Journal of Food Engineering 115(2013)443-451 Contents lists available at SciVerse ScienceDirect ood enne Journal of Food Engineering ELSEVIER journal homepage:www.elsevier.com/locate/jfoodeng Encapsulation efficiency and oxidative stability of flaxseed oil microencapsulated by spray drying using different combinations of wall materials Helena C.F.Carneiro,Renata V.Tononb,Carlos R.F.Grosso,Miriam D.Hubinger* Departmentof Food and Nutrition Focuy of Food Engineering University of Campinas,Campinas.SP.Brazil ARTICLE INFO ABSTRACT This study aimed at evaluating the potential of maltodextrin combination with different wall materials in the microencapsulation of flaxseed oil by spray drying.in order to maximize encapsulation efficiency and Maltodextrin (MD)was mixed with gum Arabic (CA).whey protein concenta ed starch(Hi-Cap】 zad for and dr The bes encapsulation efficiency was obtained for MD:Hi-Cap followed by the MD:Capsul combination,while the lowest encapsulation efficiency was obtained for MD:WPC.which also showed poorer emulsion h the active materal em on was th 2012 Elsevier Ltd.Open aess under the Elsevier OAlicense 1.Introduction low water activity.easir handiand storage andalso protects th There is an increasing demand for nutritive and healthy foods in selection and emulsion properties (stability.viscosity and droplet the market and this fact has led the food industry to focus their size)can affect the process efficiency and the microencapsulated research in products of this nature.Flaxseed oil is a polyunsatu product stability.A successful microencapsulation must result in rated oil extracted from the flax plant (Linum usitatissimim)rich powder with minimum surface oil and maximum retention of in o-linolenic acid(ALA).the essential fatty acid omega()-3. he active material which represents about 57%of its total fatty acids.The high con- Gum Arabic is one of the most common wall materials used in tent of n-3 fatty acid present in this oil allows the attribution of microencapsulation by spray drying.Although it presents many functional food,which means that besides the nutritional func- desirable characteristics to be a good encapsulating agent(high sol tions.its consumption may have beneficial effects on health ubility.low viscosity and good emulsifying properties).the oscilla- (Vaisey-Genser and Morris,2003). tion in supply.as well as the increasing prices,is leading researches On the other hand,one of the major problems associated with to look for alternative wall materials that could replace it or be used oils rich in polyunsaturated fatty acids(PUFAs)is their high sus- ceptibility to oxidative deterioration and consequent production of undesirable flavor.Thus,there is a need to protect these oils in n e and aste,low visc idely used for on aga xidatio However.the biggest problem 2 ma vith good quality its low em tive bio odified stm tal 2008:Bule t et al,2003:Bule 2010) sp892Tt45s1995219036r5193521402 E-mail addres The selection of wall material combinations affects both the unicm(MDe emulsion properties and the particles'characteristics after drying
Encapsulation efficiency and oxidative stability of flaxseed oil microencapsulated by spray drying using different combinations of wall materials Helena C.F. Carneiro a , Renata V. Tonon b , Carlos R.F. Grosso c , Míriam D. Hubinger a,⇑ aDepartment of Food Engineering, Faculty of Food Engineering, University of Campinas, Campinas, SP, Brazil b Embrapa Food Technology, Rio de Janeiro, RJ, Brazil cDepartment of Food and Nutrition, Faculty of Food Engineering, University of Campinas, Campinas, SP, Brazil article info Article history: Available online 3 April 2012 Keywords: Microencapsulation Spray drying Flaxseed oil Wall material Encapsulation efficiency Oxidative stability abstract This study aimed at evaluating the potential of maltodextrin combination with different wall materials in the microencapsulation of flaxseed oil by spray drying, in order to maximize encapsulation efficiency and minimize lipid oxidation. Maltodextrin (MD) was mixed with gum Arabic (GA), whey protein concentrate (WPC) or two types of modified starch (Hi-Cap 100TM and Capsul TA) at a 25:75 ratio. The feed emulsions used for particle production were characterized for stability, viscosity and droplet size. The best encapsulation efficiency was obtained for MD:Hi-Cap followed by the MD:Capsul combination, while the lowest encapsulation efficiency was obtained for MD:WPC, which also showed poorer emulsion stability. Particles were hollow, with the active material embedded in the wall material matrix, and had no apparent cracks or fissures. During the oxidative stability study, MD:WPC combination was the wall material that best protected the active material against lipid oxidation. 2012 Elsevier Ltd. 1. Introduction There is an increasing demand for nutritive and healthy foods in the market and this fact has led the food industry to focus their research in products of this nature. Flaxseed oil is a polyunsaturated oil extracted from the flax plant (Linum usitatissimim) rich in a-linolenic acid (ALA), the essential fatty acid omega (x)-3, which represents about 57% of its total fatty acids. The high content of n-3 fatty acid present in this oil allows the attribution of functional food, which means that besides the nutritional functions, its consumption may have beneficial effects on health (Vaisey-Genser and Morris, 2003). On the other hand, one of the major problems associated with oils rich in polyunsaturated fatty acids (PUFAs) is their high susceptibility to oxidative deterioration and consequent production of undesirable flavor. Thus, there is a need to protect these oils in order to make them more stable during handling, processing and storage (Augustin et al., 2006). Spray drying is a process widely used for microencapsulation of oils and flavours (Fuchs et al., 2006; Ahn et al., 2008; Bae and Lee, 2008; Partanen et al., 2008). It results in powders with good quality, low water activity, easier handling and storage and also protects the active material against undesirable reactions. Both wall material selection and emulsion properties (stability, viscosity and droplet size) can affect the process efficiency and the microencapsulated product stability. A successful microencapsulation must result in a powder with minimum surface oil and maximum retention of the active material. Gum Arabic is one of the most common wall materials used in microencapsulation by spray drying. Although it presents many desirable characteristics to be a good encapsulating agent (high solubility, low viscosity and good emulsifying properties), the oscillation in supply, as well as the increasing prices, is leading researches to look for alternative wall materials that could replace it or be used in combination with it (Charve and Reineccius, 2009). Maltodextrin is a hydrolyzed starch commonly used as wall material in microencapsulation of food ingredients (Gharsallaoui et al., 2007). It offers advantages such as relatively low cost, neutral aroma and taste, low viscosity at high solids concentrations and good protection against oxidation. However, the biggest problem of this wall material is its low emulsifying capacity. Therefore, it is desirable to use maltodextrin in combination with other surface active biopolymers, such as gum Arabic (Fernandes et al., 2008; Bule et al., 2010), modified starches (Soottitantawat et al., 2003; Bule et al., 2010) and proteins (Hogan et al., 2003; Bae and Lee, 2008) in order to obtain an effective microencapsulation by spray drying. The selection of wall material combinations affects both the emulsion properties and the particles’ characteristics after drying 0260-8774 2012 Elsevier Ltd. http://dx.doi.org/10.1016/j.jfoodeng.2012.03.033 ⇑ Corresponding author at: Faculty of Food Engineering – UNICAMP, 80 Monteiro Lobato Street, Postal code 13083-862, Cidade Universitaria Zeferino Vaz, Campinas/ SP, Brazil. Tel.: +55 19 3521 4036; fax: +55 19 3521 4027. E-mail addresses: htina01@fea.unicamp.br (H.C.F. Carneiro), renata.tonon@ ctaa.embrapa.br (R.V. Tonon), grosso@fea.unicamp.br (C.R.F. Grosso), mhub@fea. unicamp.br (M.D. Hubinger). Journal of Food Engineering 115 (2013) 443–451 Contents lists available at SciVerse ScienceDirect Journal of Food Engineering journal homepage: www.elsevier.com/locate/jfoodeng Open access under the Elsevier OA license. Open access under the Elsevier OA license.
H.CF. nal of Food E mg115(2013到443-451 and during It is well described that emulsion e measured after 24 h.The stability was mea as well as (1) actoote und in the literature or n mi found in literature have reporte the micr 232.Emulsic ()( (Tonon et).Therefore,even though they could show asim will present the same behavior during spray 25Cby a Peltier n is lable on m lation r et a on et al enttypes of wall the 2.3.3.Emulsion droplet siz The objective ofthiswork as to evaluate the as for microe capsulation of nax during each measurem ntil successive readingsbec mean diameter. t Si le the microcapsule partide size,bulk density,morphology and oxidative stability. encapsulation by spray drying 2.Material and methods 2.1.Material chamber of mm.The emulsions were fed into the mai Flaxseed oil (Paranambi)was used 457%18:0.21.11%18:1.1430%18: and 53.3518:3.The wal /h and re180±2and1io±2C,resp s.Mogi Guacu.Brazil (MD).whe feed flow rate was 12 t2 g/min 2.5.Powders analysis no sul TA (de (derived from waxy maize)(National Starch.ao 2 51 Encapsulation efficienc 22.Emulsion preparation hexane were added powder n glass jar with alid The wall materials were e added to distilled waterat25Cand 2 min. mixture on the filter was rinsed three times with 20mLof hexane.Ther 2008 and Reineccius.200 d Brazil)operating at 18.000 pm for5min. 20M 200b)Total oil was obe equal to the initial oil since which was exp ted,since flaxs oil is not efficiency(EE)was calcula ots of E=(。9)×10 where to is the total oil content and so is the surface oil content
and during storage. It is well described that emulsion characteristics such as stability, viscosity, droplets size, as well as powder properties such as surface oil, particle size, density, morphology and oxidative stability, are influenced by the type of encapsulating agent used (Jafari et al., 2008b). Most of the works found in the literature on microencapsulation of PUFA-rich oils uses fish oil as active material (Hogan et al., 2003; Augustin et al., 2006; Serfert et al., 2009; Anwar and Kunz, 2011). Fish oil and flaxseed oil are very rich in polyunsaturated fatty acids, but they do not have the same composition. Some works found in literature have reported the microencapsulation of fish oil containing approximately 27% (Serfert et al., 2009), 29% (Aghbashlo et al., 2012), or 33% (Drusch et al., 2009b) of omega-3 fatty acids, while the flaxseed oil contains bout 53% linolenic acid (Tonon et al., 2011). Therefore, even though they could show a similar tendency when microencapsulated, it is not possible to assure they will present the same behavior during spray drying and storage. Very little information is available on microencapsulation of flaxseed oil (Partanen et al., 2008; Omar et al., 2009; Tonon et al., 2011) and none of the published works reported the influence of different types of wall materials on the encapsulation efficiency and oxidative stability of this oil. The objective of this work was to evaluate the potential of maltodextrin combination with four types of wall materials (gum Arabic, whey protein concentrate and two types of modified starches), as alternative materials for microencapsulation of flaxseed oil by spray drying. The feed emulsions were characterized for stability, viscosity and droplet size, while the microcapsules were characterized for encapsulation efficiency, moisture content, particle size, bulk density, morphology and oxidative stability. 2. Material and methods 2.1. Material Flaxseed oil (Lino Oil, Paranambi, Brazil) was used as active material with the following fatty acid composition: 5.77% 16:0, 4.57% 18:0, 21.11% 18:1, 14.30% 18:2 and 53.35% 18:3. The wall materials used were: maltodextrin MOR-REX 1910 with 10 DE (Corn Products, Mogi Guaçu, Brazil) (MD), whey protein concentrate WPC 80 (Alibra Ingredients, Campinas, Brazil) (WPC), gum Arabic Instantgum BA (Colloids Naturels CNI, São Paulo, Brazil) (GA) and two chemically n-octenyl succinic andydrid (OSAN)-modified starches: Capsul TA (derived from Tapioca starch) and Hi-Cap 100TM (derived from waxy maize) (National Starch, São Paulo, Brazil). 2.2. Emulsion preparation The wall materials were added to distilled water at 25 C and the mixture was stirred until completely dissolved. The total solid concentration (wall material + oil) was fixed at 30%. Flaxseed oil was then added to the wall material solution at a concentration of 20% with respect to total solids (Ahn et al., 2008; Jafari et al., 2008a; Charve and Reineccius, 2009). Emulsions were formed using an Ultra-Turrax homogenizer MA-102 (Marconi, Piracicaba, Brazil) operating at 18,000 rpm for 5 min. 2.3. Emulsion characterization 2.3.1. Emulsion stability Immediately after the emulsion preparation, 25 mL aliquots of each sample were transferred to graduated cylinders of 25 mL, sealed, stored at room temperature for one day, and the volume of the upper phase measured after 24 h. The stability was measured by % of separation and expressed as: %Separation ¼ H1 H0 100 ð1Þ Where: Ho represents the emulsion initial height and H1 is the upper phase height. 2.3.2. Emulsion viscosity Emulsion viscosity was measured thought the determination of steady-shear flow curves using a Physica MCR301 Rheometer (Anton Paar, Graz, Austria). Measurements were made in triplicate, using stainless steel plate-plate geometry with a diameter of 75 mm and a gap of 0.2 mm. Temperature was controlled at 25 C by a Peltier system. Rheograms were analyzed according to empirical models and the emulsions viscosity was calculated as the relationship between shear stress and shear rate. 2.3.3. Emulsion droplet size The droplet size distribution was measured using a laser light diffraction instrument, Mastersizer S (Malvern Instruments, Malvern, UK). A small sample was suspended in water using magnetic agitation, and the droplet size distribution was monitored during each measurement until successive readings became constant. The emulsion droplet size was expressed as D32, the Sauter mean diameter. 2.4. Microencapsulation by spray drying Spray drying process was performed in a laboratory scale spray dryer Lab Plant SD-05 (Huddersfield, England), with a nozzle atomization system with 0.5 mm diameter nozzle and main spray chamber of 500 215 mm. The emulsions were fed into the main chamber through a peristaltic pump and the feed flow rate was controlled by the pump rotation speed. Drying air flow rate was 73 m3 /h and compressor air pressure was 0.06 MPa. Inlet and outlet air temperature were 180 ± 2 and 110 ± 2 C, respectively, and feed flow rate was 12 ± 2 g/min. 2.5. Powders analysis 2.5.1. Encapsulation efficiency Encapsulation efficiency (EE) was determined according to the method described by Bae and Lee (2008). Fifteen milliliters of hexane were added to 1.5 g of powder in a glass jar with a lid, which was shaken by hand for the extraction of free oil, during 2 min, at room temperature. The solvent mixture was filtered through a Whatman filter paper n 1 and the powder collected on the filter was rinsed three times with 20 mL of hexane. Then, the solvent was left to evaporate at room temperature and after at 60 C, until constant weight. The non-encapsulated oil (surface oil) was determined by mass difference between the initial clean flask and that containing the extracted oil residue (Jafari et al., 2008b). Total oil was assumed to be equal to the initial oil, since preliminary tests revealed that all the initial oil was retained, which was expected, since flaxseed oil is not volatile. Encapsulation efficiency (EE) was calculated from Eq. (2). EE ¼ TO SO TO 100 ð2Þ where TO is the total oil content and SO is the surface oil content. 444 H.C.F. Carneiro et al. / Journal of Food Engineering 115 (2013) 443–451
HCF.Carneiro/Engineering 115(201)443-45 44 26.Statistical analysi 3.Results and discus red using a laser light 3.1.Emulsion characterization uspended inethyl alcohol (99 ive readings ume weighted mean diameter. slze was exp as Das.the vo 2.5.5.Scanning electron microscopy (SEM) Microcaps on layer and om phase According to copy.Oxf nd)operating a hich would enhan cimens (stubs)of 12 mm diam ron (V ace may ndary and t netalliza ues wh cqSitiomthagaenicatio by the L version 3.01. ered to in this unusual beh that the fp stability of pr o de ace d before.emu ion viscosity was determined throu the ling to which visc 5 e ented ad heolog at p character ccording to the IDF standard alue (m)inferior to( viscosity of emulsions produced epared with MD: nbed by Shantha and Dec e with long chains, can b ofdsonsiblef 25.6.2.Heads Heads gas c epodtcio of and ar on of em ns prepared with different types of wall material Dx(um) MD:Hi-Ca Ana 16.8±001 2.11±0.01 ith rgas was helium(99),used at a flov ure was 200
2.5.2. Moisture content Powders’ moisture content was determined gravimetrically by drying in a vacuum oven at 70 C until constant weight (AOAC, 2006). 2.5.3. Bulk density For determination of bulk density, 2 g of powder were transferred to a 50 mL graduated cylinder. Packed bulk density was calculated from the height of powder in the cylinder after being tapped by hand on a bench 50 times from a height of 10 cm (Goula and Adamopoulos, 2004). 2.5.4. Particle size distribution Particle mean diameter was measured using a laser light diffraction instrument, Mastersizer S (Malvern Instruments, Malvern, UK). A small sample was suspended in ethyl alcohol (99.9%) using magnetic agitation, and the particle size distribution was monitored during each measurement until successive readings became constant. The particle size was expressed as D43, the volume weighted mean diameter. 2.5.5. Scanning electron microscopy (SEM) Microcapsules were observed in a Scanning Electron Detector microscope with Energy Dispersive X-ray, SEM and EDX Leo 440i 6070 (LEO Electron Microscopy, Oxford, England) operating at 15 kV and electron beam current of 100 pA. The samples were fixed directly on door-metallic specimens (stubs) of 12 mm diameter and then subjected to metallization (sputtering) with a thin layer of gold/palladium in a Sputter Coater SC7620 polaron (VG Microtech, England) at a coverage rate of 0.51 Å/s for 180 s, with a current of 3.5 mA, 1 V and 2 102 Pa. After metallization, the samples were observed with magnifications of 4000, 5000 and 10,000. Image acquisition was performed by the LEO software, version 3.01. 2.5.6. Oxidative stability For the stability tests, the microcapsules were sealed in a glass vial (20 mL), stored at 45 C in order to accelerate the oxidation process, and evaluated for oxidation by two distinct methods (peroxide value and headspace analysis) at time zero (right after drying) and over four weeks of storage. 2.5.6.1. Peroxide value. The oil extraction was performed according to Partanen et al. (2008). The peroxide value determination was carried out spectrophotometrically according to the IDF standard method 74A:1991 using the Unico 2800UV/VIS spectrophotometer (United Products & Instruments Inc., New Jersey, USA). Measurements were performed in triplicate. Hydroperoxide concentrations were determined using a Fe+3 standard curve with iron concentration varying from 1 to 24 lg, as described by Shantha and Decker (1994). 2.5.6.2. Headspace analysis. Headspace gas chromatography was used to determine the production of propanal and hexanal, considered as indicators of flaxseed oil oxidation (Augustin et al., 2006; Boyde et al., 1992; Jimenez et al., 2006; Drusch et al., 2007). Powder samples (1 g) of each combination were placed into an amber glass vial (20 mL) and stored under the same conditions previously described. To quantify the propanal and hexanal production, new and fresh samples were dopped with a known amount of aldehydes (purity standards known) in order to built a calibration curve. Analyses were made in a gas chromatograph type CombiPal Headspace – CTC Analytics Shimadzu – model GC-2010 (Kyoto, Japan), with a Rtx-01, 100% dimethyl polysiloxane column (30 m 0.32 mm, 3.0 mm film). The carrier gas was helium (99.999%), used at a flow rate of 1.0 mL/min. The injector temperature was 200 C, split mode (1:30). Finally, the detector was flame ionization and temperature used was 250 C. Analyses of headspace were performed in triplicate. 2.6. Statistical analysis Results were statistically analysed by Analysis of Variance, using the software Statistica 8.0 (Statsoft Inc., Tulsa, USA). Mean analysis was performed using Duncan’s procedure at p 6 0.05. 3. Results and discussion 3.1. Emulsion characterization The percentage of separation and the droplets mean diameter (D32) observed in the emulsions produced with different types of wall materials are shown in Table 1. The stability study revealed that most of the emulsions were kinetically stable, with exception of those prepared with WPC and maltodextrin, which showed the formation of a small separation layer and a foam phase, 24 h after its homogenization. This was unexpected, since whey proteins are well known by their good emulsifying capacity. According to Dickinson and Matsumura (1991) this result may have been caused by the unfolding of the protein molecules at the droplets surface, which would enhance protein–protein interaction leading to flocculation during emulsifi- cation and consequently reducing the emulsion stability. The unfolding of protein molecules of the oil–water interface may lead to changes in secondary and tertiary structure, and consequently exposure of their residues which would be linked (–S–S– linkages or disulphide linkages) within the native globular structure leading to the formation of intermolecular interaction at the oil–water interface and flocculating. Another hypothesis that can be considered to explain this unusual behaviour is that the stability of protein-stabilized emulsions is a function of pH and other parameters. So, depending on the emulsions’ pH, the emulsifying capacity of WPC may have been lower than usual (Huynh et al., 2008), affecting the emulsion stability. As stated before, emulsion viscosity was determined through steady-shear flow curves. Experimental data were better adjusted by the Newtonian model, according to which viscosity is constant with shear rate. The MD:Capsul mixture was the only one that presented a different rheological behaviour, being characterized as non-Newtonian (Power-law) fluid, with flow behaviour index value (n) inferior to 1 (0.830). The viscosity of emulsions produced with different wall materials combination is shown in Fig. 1. The emulsion prepared with MD:GA showed the highest viscosity, followed by that prepared with MD:Capsul combination. Gum Arabic is generally used as a thickening agent in foodstuffs, showing a ramified structure with long chains, which can be responsible for its higher viscosity. Despite the greater difficulty of dissolution of Hi-Cap during the initial solution preparation, the emulsions prepared with this wall material and maltodextrin showed the lowest Table 1 Characterization of emulsions prepared with different types of wall materials. Formulation % Separation D32 (lm) MD:Hi-Cap – 2.11 ± 0.01b MD:WPC 16.8 ± 0.01 2.10 ± 0.02b MD:GA – 1.73 ± 0.02c MD: Capsul – 2.19 ± 0.03a Different letters indicate significant difference between samples at p 6 0.05. (MD: Maltodextrin, Hi-Cap: Hi-Cap 100, WPC: Whey Protein Concentrate, GA: Gum Arabic, Capsul: Capsul TA). H.C.F. Carneiro et al. / Journal of Food Engineering 115 (2013) 443–451 445
46 HCF.Cmer f Food ineering 115()43-451 0.100 3.Encapsulation CapsulMD encapsulation effceyom62.3 to es .010 0 50 100 150 200 250300 Shear rate (s") s(2009)0 d t stud .ther ed oil and the differ rials combinati a2a8品co2nee ent retention properties and flm-forming capacity. 3.3.Particle characterization ed to the other Table2presents the f particles prepared with the biggest droplets w n com hou the mixture of maltodextrin and Capsul has show n in bulk densit ed only by the of each type of material. pared to products with lower densities.Moreover,higher bulk 。-MD:GA 9 01 Hi-Cap/MD WPC/MD GAMD Fig.2.Droplets size distri
viscosity values compared to other combinations. Soottitantawat et al. (2003), Fernandes et al. (2008), Bule et al. (2010) also verified that mixtures of gum Arabic and maltodextrin presented higher emulsion viscosity when compared to emulsions prepared with maltodextrin and other polymers. The homogenization of flaxseed oil and the different wall materials combinations resulted in emulsions with droplet diameter ranging from 0.6 to 26 lm. According to Fig. 2, most of the curves presented a monomodal distribution with one peak representing a predominant size, with exception of the emulsion prepared with MD:Capsul, which showed a bimodal distribution. The use of different wall materials had significant influence on emulsions droplet size (Table 1). The emulsion prepared with MD:Capsul had the biggest droplets when compared to the other materials, while the emulsion prepared from the MD:GA mixture showed the smallest ones. This last result can be related to the highest viscosity presented by the MD:GA emulsion, which implies in a greater resistance to droplets movement, avoiding coalescence and resulting in smaller diameters. Although the mixture of maltodextrin and Capsul has shown higher viscosity in comparison to those produced with WPC and Hi-Cap, these last two samples showed smaller droplet mean diameters, indicating that droplets size was not affected only by the emulsion viscosity, but also by the intrinsic emulsifying properties of each type of material. 3.2. Encapsulation efficiency According to Fig. 3, the encapsulation efficiency of samples was significantly influenced by the type of wall material used, since emulsions prepared with Hi-Cap resulted in particles with considerably lower surface oil than those prepared with the other ones. The encapsulation efficiency values varied from 62.3% to 95.7%, being the lowest value obtained for MD:WPC. Analyzing the results, these two combinations (MD:WPC and MD:Hi-Cap) had similar rheology and droplet mean diameter characteristics. However, the emulsion containing MD:WPC showed the poorest stability, which may have influenced on the encapsulation efficiency of its resulting powders. According to Barbosa et al. (2005), the more stable the emulsion, the higher the encapsulation efficiency is, i.e., the lower the amount of nonencapsulated material on particles surface. Charve and Reineccius (2009) obtained a similar result studying the volatile retention in microcapsules prepared by spray drying, where the microencapsulated particles produced with modified starch showed higher oil retention when compared to the particles encapsulated with gum Arabic and with whey protein. Many studies have shown that the reduction of emulsion droplets size, which generally represents an increased stability, results in greater retention of active material (Liu et al., 2001; Soottitantawat et al., 2005; Jafari et al., 2008a). In the present study, the results obtained for encapsulation efficiency could not be related to the emulsions droplet size or viscosity and the differences between them can probably be attributed to the differences between the polymer matrices formed by each one of the wall materials used, which have different retention properties and film-forming capacity. 3.3. Particle characterization Table 2 presents the characterization of particles prepared with different wall materials. In general, samples did not show signifi- cant differences in moisture content when different wall materials were used. Only the MD:Capsul combination had a moisture content value slightly lower than the others. Hogan et al. (2001) observed moisture content values from 1% to 3% in soybean oil microencapsulated by spray-drying and these values were not affected by the type of wall material as well. Microcapsules showed significant variation in bulk density, according to the type of wall material used. Bulk density values ranged from 0.28 (MD:WPC) to 0.40 g/cm3 (MD:GA), being gum Arabic the material that resulted in the most dense particles. The advantage of obtaining powders with higher density is that they can be stored in large amounts into smaller containers when compared to products with lower densities. Moreover, higher bulk 0.010 0.100 0 50 100 150 200 250 300 Shear rate (s-1) Viscosity (Pa.s) Hi-Cap/MD WPC/MD GA/MD Capsul/MD Fig. 1. Emulsions viscosity as a function of shear rate. (MD: Maltodextrin, Hi-Cap: Hi-Cap 100, WPC: Whey Protein Concentrate, GA: Gum Arabic, Capsul: Capsul TA). 0 3 6 9 12 15 0.01 0.1 1 10 100 1000 Droplet size (µm) Volume (%) MD:Hi-Cap MD:WPC MD:GA MD:Capsul Fig. 2. Droplets size distribution of emulsions prepared with different wall materials. (MD: Maltodextrin, Hi-Cap: Hi-Cap 100, WPC: Whey Protein Concentrate, GA: Gum Arabic, Capsul: Capsul TA). 0 10 20 30 40 50 60 70 80 90 100 Hi-Cap/MD WPC/MD GA/MD Capsul/MD Encapsulation Efficiency (%) a d c b Fig. 3. Encapsulation efficiency of powders produced with different wall materials. Different letters indicate significant difference between samples at p 6 0.05. (MD: Maltodextrin, Hi-Cap: Hi-Cap 100, WPC: Whey Protein Concentrate, GA: Gum Arabic, Capsul: Capsul TA). 446 H.C.F. Carneiro et al. / Journal of Food Engineering 115 (2013) 443–451
44 FormulationMoisture content Bulk density(g/cm)D(um) 1.65±0.160 035±0021 19.79±0.05 145000 combinatio 1.11±011 pheres with ere foun at et al(0 ng t intemal morphology.all e ho d to th op he wa materials matrix It is an r char the final stages of ment with the results ed the bi st num 35.Lipid oxidation ing a predominant used a mhinations have resulted i ide.the with th hese partic (Laine 12)found T values of were 2010) trinand whey protein concentrate. .Maltode xtrin had a of ap roximately 110 3.4.Powder morphology though there is ure fo nal)of 4s 12%fm isture conter a gy,part rious sizes with n herhan50cahteaotepotodwheathad517%ofmo ure cont .WPC/MD GA/MD mples encapsulated w Capsul/MD 的 /kg oi e continued to oxidize and in the fourth week.MD:GA and 0.0 01 100 Particle size（um (006)encapsulatedc gated ces as wa als
density may indicate lower amount of air occluded in the spaces between particles, which can help to prevent lipid oxidation. Particle mean diameters varied from 17.98 to 23.03 lm. The microcapsules produced from mixtures of maltodextrin and gum Arabic showed greater size, probably due their higher emulsion viscosity. According to Masters (1991), the atomized droplet size varies directly with emulsion viscosity at a constant atomization speed. The higher the emulsion viscosity, the larger are the droplets formed during atomization, and therefore, the larger are the powdered particles obtained. This is in agreement with the results published by Hogan et al. (2001) for fish oil encapsulated using sodium caseinate and carbohydrates of varying dextrose equivalence (DE) as wall materials. Fig. 4 shows the particle size distribution of powders produced with different combinations of wall materials. The particles exhibited a very large size range, with diameters varying from 0.02 to 160.0 lm, approximately, and showed a bimodal distribution with two distinct peaks, each one representing a predominant size. This is particularly interesting in the case of powders, once the ‘‘population’’ of smaller particles can penetrate into the spaces between the larger ones, thus occupying less space. Although all the wall materials combinations have resulted in similar particle size distributions, the powders produced with the MD:WPC mixture showed a wider distribution, which means that these particles were less homogeneous. It can also be related to the lower stability of the feed emulsions produced with maltodextrin and whey protein concentrate. 3.4. Powder morphology Fig. 5 shows the SEM microstructures (internal and external) of powders produced with different wall material combinations. Observing the external morphology, particles showed a spherical shape and various sizes with no apparent cracks or fissures, which is an advantage, since it implies that capsules have lower permeability to gases, increasing protection and retention of the active material. Moreover, the variety in size is a typical characteristic of particles produced by spray drying. The mixtures of different wall materials influenced on microparticles morphology. This is more evident when comparing the images from the combination of MD:Hi-Cap with the others, since this mixture resulted in microspheres with smoother surface and fewer teeth or roughness. According to Ré (1998), such imperfections (teeth) are formed when there is a slow process of film formation during drying of the atomized droplets, associating the presence of surface depressions to the collapse suffered by the droplets during the initial stages of drying. Similar morphological characteristics were found by Tonon et al. (2011) and Soottitantawat et al. (2005)). Analysing the internal morphology, all microspheres were hollow and the active material was adhered to the surface as small droplets embedded in the wall materials matrix. It is another characteristic of particles obtained by spray drying. This emptiness is a result of the quick particles expansion during the final stages of drying (Jafari et al., 2008b). According to Fig. 5, the particles that showed the biggest number of droplets were obtained by the MD:Hi-Cap combination. This mixture also presented the best encapsulation efficiency (96%, approximately) which means that more active material was encapsulated and embedded in the wall material matrix, as mentioned before. 3.5. Lipid oxidation The oxidative stability of flaxseed oil encapsulated into the different wall materials combinations was evaluated by two different methods (peroxide value and headspace gas chromatography), during four weeks of storage, at 45 C. All the powders were in the glassy state at this temperature, since the materials used as carrier agents were expected to have glass transition temperatures higher than 45 C. Gum Arabic was reported to show a Tg equal to 92 C when stored at aw of 0.11, and 62 C when stored at aw of 0.24 (Laine et al., 2010). García et al. (2012) found Tg values of 53 C and 51 C for WPC with 9.6% and 10% of moisture content, respectively. Maltodextrin 10DE had a Tg of approximately 110 C at aw of 0.1 (Roos, 1995). Regarding Capsul (derived from tapioca starch) and Hi-Cap (derived from waxy starch), although there is no data in the literature for these specific modified starches, Tran et al. (2007) found Tg values of 94 C and 88–92 C for native and modified tapioca starches with 12% of moisture content, while Yuan and Thompson (1994) showed that Tg of waxy starch was higher than 50 C (values not reported) when it had 5.17% of moisture content. 3.5.1. Peroxide value Peroxide value variations of flaxseed oil encapsulated with different wall materials are shown in Fig. 6. At time zero, all samples showed a low level of oxidation, ranging from 6.12 to 8.77 meq peroxide/kg oil. The samples encapsulated with MD:HiCap and MD:GA presented higher peroxide concentration after one week, reaching values of 22.6 and 24.8 meq peroxide/kg oil, respectively. Samples encapsulated with Hi-Cap, GA or Capsul with maltodextrin suffered a significant increase in oxidation at the third week of storage. Between the third and fourth weeks, samples continued to oxidize and in the fourth week, MD:GA and MD:Capsul did not differ from each other. Powders oxidative stability was strongly influenced by the wall material combination. Although the MD:WPC mixture has shown the lowest encapsulation efficiency, this combination presented the highest oil protection against oxidation, while MD:GA and MD:Capsul showed the poorest oxidative stability. Jimenez et al. (2006) encapsulated conjugated linoleic acid (CLA) using polymeric matrices as wall materials and also studied the microparticles Table 2 Characterization of particles prepared with different wall materials. Formulation % Moisture content Bulk density (g/cm3 ) D43 (lm) MD:Hi-Cap 1.65 ± 0.160a 0.35 ± 0.027b 19.79 ± 0.05b MD:WPC 1.54 ± 0.060a 0.28 ± 0.010c 17.98 ± 0.88c MD:GA 1.45 ± 0.050a 0.40 ± 0.012a 23.03 ± 0.31a MD: Capsul 1.11 ± 0.110b 0.36 ± 0.023b 15.32 ± 0.01d Different letters indicate significant difference between samples at p 6 0.05. (MD: Maltodextrin, Hi-Cap: Hi-Cap 100, WPC: Whey Protein Concentrate, GA: Gum Arabic, Capsul: Capsul TA). 0 2 4 6 8 10 0.01 0.1 1 10 100 1000 Particle size (µm) Volume (%) Hi-Cap/MD WPC/MD GA/MD Capsul/MD Fig. 4. Particle size distribution of powders produced with different wall materials combinations. (MD: Maltodextrin, Hi-Cap: Hi-Cap 100, WPC: Whey Protein Concentrate, GA: Gum Arabic, Capsul: Capsul TA). H.C.F. Carneiro et al. / Journal of Food Engineering 115 (2013) 443–451 447