room temperature (25°C) using a two-stage homogenizer at pilot scale (SIMES

ED

S.A:, Argentina) to reduce droplet sizes. Pressures assessed in the first stage for primary emulsions were 3.5, 7 or 10.5 MPa and in the second stage 1.5, 3 and 4.5, respectively (total pressure: 5, 10

PT

and 15 MPa). Total processing pressure corresponds to the sum of the pressures at both stages

CE

(Table1) (Ruiz-Montañez et al., 2017). It should be highlighted that pressures for both stages were chosen to yield a Thoma number of 0.3 (Th; counter pressure/total pressure drop), which is in the

AC

range suggested by some authors to decrease cavitation intensity after the first orifice (Finke et. al 2014, Schlender et. al 2015). This first part was intended to evaluate the influence of homogenization pressure on primary WPC-monolayer emulsions. Primary emulsion was then diluted adding sodium alginate (SA) dispersion on continuous stirring using a magnetic mixer, and then adjusting pH to different values (7, 6, 5, 4) with HCl 0.3N to form secondary emulsions (4.5% oil, 1%WPC, 0.25%SA). Both WPC and SA concentrations to form secondary double-layer emulsions were selected considering optimum WPI:SA ratio reported

ACCEPTED MANUSCRIPT in previous work to produce stable systems (Fioramonti et al., 2015). This second part was intended to evaluate the influence of pH on WPC-SA double interfacial layer formation. After choosing pH 5 to form WPC-SA interfacial double-layer, secondary emulsions were then diluted adding MD solutions at the same pH to increase solids content. Final composition was:

T

4.5% oil, 1%WPC, 0.25%SA and 0.8-7% MD (Table 1). Maltodextrin concentrations were chosen

IP

based on oil-to-wall ratios 1:2 and 2:1, to obtain 33% and 66% oil load within powdered

CR

microcapsules, respectively. Homogenization pressures showed in Table 1 were selected after primary WPC emulsion analysis (15 MPa was disregarded). Individually prepared replicates were

AN

US

assayed for each condition.

Procedure 2

M

After choosing one homogenization pressure from previous procedure, slight modifications were

ED

introduced to obtain double-layer emulsions. One of them was to combine some steps mixing WPC and SA dispersions from the beginning and lowering pH during pressure homogenization to make

PT

the entire procedure more scalable. The other one was to increase MD concentration in emulsions to

CE

modify oil-to-wall ratio and provide better supporting matrix to spray-dried powders (Table 1) with the aim to improve oil encapsulation efficiency. So, this time, primary emulsions were prepared

AC

blending 18% (w/w) oil phase with 82% (w/w) mixed aqueous phase of both 4% WPC and 1% SA solutions (1:1) at pH 7 using the same conditions described above. This pre-emulsion was then subjected to two passes at low pressure homogenization (10 MPa) and pH was adjusted to 5 with HCl 0.3 N while systems were being processed in the homogenizer. Secondary emulsions were then diluted adding a MD solution at pH 5 while continuous stirring in a magnetic mixer. Final composition was: 5% oil, 1%WPC, 0.25%SA and 13% MD, having a core-to-wall ratio of 1:3 (25% oil load) in spray-dried powders (Table 1).

ACCEPTED MANUSCRIPT

2.3. Characterization of emulsions

2.3.1. Droplet Size

T

Droplet size of emulsions was measured by dynamic light scattering using a Zeta Sizer Nano ZS90

IP

(Malvern Instruments, UK) provided with a He-Ne laser (633 nm). Measurements were carried out

CR

at 25°C and a fixed scattering angle of 90°, within the range of 0.6 nm to 6 µm, according to equipment specifications. Samples were diluted 1/70 using filtered ultrapure water (through 0.45

US

and 0.22 µm nitrocellulose filters) (Merck Millipore, Ireland) and then placed into disposable

AN

polystyrene cuvettes of 1 cm pathlength. Refractive index of both the dispersed (1.48) and continuous phase (1.33) were used. Droplet size distributions were obtained as plots of the relative

M

intensity of light scattered by particles of different sizes and are reported as the average of ten

PT

2.3.2. Zeta Potential

ED

readings.

CE

Zeta potential of emulsions was also determined using the Zeta Sizer Nano ZS90 (Malvern Instruments, UK). In this case, systems were diluted 1/500 using filtered ultrapure water (through

AC

0.45 and 0.22 µm nitrocellulose filters) at different pH (the same one of the emulsion itself) and then placed into disposable polycarbonate cuvettes with inbuilt gold plated copper electrodes and bent capillary tube. Measurements were reported as the average and standard deviation of five determinations per sample.

2.4. Microcapsules obtention

ACCEPTED MANUSCRIPT Powdered flaxseed oil microcapsules were obtained dehydrating secondary emulsions by spraydrying using a Yamato ADL311S equipment (China) with a standard 0.406 mm nozzle, using the following conditions: 4 mL/min emulsion feed rate, 0.12 m3 /s drying air flow-rate, 0.15 MPa atomizing air pressure, drying air inlet temperature of 140°C (emulsions from Procedure 1) and

T

170°C (emulsions from Procedure 2), drying air outlet temperature between 85°C-95°C. It should

IP

be highlighted that this equipment only allows regulation of one drying air temperature, so inlet

CR

temperature was fixed and the other one was monitored during spray-drying. Dried microparticles were collected in a collection vessel and separately packed in non-modified atmosphere into plastic

US

eppendorfs covered with non-adhesive Teflon tape and wrapped in aluminium foil (to avoid light).

AN

Samples were stored at three different temperatures: -18°C, 4°C and 20°C, which were chosen considering storage conditions of different types of plausible food matrices (ice-creams, yogurts,

M

bakery products). Both microcapsules dehydration and storage experiments were performed in two

ED

replicates.

CE

PT

2.5. Oxidative stability of flaxseed oil

Oxidative stability of flaxseed oil was examined by measuring primary and secondary oxidation

AC

products at different processing steps (initial oil, emulsion formation, spray-drying), and during storage (i) every week up to 6 weeks and (ii) one-time measurement at 6 months to determine if microcapsules could have extended shelf-life for a long period, using the techniques described below. It should be highlighted that all spray-dried powders were easily rehydrated when adding water to reconstitute emulsions to assess oxidative stability of the encapsulated oil.

2.5.1. Primary oxidation compounds

ACCEPTED MANUSCRIPT Lipid hydroperoxides were determined as described in Fioramonti et al. 2017. Briefly, 0.3 mL of emulsion (fresh or reconstituted after adding water to spray-dried powders) was added to a mixture of 1.5 mL isooctane/isopropanol (3:2 v/v) followed by vortexing and centrifugation (3000 g, 15 min). Next, 0.2 mL of the upper organic phase was added to 2.8 mL of chloroform/methanol (7:3

T

v/v), followed by 15 L of ammonium thiocyanate solution (3.94 M) and 15 L of ferrous iron

IP

solution (prepared by mixing 0.132 M BaCl2 and 0.144 M FeSO4 in acidic solution). The mix was

CR

vortexed for 4 s and the absorbance at 500 nm was measured after 10 min incubation at room

US

temperature (25°C). The entire procedure was conducted in dim light and test tubes were covered with aluminum foil to avoid photo-oxidation of the samples. Lipid hydroperoxide concentrations

AN

were determined using a Fe3+ standard curve within the range 0-7 g Fe3+/mL. Peroxide value (PV)

( m  5 5 .8 4  m 0  2 )  4 .5

(1)

ED

( AS  ASB  ARB )

PT

PV 

M

was expressed as milliequivalents of peroxide per kilogram of sample as follows:

CE

where AS is the absorbance of the sample, ASB is the absorbance of the sample blank, ARB is the absorbance of the reagent blank, m is the slope obtained from the calibration curve, m 0 is the mass

AC

of the oil in grams. Each sample was measured in duplicate.

2.5.2. Secondary oxidation compounds Thiobarbituric acid-reactive substances (TBARS) were also determined as described in Fioramonti et al. 2017, where 1 mL of reconstituted emulsion was combined with 2 ml of TBA reagent and placed in a boiling water bath for 15 min. After cooling down to room temperature (25°C) for 10 min, samples were centrifuged (8000 g, 15 min) and the absorbance was measured at 532 nm.

ACCEPTED MANUSCRIPT Concentrations of TBARS were determined from a standard curve prepared using 1,1,3,3tetramethoxypropane.

T

2.6. Powdered microcapsules characterization

IP

2.6.1. Morphological analysis

CR

Microstructure of spray-dried samples was assessed by scanning electron microscopy (JSM-35C, JEOL, Japan). Powders were placed on the SEM stubs using a two-sided adhesive tape and were

US

coated with gold using a magnetron sputter coater. Coated samples were analyzed at accelerating

AN

voltage of 20 kV.

M

2.6.2. Encapsulation efficiency

ED

Encapsulation efficiency of powdered microcapsules was determined by the method described in Klinkesorn et al. (2006) with modifications. Briefly, extractable oil, usually referred as free oil (FO)

PT

was determined by adding 12 mL of hexane to 2 g powder, and after leaving 5 min at room

CE

temperature (25°C) the mixture was then centrifuged (3000 g, 5 min) (Heal Force, China). Supernatant was filtered with Munktell 00R filter paper, powder residue was rinsed twice with

AC

hexane, and hexane was evaporated in a rotatory evaporator at 60°C for 15 min. The solvent-free extract was dried at 105°C. The amount of free oil was determined gravimetrically. Total oil (TO) was assumed to be equal to the initial oil as flaxseed oil is not volatile and it is expected to be retained (Carneiro et al. 2013). All determinations were done in duplicate. Encapsulation efficiency was calculated as:

ACCEPTED MANUSCRIPT % EE 

T O ( g /100 g pow der)  F O ( g /100 g pow der)

 100

(2)

T O ( g /100 g pow der)

2.6.3. Water activity

T

Water activity of powdered microcapsules was determined at ambient temperature (25°C) using an

CR

IP

Aqualab system (Washington, USA). Samples were measured in triplicate.

2.7. Statistical analysis

US

Each experiment was carried out with its corresponding replication (two replicates). All assays were

AN

performed at least in duplicate. Averages and standard deviations were calculated from these measurements. Differences between means were determined by applying analysis of variance using

M

LSD test at p<0.05 significance level. When homogeneity of variance assumption was not satisfied,

ED

Kruskal Wallis test at p<0.05 and boxplots were used to identify significant differences.

CE

PT

3. RESULTS AND DISCUSSION

3.1. Procedure 1: Effect of homogenization pressure, pH and maltodextrin concentration on

AC

microcapsules formation

Emulsion stability

The first step to properly microencapsulate edible oils is to produce stable emulsions (Gharsallaoui et al., 2007) and droplet size is one of the main parameters to take into account as it is related to destabilization of emulsions due to the upward movement of droplets. As postulated by

ACCEPTED MANUSCRIPT McClements (1999), the creaming velocity of an individual droplet is directly proportional to the square of its radius and to the density difference between the dispersed and the continuous phase. Therefore, one of the strategies to reduce creaming destabilization is to produce emulsions with lower droplet sizes (Robins, 2000), and the first hypothesis was that higher homogenization

T

pressures would produce emulsions with lower droplet sizes. Using such criteria, single layered

IP

primary emulsions were formulated with flaxseed oil and WPC dispersions and the effect of

CR

homogenization pressure on droplet size distributions was studied (Fig. 1A). It can be observed that all emulsions exhibited two main populations, one predominant peak at higher droplet sizes and one

1

µm and

another smaller peak

around

0.19

µm.

When increasing

AN

maximum around

US

smaller peak at lower sizes. Emulsions homogenized at 5 MPa showed one predominant peak with a

homogenization pressure, a left shift of both peaks toward smaller droplet sizes was observed. At

M

first glance there is no such a big difference between droplet sizes of emulsions prepared at 10 and

ED

15 MPa, although systems obtained at the highest homogenization pressure exhibited a better resolution of the two droplet populations, as the peaks are narrower and well separated from each

PT

other. These results are consistent with those observed by Santana et al. (2011) when studying the

CE

effect of different homogenization pressures (20-100 MPa) on droplet size of flaxseed oil–whey protein emulsions, where it was shown that increasing homogenization pressures produced higher

AC

droplet populations of smaller sizes. In contrast to this, Juttulapa et al. 2017 reported that homogenization pressure did not affect droplet sizes of pectin-zein stabilized rice bran oil emulsions prepared at 50, 100 and 150 MPa, and they have attributed this to high oil concentration (20% w/w) used in the formulations, suggesting this could have increased the system viscosity, resulting in larger droplet sizes. Nevertheless, it must be born in mind that emulsions in these other studies were prepared using higher homogenization pressures (above 20 MPa) than the range assessed in the

ACCEPTED MANUSCRIPT present work. In our case, as homogenization at 15 MPa did not produce further reduction of droplet sizes, we decided not to continue working on this treatment. Zeta potential measurements were also performed on primary WPC-flaxseed oil emulsions at pH 7 (Fig. 1B) and it was observed that homogenization pressure did not have a significant effect (p >

T

0.05) on emulsions surface potential. This was expected as zeta potential of particles mainly

IP

depends on pH and electrolyte composition (ionic strength) of dispersions (Kasha et al. 2015,

CR

Bhattacharjee, 2016) and in this case, those variables were not affected. It also bears noting that emulsions presented zeta potential values around -44 mV, thus indicating highly stable systems as

US

electrostatic repulsion between droplets with absolute zeta potential above 30 mV would be

AN

enough to overcome attractive droplet-droplet interactions (Bhattacharjee, 2016). The next step was to perform electrostatic deposition of alginate molecules onto the protein

M

interfacial film surrounding the droplets to form double layer emulsions. As stated before, SA

ED

molecules tend to be negatively charged across a wide range of pH (pH > 3) due to ionization of carboxyl groups (Wu et al., 2018), whereas proteins behave as ampholytes, as their net charge

PT

varies at different pH. Taking this into account, WPC net charge will be strongly influenced by the

CE

pI of their main proteins, β-lactoglobulin (4.7-5.2) α-lactalbumin (4.2-4.5) (El-Salam et al., 2009). In previous work, the influence of pH on the formation of alginate double layers onto ultrasonically

AC

generated flaxseed oil-WPI emulsions was studied (Fioramonti et al., 2015). But in this case, WPC contains a higher content of impurities than WPI (lactose, albumins, fats, ash) that might modify surface potential of emulsion droplets. So, this time, the influence of pH on zeta potential of WPCSA flaxseed oil emulsions was assessed and results are shown in Fig. 2. It was observed that, in the absence of SA, surface potential of primary WPC emulsions decreased from -41.9 mV to -10.7 when lowering pH from 7 to 4. This could be attributed to charge variation on aminoacidic residues

ACCEPTED MANUSCRIPT of the protein adsorbed onto the droplets’ surface, becoming less negative at pH values near the isoelectric point (Pongsawatmanit et al., 2006). It is interesting to note that zeta potential values of primary emulsions prepared with WPC differ somewhat from those prepared with WPI in previous study (Fioramonti et al., 2015), e.g WPI emulsions at pH 5 presented zeta potential of -11 mV

T

whereas in WPC emulsions this value was -29 mV. This difference would be consistent with the

IP

presence of higher content of impurities and electrolyte species in WPC (lactose, ash and fats) that

CR

could interact with charged surfaces and thus modify zeta potential of emulsions (Bhattacharjee et al., 2016). That could also explain that, at pH 4, WPC primary emulsions still presented a slightly

AN

the range of pI of WPC main proteins (4.2-5.2).

US

negative zeta potential (-10.74 mV), although this condition corresponds to a pH value that is below

When adding SA to form secondary emulsions, it can be observed that both WPC and WPC-SA

M

systems presented values of zeta potential around -42 mV at pH 7, showing no significant

ED

differences (p>0.05) and indicating that addition of the polysaccharide did not affect droplets’ surface potential. Therefore, under these conditions, SA molecules might not adsorb onto the

PT

protein interface surrounding the droplets as both biopolymers would present a net negative charge

CE

and might be mutually excluded by electrostatic repulsion (Rodriguez Patino & Pilosof, 2011). When decreasing pH from 6 to 4, secondary WPC-AS emulsions exhibited significant (p>0.05)

AC

differences in their zeta potential, showing more negative values when compared to primary WPC emulsions. The contrast was higher at pH 4 and 5. These results suggest that SA molecules would have been adsorbed onto the protein layer surrounding oil droplets to form double interfacial layers (Klinkesorn et al., 2005; Gu et al., 2005). And this could be explained taking into account that, although at these pH values primary emulsions presented a net negative surface potential, positively charged patches could still be exposed onto the protein surface, which would electrostatically

ACCEPTED MANUSCRIPT interact with SA negatively charged carboxyl groups, thereby promoting the formation of the double layer by self-assembly (Jones & McClements, 2011, De Kruif et al., 2004). It bears noting that at pH 4, zeta potential difference between primary and secondary emulsions was higher than at pH 5, suggesting the adsorption of an increased number of SA molecules to the

T

interfacial protein layer. This might promote a higher protection of the oil in secondary emulsions,

IP

since a double layer of greater thickness would be formed. However, at the experimental level, it

CR

was difficult to achieve such low pH values as emulsions became visibly thicker and droplets agglomerates started to form, thereby hindering the acidification process. That is why pH 5 was

US

chosen to continue with the studies.

AN

After selecting the best conditions to form the WPC-AS double layer around droplets, maltodextrin was added to emulsions to increase solid content of formulations. As stated before, MD is a neutral

M

polysaccharide that has no surface activity (it does not adsorb to the oil-water interface) but acts as

ED

supporting material that becomes part of the microcapsules wall once emulsions are dehydrated (Gharsallaoui et al., 2007). In this section, the effect of both homogenization pressure and MD

PT

concentration on the stability of secondary emulsions was studied. When analyzing zeta potential

CE

measurements of secondary emulsions containing MD presented in Fig. 3B, it was observed that addition of MD to systems prepared at pH 5 did not produced a significant effect (p<0.05). Indeed,

AC

emulsions obtained at both homogenization pressures and MD concentrations showed zeta potential values similar to their corresponding secondary emulsions containing no MD at pH 5 (-46.7 mV) (Fig. 2). These results were expected as MD is a neutral polysaccharide with no surface activity, so it should not interact with the interfacial WPC-SA bilayer surrounding oil droplets (Fioramonti et al., 2015b). It should also be noticed that all emulsions exhibited absolute surface potential values above 30 mV, thereby constituting highly stable colloidal systems as electrostatic repulsive forces

ACCEPTED MANUSCRIPT would be enough to overcome droplet-droplet interactions thorough van der Waals forces (Bhattacharjee et. al 2016, Fioramonti et. al, 2015a). Droplet size distributions of WPC-SA secondary emulsions containing MD are presented in Fig. 3A where it can be observed one predominant peak at droplet sizes between 0.5 – 6 μm, and another

T

small population of droplet sizes between 0.1 – 0.5 μm (almost negligible). When compared with

IP

primary emulsions obtained at the same homogenization pressure (10 MPa) presented in Fig. 1A, a

CR

shift of both peaks towards bigger droplet sizes was evidenced in secondary emulsions. This could be due both to formation of a thicker interfacial WPC-SA bilayer and possibly to a partial

US

agglomeration of some droplets since alginate macromolecules are likely to bridge them.

AN

Gharsallaoui et al. (2010) also reported a left shift towards higher droplet sizes (d43 ~ 300 nm) when comparing O/W single-layer protein-stabilized emulsions and double-layer protein-pectin

M

emulsions. And the difference in oil droplet sizes was attributed to pectin adsorbed onto the protein-

ED

coated interfacial surfaces.

When analyzing droplet size distributions of systems obtained at different homogenization

PT

pressures and MD concentrations, they were almost the same, except for emulsions prepared at 10

AC

3A).

CE

MPa with the lowest MD concentration which showed a slight shift towards smaller drop sizes (Fig.

Oxidative stability of the oil

The next step was to assess oxidative stability of flaxseed oil throughout the encapsulation process, from starting material to powdered microcapsules obtention. Table 2 shows PV and TBARS values of the oil at different microencapsulation stages (initial bulk oil, after emulsification, after spray-

ACCEPTED MANUSCRIPT drying) for each system formulated with different MD concentration and homogenization pressures. It was clearly observed that encapsulation process produced a gradual increase on the concentration of hydroperoxides, as significant differences (p > 0.05) were found on PV during different processing stages, for each formulated system. This would be naturally expected as during

T

emulsification stage, lipid oxidation might have been favored due to the increment of interfacial

IP

area, leading to a large contact surface between the oxidizable oil droplets and water-soluble

CR

compounds (including oxygen and prooxidants), which could contribute to the initiation and propagation of oxidation reactions (Waraho et al., 2010). In addition, high temperatures during

US

spray drying of emulsions could have increased hydroperoxides concentration even more. And

AN

oxidative deterioration of flaxseed oil would be more related to exposure of surface oil to high temperatures during microcapsules crust formation after emulsion atomization inside the spray

M

dryer. These results are consistent with those reported by Hogan et al. (2003) when encapsulating

ED

menhaden oil by spray-drying using sodium caseinate and carbohydrates as wall materials, as they also found that PV of fresh powders were greater than those of their corresponding emulsions.

PT

According to Huang et al. (2012), high temperatures during spray drying might provide more

CE

energy for the lipid oxidation process to occur. However, it bears noting that PV of all systems were still within the range allowed for cold pressed

AC

oils (15 meq of active oxygen/Kg oil) (Codex Alimentarius, 2001). When focusing on secondary oxidation products, in Table 2 it can be observed that all systems presented low TBARS values at different processing stages, thereby indicating low hydroperoxides decomposition throughout processing steps. These results are in agreement with low levels of PV found in all formulations and the absence of rancid off-flavors in spray-dried powders (Pingret et al., 2013). Although there are no standardized reference values of TBARS - as it is for PV in the Codex Alimentarius - generally speaking, TBARS less than 1 mmol/Kg oil would be acceptable.

ACCEPTED MANUSCRIPT Indeed, Avramenko et al. (2016) and Karaca et al. (2013) reported TBARS of 1 mmol/Kg oil and 3.9 mmol/Kg oil, respectively, for bulk flaxseed oil during storage at room temperature when PV was almost reaching 15 meq/Kg oil.

IP

T

Characterization of powdered microcapsules

CR

Table 3 shows aw of powdered microcapsules and it can be noticed that all values were between the range 0.2-0.4, which is considered as stable for lipid oxidation, microbial growth, browning and

US

enzymatic reactions (Leung, 1987, Caliskan et al., 2016). Besides, microcapsules obtained at each

AN

homogenization pressure, exhibited lower aw at higher MD concentrations. This latter is consistent with previous studies as several authors have reported a decrease of a w with increasing MD content

M

in spray-dried powders (Caliskan et al., 2016; Oberoi et al., 2015; Quek et al., 2007; Ekpong et al.,

ED

2016) which would be related to the ability of polysaccharides to bind water molecules through hydrogen bonding to their hydroxyl groups, thereby modifying free water availability (Fenemma,

PT

1997). Table 3 also shows the values of free oil (FO), total oil (TO) and encapsulation efficiency

CE

(EE) for all formulated systems. It was observed that EE depended mainly on MD content, as microcapsules with the lowest MD concentration had EE values less than 35%, while those with the

AC

highest MD content showed EE values greater than 50%. These results agree with those discussed by Gharsallaoui et al. (2007), who suggested that the EE could be improved by increasing total solids concentration in emulsions, since they become part of the wall structure during dehydration processes acting as a supporting matrix surrounding microcapsules. Besides, when comparing the effect of homogenization pressure at a fixed MD concentration, it can be noticed that EE increased at higher homogenization pressures. The best EE was reached in systems obtained at 10 MPa at the highest MD, where almost 60% of oil encapsulation was achieved. This data suggest that smaller

ACCEPTED MANUSCRIPT droplet sizes and higher total solids content of emulsified systems would promote a better encapsulation of flaxseed oil. Although obtaining 60% EE is not a promising result for flaxseed oil microencapsulation, Gallardo et al. (2013) obtained 25.5% EE when producing spray-dried microcapsules with flaxseed oil, maltodextrin and methyl cellulose as wall materials at the same oil-

T

to-wall ratio (1:2). Hence, the formulation proposed in the present work would represent a better

IP

encapsulation system in this particular case.

CR

When examining macroscopic appearance of spray-dried microcapsules it was clearly seen that powders formulated with the lowest MD concentration presented a more yellow color compared to

US

those containing higher MD. This is consistent with the greater surface oil found in the former

AN

systems (Table 3). The increase in MD content remarkably improved the appearance of powders, thus evidencing better protection of flaxseed oil.

M

Morphology of the surface and structure of spray-dried microcapsules formulated with the highest

ED

MD concentration – those exhibiting better EE – were observed using SEM (Fig. 4B and C). Spherical shaped particles of different sizes ranging from 1-10 μm can be identified. Microcapsules

PT

presented smooth surfaces with a few wrinkles on some of them and these results are consistent

CE

with those reported by other authors when using MD to obtain spray-dried microcapsules (Silva et al. 2013; Ghani et al., 2017). According to Gharsalloui et al. (2012), low-molecular-weight sugars

AC

contained in MD of certain DE may act as plasticizer preventing irregular shrinkage of the surface during drying and thus promoting the formation of spherical microcapsules with smooth surface. As suggested by Tonon (2009), particles with rough surfaces showing high indentation have larger contact areas when comparing with smooth surfaces, thereby making microcapsules more susceptible to degradation reactions, such as oxidation. Moreover, flow properties of powders would be improved when microspheres have fewer depressions on their surface (Silva et al., 2013).

ACCEPTED MANUSCRIPT In systems obtained at 5 MPa, there seems to be a higher proportion of microcapsules of larger sizes (Fig. 4B) when compared with those obtained at 10 MPa. These results should, however, be interpreted with caution as it must be born in mind that microscopy is a qualitative technique and a considerable high number of fields needs to be counted to infer about particle sizes. It should also

T

be clarified that microcapsules containing the lowest MD concentration were not analyzed by SEM

IP

due to the high content of unencapsulated surface oil, which could have caused malfunction of the

CR

equipment.

Although the application of Procedure 1 allowed us to produce stable WPC-SA double-layer

US

emulsions that after spray drying led to powdered microcapsules with suitable values of aw, PV and

AN

TBARS, the encapsulation efficiency still does not seem to be high enough to properly protect the oil from further oxidation. As stated by Karaca et al. (2013), it is crucial to minimize the amount of

M

free oil in lipid encapsulation, as this material can undergo oxidative deterioration faster than the

ED

encapsulated oil, thereby reducing shelf life of the functional ingredient. Taking this into account, the homogenization pressure that produced the highest encapsulation efficiency (10 MPa) was

PT

chosen and slight modifications were made during formulation of emulsions as previously described

CE

in Procedure 2. The first modification was to combine some steps to make the entire procedure more scalable. In Procedure 1 (i) primary WPC emulsions were first prepared using the high-speed

AC

blender and the valve-homogenizer, (ii) then transferred to a beaker and SA was added while mixing in a magnetic stirrer, (iii) then pH was lowered from 7 to 5 while using the magnetic stirrer. Whereas in Procedure 2, we tried to combine all these steps in one to make the entire procedure more scalable: primary emulsions including WPC and SA were prepared in the high-speed blender at pH 7, then passed through the valve-homogenizer while pH was lowered dropwise to 5 in the same equipment. The second modification was to increase MD concentration to increment total solids content of emulsions from 13% to 20% to provide a better supporting matrix to protect

ACCEPTED MANUSCRIPT microcapsules during spray-drying (Gharsallaoui et al., 2007). And the third modification was to raise spray-dryer inlet temperature to promote better oil encapsulation during dehydration. In Procedure 1, an inlet temperature of 140°C was chosen in order to protect polyunsaturated fatty acids from oxidative deterioration and considering that Tonon et al. (2011) reported lipid oxidation

T

of microencapsulated flaxseed oil was minimized at inlet air temperatures between 140°C and

IP

170°C. On the other hand, some authors also suggest that the inlet air temperature is an important

CR

factor that influences morphology of oil microcapsules. And that low inlet air temperatures may not be sufficient for droplet drying and proper crust formation around microcapsules, leading to oil

US

leaking and an inefficient encapsulation (Encina et al., 2016; Jimenez-Martín et al., 2015; Tonon et

AN

al., 2011). That is why in Procedure 2, inlet air temperature was raised to 170°C, which still was

M

within the range reported by Tonon et al. (2011) to reduce flaxseed oil oxidative deterioration.

ED

3.2. Procedure 2: Effect of different storage temperatures on oxidation stability of powdered

PT

microcapsules

CE

Table 4 summarizes parameters obtained for flaxseed oil microcapsules using Procedure 2. Spraydried microcapsules using this procedure, showed an increment in EE as 84.39% of oil

AC

encapsulation was achieved (Table 4). Droplet size distribution (data not shown) and zeta potential (-44.95 mV) of emulsions were checked and both were within the range reported in previous section for systems formulated with MD (Fig. 3A and 3B). Spray dried powders showed PV of 6.8 meq/Kg oil and negligible TBARS values (less than 0.01 mmol/Kg oil), which were also within the range reported in Table 3. Nevertheless, it should be noted that PV of spray-dried powders obtained by Procedure 2 were slightly higher than those from microcapsules obtained with Procedure 1. Indeed, this might not have been related to the process itself but to initial flaxseed oil hydroperoxides

ACCEPTED MANUSCRIPT concentration for Procedure 2 (1.84 meq/Kg oil), thus being slightly higher than those reported for Procedure 1 (Table 3). Powdered microcapsules obtained by Procedure 2 were stored at different temperatures and oxidation stability of the oil was measured during storage. Fig. 5A shows PV for microcapsules

T

stored at -18°C, 4°C and 20°C, respectively. It was observed that powders stored at -18 and 4°C up

IP

to 6 months (26 weeks), maintained almost the same hydroperoxide concentration they had before

CR

storage, and all PV were below the maximum allowed for cold pressed oils (15 meq of active oxygen/Kg oil) (Codex Alimentarius, 2001). Hence, this would be evidencing a suitable protection

US

of flaxseed oil within the structure of the microcapsule, that comprises a WPC-AS double layer

AN

surrounded by a MD outer coating. The same behavior was observed for powders kept at 20°C up to 6 weeks, but after 6 months PV reached 20 meq/Kg oil thereby exceeding the maximum allowed

M

for cold pressed oils. These latter results can be attributed to the fact that higher oil oxidation rates

ED

are promoted at higher temperatures, as proposed by other authors. Zuta et al. (2007), who studied the oxidative stability of mackerel oil at different storage temperatures (-40, 4 and 30°C), reported

PT

higher PV at the highest temperature. This trend has also been found by Hogan et al. (2003) when

and 30°C).

CE

assessing PV of menhaden oil spray-dried microcapsules at different storage temperatures (4, 23

AC

When analyzing secondary oxidation products (Fig 5B), again, it can be observed that powders showed almost negligible TBARS values at all temperatures, thus indicating low hydroperoxides decomposition, which is consistent with low levels of PV discussed above.

4. CONCLUSIONS Flaxseed oil powdered microcapsules were produced by spray-drying of double layer WPC-SA emulsions. When formulating WPC single layered primary emulsions, smaller droplet sizes were

ACCEPTED MANUSCRIPT observed with higher homogenization pressures up to 10 MPa. After adding SA to produced secondary emulsions, pH 5 was selected as the best condition to form self-assembled interfacial WPC-SA double layers around oil droplets. Further addition of different MD concentrations to secondary emulsions did not have a significant effect on surface potential of emulsion droplets, thus

T

preserving the WPC-SA interfacial membrane. Encapsulation efficiencies up to 84% were obtained

IP

in powdered microcapsules, having the best results when increasing MD concentration. Regarding

CR

oxidative stability of flaxseed oil, encapsulation process produced a gradual increase on the concentration of hydroperoxides, however, these values were maintained relatively constant and did

US

not exceed the maximum allowed for cold pressed oils during 6 months storage at -18 and 4°C, and

AN

6 weeks storage at 20°C. These experiments have shown that it was possible to turn a highly oxidizable liquid ingredient into a solid easy-to-handle powder with peroxide values within the

M

range allowed for foods. This technology could also be applied for encapsulation of other oils with

ED

high content of omega-3 fatty acids (chia, fish, krill) to produce bioactive microcapsules as a nutraceutical ingredient that could be then incorporated into different food matrices (ice-creams,

PT

yogurts, bakery products). Further work will be needed to establish the influence of food matrices

CE

on stability of microencapsulated oils during shelf life.

AC

Acknowledgements

This research was supported by the project CAI+D PI 2011 from Universidad Nacional del Litoral (UNL, Argentina). Authors would also like to thank for the financial support of Consejo Nacional de Investigaciones Científicas y Técnicas de la República Argentina (CONICET).

ACCEPTED MANUSCRIPT

REFERENCES Aoki, T., Decker, E. A., & McClements, D. J. (2015). Influence of environmental stresses on stability of O/W emulsions containing droplets stabilized by multilayered membranes produced

T

by a layer-by-layer electrostatic deposition technique. Food Hydrocolloids, 19, 209-220.

IP

Avramenko, N.A., Chang, C., Low, N.L., & Nickerson, M.T. (2016). Encapsulation of flaxseed oil

CR

within native and modified lentil protein-based microcapsules. Food Research International, 81, 17-24.

US

Bhatacharjee, S. (2016). DLS and zeta potential – What they are and what they are not? Journal of

AN

Controlled Release, 235, 337-351.

Caliskan, G., & Dirim, S. N. (2016). The effect of different drying processes and the amounts of

M

maltodextrin addition on the powder properties of sumac extract powders. Powder Technology,

ED

287, 308-314.

Carneiro, H. C. F., Tonon, R. V., Grosso, C. R. F., & Hubinger, M. D. (2013). Encapsulation

PT

efficiency and oxidative stability of flaxseed oil microencapsulated by spray drying using

CE

different combinations of wall materials. Journal of Food Engineering, 115 (4), 443-451. Codex Alimentarius – Volume 8. (2001). Fats, oils and related products. (2nd ed.). Rome: Food

AC

and Agriculture Organization of the United Nations. De Kruif, C.G., Weinbreck, F., & de Vries, R. (2004). Complex coacervation of proteins and anionic polysaccharides. Current Opinion in Colloid & Interface Science, 9, 340-349. Domian, E., Cenkier, J., Górska, A., & Brynda-Kopytowska, A. (2018). Effect of oil content and drying method on bulk properties and stability of powdered emulsions with OSA starch and linseed oil. LWT – Food Science and Technology, 88, 95-102.

ACCEPTED MANUSCRIPT Ekpong, A., Phomkong, W., & Onsaard, E. (2016). The effects of maltodextrin as a drying aid and drying temperature on production of tamarind powder and consumer acceptance of the powder. International Food Research Journal, 23, 300-308. El-Salam, M.H., El-Shibiny, S., & Salem, A. (2009). Factors affecting the functional properties of

T

whey protein products: A review. Food Reviews International, 25, 251-270.

IP

Encina, C., Vergara, C., Giménez, B., Oyarzún-Ampuero, F., & Robert, P. (2016). Conventional

CR

spray-drying and future trends for the microencapsulation of fish oil. Trends in Food Science & Technology, 56, 46-60.

US

Fenemma, O.R. (1997). Food Chemistry. (3rd ed.). New York: Marcel Dekker.

Goymann,

C.

C.

(2014).

Multiple

AN

Finke, J. H., Niemann, S., Richter, C., Gothsch, T., Kwade, A., Büttgenbach, S., & Müllerorifices

in

customized

microsystem high-pressure

M

emulsification: The impact of design and counter pressure on homogenization efficiency.

ED

Chemical Engineering Journal, 248, 107-121. Fioramonti, S. A., Perez, A. A., Aríngoli, E. E., Rubiolo, A. C., & Santiago, L. G. (2014). Design

PT

and characterization of soluble biopolymer complexes produced by electrostatic self-assembly

CE

of a whey protein isolate and sodium alginate. Food Hydrocolloids, 35, 129-136. Fioramonti, S. A., Arzeni, C., Pilosof, A. M. R., Rubiolo, A. C., & Santiago, L. G. (2015b).

AC

Influence of freezing temperature and maltodextrin concentration on stability of linseed oil-inwater multilayer emulsions. Journal of Food Engineering, 156, 31-38. Fioramonti, S. A., Martínez, M. J., Pilosof, A. M. R., Rubiolo, A. C., & Santiago, L. G. (2015a). Multilayer emulsions as a strategy for linseed oil microencapsulation: Effect of pH and alginate concentration. Food Hydrocolloids, 43, 8-17.

ACCEPTED MANUSCRIPT Fioramonti, S. A., Rubiolo, A. C., & Santiago, L. G. (2017). Characterization of freeze-dried flaxseed oil microcapsules obtained by multilayer emulsions. Powder Technology, 319, 238244. Gallardo, G., Guida, L., Martinez, V., López, M. C., Bernhardt, D., Blasco, R., Pedroza-Islas, R., &

IP

application. Food Research International, 52, 437-482.

T

Hermida, L. G. (2013). Microencapsulation of linseed oil by spray drying for functional food

spray-drying in microencapsulation of food

ingredients: An overview. Food Research

US

International, 40, 1107-1121.

CR

Gharsallaoui, A., Roudant, G., Chambin, O., Voilley, A., & Saurel, R. (2007). Applications of

AN

Gharsallaoui, A., Saurel, R., Chambin, O., Cases, E., Voilley, A., & Cayot, P. (2010). Utilisation of

Food Chemistry, 122, 447-454.

A., Adachi, S., Shiga, H., Neoh, T. L., Adachi, S., & Yoshii, H. (2017). Effect of

ED

Ghani, A.

M

pectin coating to enhance spray-dry stability of pea protein-stabilised oil-in-water emulsions.

different dextrose equivalents of maltodextrin on oxidation stability in encapsulated fish oil by

PT

spray drying. Bioscience, Biotechnology, and Biochemistry, 81 (4), 705-711.

CE

Gharsallaoui, A., Roudaut, G., Chambin, O., Voilley, A., & Rémi, S. (2007). Applications of spraydrying in microencapsulation of food ingredients: An overview. Food Research International,

AC

40,1107-1121.

Gharsallaoui, A., Saurel, R., Chambin, O., Cases, E., Voilley, A., & Cayot, P. (2010). Utilisation of pectin coating to enhance spray-dry stability of pea protein-stabilised oil-in-water emulsions. Food Chemistry, 122, 447-454. Gharsallaoui, A., Saurel, R., Chambin, O., & Voilley, A. (2012). Pea (Pisum sativum, L.) protein isolate stabilized emulsions: A novel system for microencapsulation of lipophilic ingredients by spray drying. Food Bioprocess Technology, 5, 2211-2221.

ACCEPTED MANUSCRIPT Gu, Y.S., Decker, E.A., & McClements, D.J. (2005). Influence of pH and carrageenan type on properties of -lactoglobulin stabilized oil-in-water emulsions. Food Hydrocolloids, 19, 83-91. Hebishy, E., Buffa, M., Guamis, B., Blasco-Moreno, A., & Trujillo, A. J. (2015). Physical and oxidative stability of whey protein oil-in-water emulsions produced by conventional and ultra

T

high-pressure homogenization: Effects of pressure and protein concentration on emulsion

IP

characteristics. Innovative Food Science & Emerging Technologies, 32, 79-90.

CR

Hogan, S. A., O’Riordan, E. D., & O’Sullivan, M. (2003). Microencapsulation and oxidative

US

stability of spray-dried fish oil emulsions. Journal of Microencapsulation, 20 (5), 675-688. Huang, H., Hao, S., Li, L., Yang, X., Cen, J., Lin, W., & Wei, Y. (2014). Influence of emulsion

AN

composition and spray-drying conditions on microencapsulation of tilapia oil. Journal of Food Science and Technology, 51 (9), 2148-2154. E.,

&

Gharsallaoui,

A.

M

Jiménez-Martínez,

(2015).

Suitability of using monolayered

and

ED

multilayered emulsions for microencapsulation of -3 fatty acids by spray drying: Effect of

PT

storage at different temperatures. Food and Bioprocess Technology, 8, 100-111. Jones, O. G., & McClements, D. J. (2011). Recent progress in biopolymer nanoparticle and formation

by

heat-treating

electrostatic

protein-polysaccharide

complexes.

CE

microparticle

Advances in Colloid and Interface Science, 167, 49-62.

AC

Juttulapa, M., Piriyaprasarth, S., Takeuchi, H., & Sriamornsak, P. (2017). Effect of high-pressure homogenization on stability of emulsions containing zein and pectin. Asian Journal of Pharmaceutical Sciences, 12, 21-27. Karaca, A. C., Low, N., & Nickerson, M. (2013). Encapsulation of flaxseed oil using a benchtop spray dryer for legume protein-maltodextrin microcapsule preparation. Journal of Agricultural and Food Chemistry, 61, 5148-5155.

ACCEPTED MANUSCRIPT Kasha, A., Al-Hashim, H., Abdallah, W., Taherian, R., & Sauerer, B. (2015). Effect of Ca 2+, Mg2+ and SO 4 2- ions on the zeta potential of calcite and dolomite particles aged with stearic acid. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 482, 290-299. Kearney, J. (2010). Food comsumption trends and drivers. Philosophical Transactions of the Royal

T

Society, 365, 2793-2807.

IP

Klinkesorn, U., Sophanodora, P., Chinachoti, P., McClements, D. J., & Decker, E. A. (2005).

CR

Increasing the oxidative stability of liquid and dried tuna oil-in-water emulsions with electrostatic layer-by-layer deposition technology. Journal of Agricultural and Food Chemistry,

US

53, 4561-4566.

AN

Klinkesorn, U., Sophanodora, P., Chinachoti, P., Decker, E.A., & McClements, D.J. (2005). Encapsulation of emulsified membranes prepared using electrostatic layer-by-layer depositon.

M

Food Hydrocolloids, 19 (6), 1044-1053.

ED

Klinkesorn, U., Sophanodora, P., Chinachoti, P., Decker, E.A., & McClements, D.J. (2006). Characterization of spray-dried tuna oil emulsified in two-layered interfacial membranes

PT

prepared using electrostatic layer-by-layer deposition. Food Research International, 39, 449-

CE

457.

Kuhn, K. R., & Cunha R. L. (2012). Flaxseed oil – Whey protein isolate emulsions: Effect of high

AC

pressure homogenization. Journal of Food Engineering, 111, 449-457. Leung, H. K. (1987). Influence of water activity on chemical reactivity. In: L. B. Rockland, & L. R. Beuchar (Eds.), Water Activity: Theory and Applications to Food (pp. 27-54). Chicago: CRC Press. Lim, A. S. L., & Roos, Y. H. (2017). Carotenoids stability in spray dried high solids emulsions using layer-by-layer (LBL) interfacial structure and trehalose-high DE maltodextrin as glass former. Journal of Functional Foods, 33, 32-39.

ACCEPTED MANUSCRIPT McClements, D. J. (1999). Food Emulsions: Principles, Practice, and Techniques. Boca Ratón: CRC Press. McClements, D. J., Decker, E. A., & Weiss, J. (2007). Emulsion-based delivery systems for lipophilic bioactive components. Journal of Food Science, 72 (8), 109-124.

T

Oberoi, D. P. S., & Sogi, D.S. (2015). Effect of drying methods and maltodextrin concentration on

IP

pigment content of watermelon juice powder. Journal of Food Engineering, 165, 172-178.

CR

Ogawa, S., Decker, E. A., & McClements, D. J. (2003). Influence of environmental conditions on the stability of oil in water emulsions containing droplets stabilized by lecithin-chitosan

US

membranes. Journal of Agricultural and Food Chemistry, 51, 5522-5527.

AN

Pingret, D., Fabiano-Tixier, A., & Chemat, F. (2013). Degradation during application of ultrasound in food processing: A review. Food Control, 31, 593-606.

M

Quek, S. Y., Chok, N. K., & Swedlund, P. (2007). The physicochemical properties of spray-dried

ED

watermelon powders. Chemical Engineering Processing, 46, 386-392. Ramakrishnan, S., Ferrando, M., Aceña-Muñoz, L., De Lamo-Castellví, S., & Güell, C. (2013). Fish

PT

oil microcapsules from O/W emulsions produced by premix membrane emulsification. Food

CE

and Bioprocess Technology, 6, 3088-3101. Robins, M. (2000). Emulsions – Creaming phenomena. Current Opinion in Colloid and Interface

AC

Science, 5, 265-272.

Rodriguez Patino, J.M., & Pilosof, A.M.R. (2011). Protein-polysaccharide interactions at fluid interfaces. Food Hydrocolloids, 25, 1925-1937. Ruiz-Montañez,

G.,

Chevalier-Lucia,

Ragazzo-Sanchez,

J.A.,

Picart-Palmade, L., Calderón-Santoyo, M., &

D. (2017). Optimization of nanoemulsions processed by high-pressure

homogenization to protect a bioactive extract of jackfruit (Artocarpus heterophyllus Lam). Innovative Food Science & Emerging Technologies, 40, 35-41.

ACCEPTED MANUSCRIPT Santana, R. C., Perrechil, F. A., Sato, A. C. K., & Cunha, R.L. (2011). Emulsifying properties of collagen fibers: Effect of pH, protein concentration and homogenization pressure. Food Hydrocolloids, 25, 604-612. Schlender, M., Minke, K., Spiegel, B., & Schuchmann, H. P. (2015). High-pressure double stage

T

homogenization processes: Influences of plant setup on oil droplet size. Chemical Engineering

IP

Science, 131, 162-171.

CR

Silva, P. I., Stringheta, P. C., Teófilo, R. F., & Nolasco de Oliveira, I. R. (2013). Parameter optimization for spray-drying microencapsulation of jaboticaba (Myrciaria jaboticaba) peel

US

extracts using simultaneous analysis of responses. Journal of Food Engineering, 117, 538-544.

AN

Simopoulos, A. (2016). An increase in the omega-6/omega-3 fatty acid ratio increases the risk for obesity. Nutrients, 8, 128.

M

Tonon, R. V. (2009). Secagem por atomização do suco de açaí: influência das variáveis de

ED

processo, qualidade e estabilidade do produto. (PhD Thesis). Brasil: Universidade Estadual de Campinas.

PT

Tonon, R. V., Grosso, C. R. F., & Hubinger, M. D. (2011). Influence of emulsion composition and

CE

inlet air temperature on the microencapsulation of flaxseed oil by spray drying. Food Research International, 44, 282-289.

AC

Tonon, R. V., Pedro, R. B., Grosso, C. R. F., & Hubinger, M. D. (2012). Microencapsulation of flaxseed oil by spray drying: Effect of oil load and type of wall material. Drying Technology: An International Journal, 30 (13), 1491-1501. Waraho, T., Cardenia, V., Decker, E. A., & McClements, D. J. (2010). Lipid oxidation in emulsified food products. In: E. Decker, R. Elias, & D. J. McClements, Oxidation in Foods and Beverages and Antioxidant Applications (pp. 306-343). Massachusetts: Woodhead Publishing.

ACCEPTED MANUSCRIPT Wu, Z., Wu, J., Zhang, R., Shichao, Y., Lu, Q., & Yu, Yuequin. (2018). Colloid properties of hydrophobic modified

alginate: surface tension, -potential, viscosity and emulsification.

Carbohydrate Polymers, 181, 56-62. Zhong, D., Huang, X., Yang, H., & Cheng, R. (2010). New insights into viscosity abnormality of

T

sodium alginate aqueous solution. Carbohydrate Polymers, 81, 948-952.

AC

CE

PT

ED

M

AN

US

CR

oxidation of mackerel oil. Food Chemistry, 100, 800-807.

IP

Zuta, P. C., Simpson, B. K., Zhao, X., & Leclerc, L. (2007). The effect of -tocopherol on the

ACCEPTED MANUSCRIPT

Figure 1. (A) Droplet size distributions and (B) zeta potential of single layered primary emulsions 5,

10,

15 MPa).

T

(4.5% oil 1% WPC pH 7) at different total homogenization pressures (

IP

Figure 2. Effect of pH on zeta potential of 4.5 wt% oil 1 wt% WPC emulsions containing 0 wt%

5-HMD,

US

Figure 3. (A) Droplet size distributions ( 5-LMD,

CR

SA (■) and 0.25 wt% (■).

10-LMD, 10-HMD). and (B) zeta

AN

potential of double layered secondary emulsions containing MD (LMD: low maltodextrin, HMD:

M

high maltodextrin) obtained at different total homogenization pressures (5 and 10 MPa).

ED

Figure 4. Visual appearance of microencapsulated powders obtained at both homogenization pressures and maltodextrin concentrations (A). SEM of spray-dried flaxseed oil microcapsules

CE

PT

containing high maltodextrin concentrations (HMD) prepared at 5 MPa (B) and 10 MPa (C).

Figure 5. Effect of storage temperature (■ -18, ● 4 and ▲20°C) on peroxide values (A) and

AC

TBARS (B) of flaxseed oil powdered microcapsules obtained by Procedure 2.

ACCEPTED MANUSCRIPT Table 1. Combination of homogenization pressures and coating materials to prepare flaxseed oil multilayer emulsions.

WPC (%)

SA (%)

MD (%) 0.8

2:1

7

5-HMD

1:2

0.25

0.8

10-LMD

2:1

0.25

7

10-HMD

1:2

0.25

13

1.5

5

4.5

1

0.25

1

3.5

1.5

5

4.5

1

0.25

1

7

3

10

4.5

1

1

7

3

10

4.5

1

2

7

3

10

5

US

AN M ED PT CE AC

IP

3.5

* LMD (low maltodextrin), HMD (high maltodextrin)

Oil:Wall ratio

5-LMD

1

1

Condition Code*

T

Flaxseed oil (%)

CR

Procedure

Homog. Pressure (MPa) Stage Stage 2 Total 1

1:3

ACCEPTED MANUSCRIPT

Table 2. Effect of different microencapsulation stages on peroxide value (PV) and thiobarbituric acid reactive substances index (TBARS) of flaxseed oil.

TBARS (mmol/Kg oil)

Emulsion

Powder

Initial Oil

Emulsion

Powder

5-LMD

1.27±0.13c

2.50±0.41b

5.63±0.41a

0.15±0.01ª

0.11±0.00b

0.19±0.04a,b

5-HMD

0.40±0.18c

1.48±0.18b

3.46±0.02a

0.09±0.01b

0.14±0.04b

0.33±0.01a

10-LMD

1.04±0.18b

2.37±0.68b

4.27±0.07a

0.09±0.03ª

0.11±0.01ª

0.34±0.15ª

10-HMD

1.03±0.18c

4.31±0.03b

6.44±0.70a

0.23±0.10ª

0.13±0.06ª

0.33±0.09ª

US

CR

IP

Initial Oil

T

PV (meq/Kg oil) Sample

AN

Values represent the average ± standard deviation of replicates. Mean values with different letters were

M

significantly different (horizontal comparison) when LSD test was applied (p < 0.05)

ED

Table 3. Influence of maltodextrin concentration and homogenization pressure on water activity

0.43±0.07a,b

AC

5-LMD

aw

CE

Condition

PT

(aw), Free Oil (FO), Total Oil (TO) and Encapsulation Efficiency (EE) of powdered microcapsules.

FO (g oil/100 g)

TO (g oil/100 g)

EE (%)

47.74±0.86ª

68.88±0.06ª

30.69±1.31ª

10-LMD

0.45±0.01a

41.90±1.02b

69.10±0.19ª

34.39±7.22b

5-HMD

0.27±0.02c

17.22±1.46c

35.50±0.10b

50.99±3.27c

10-HMD

0.37±0.01b

14.24±0.78d

35.49±0.17b

59.88±1.04d

Values represent the average ± standard deviation of replicates. Mean values with different letters were significantly different (vertical comparison) when LSD test was applied (p < 0.05).

ACCEPTED MANUSCRIPT

Table 4. Properties of flaxseed oil microcapsules prepared by Procedure 2.

Measured Properties

Procedure 2 44.95±0.26

T

Zeta Potential of emulsion (mV)

IP

Free Oil (g oil/100 g powder)

CR

Total Oil (g oil/100 g powder) Encapsulation Efficiency (%)

US

PV initial oil (meq/Kg oil) TBARS initial oil (mmol/Kg oil)

AN

PV spray-dried powder (meq/Kg oil)

TBARS spray-dried powder (mmol/Kg oil)

3.91±0.36 26.68±0.03 84.39±1.32 1.84±0.12 0.14±0.07 6.84±0.34

less than 0.01

M

Values represent the average ± standard deviation of replicates.

ED

HIGHLIGHTS

PT

 Emulsion layer-by-layer deposition technology was used to encapsulate flaxseed oil.

CE

 Effect of low homogenization pressure and maltodextrin (MD) content was studied.  Microencapsulation process gradually increased peroxide value (PV) of the oil.

AC

 Powders did not exceed maximum PV allowed for cold pressed oils during storage.

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5