at SciVerse ScienceDirect

Food Hydrocolloids 33 (2013) 225e233

Contents lists available

Food Hydrocolloids

journal homepage: www.elsevier .com/locate/ foodhyd

Effect of relative humidity on the store stability of spray-dried beta-carotene nanoemulsions

Rong Liang a,b, Qingrong Huang c, Jianguo Ma a, Charles F. Shoemaker b,**, Fang Zhong a,*

aKey Laboratory of Food Colloids and Biotechnology, Ministry of Education, School of Food Science and Technology, Jiangnan University, Wuxi 214122, PRChinabDepartment of Food Science and Technology, University of California, Davis, CA 95616, USAcDepartment of Food Science, Rutgers University, 65 Dudley Road, New Brunswick, NJ 08901, USA

a r t i c l e i n f o

Article history:Received 23 November 2012Accepted 27 March 2013

Keywords:Beta-caroteneModified starchRetentionFilm forming propertyGlass transition temperature

* Corresponding author. Tel.: þ86 510 85328307; fa** Corresponding author.

E-mail addresses: cfshoemaker@ucdavis.edu (jiangnan.edu.cn (F. Zhong).

0268-005X/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.foodhyd.2013.03.015

a b s t r a c t

In order to overcome the limitations of liquid-base emulsion system, beta-carotene nanoemulsionsstabilized by modified starch were spray-dried to powders after the emulsification process. The powdersshowed a good dissolution in water and the reconstituted emulsions had similar particle sizes with thefresh nanoemulsions. A 30 days storage test was carried out to investigate the effect of relative humility(RH) on the storage stability of beta-carotene powders at 25.0 �C. The beta-carotene degradation profilesover time were found to fit well with a Weibull model and also closely related to the film property of thematrix, moisture sorption property and glass transition temperature of the powder. The results showedthat modified starches with lower film oxygen permeability had a higher retention of beta-caroteneduring storage. The glass transition temperature of powder in different RH also affected the rate ofbeta-carotene degradation. Overall these results provide useful information for choosing wall materialsand storage conditions to protect nutraceuticals in delivery systems.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

As a big group of nature pigments with health benefit of vitaminA, carotenoid family has received much attention from scientists,consumers and manufactures in recent years (Huang, Yu, & Ru,2010; de Paz et al., 2012; Zeb & Murkovic, 2011). Besides re-searches have been reported that carotenoids may provide a varietyof potential health-promoting functions (Yuan, Gao, Zhao, & Mao,2008). With antioxidant activity, carotenoids may help reduce therisk for some chronic diseases, such as cancer, cardiovascular dis-ease and aging by quenching singlet oxygen and preventing theoxidative damage (Boon, McClements, Weiss, & Decker, 2010; vonLintig, 2010; Singh & Goyal, 2008; Spada, Zapata Norena, FerreiraMarczak, & Tessaro, 2012). They can also strengthen the ability ofimmune system to resist infections and help the reproductivesystems function properly (Handelman, 2001). Within the carot-enoid family, beta-carotene is one of the most intensively studiedcarotenoid since it shows the most potent provitamin A activity

x: þ86 530 752 7347.

C.F. Shoemaker), fzhong@

All rights reserved.

(Ferreira & Rodriguez-Amaya, 2008; Hou et al., 2012; Kandlakunta,Rajendran, & Thingnganing, 2008; Yuan et al., 2008). It is also themost widely distributed carotenoid in foods with the approval forhuman consumption by U.S. Food and Drug Administration (FDA).

However, with the high degree of unsaturation, carotenoids arelabile to thermal and chemical oxidation, isomerization andphotosensitization when exposed to oxygen, light and high tem-perature during process and storage (Ferreira & Rodriguez-Amaya,2008). The hydrogenecarbon skeleton of the molecule endowsthem lipophilic property and makes them insoluble in water andevenmarginally soluble in oil at room temperature (Mattea, Martin,Matias-Gago, & Cocero, 2009; Ribeiro & Cruz, 2005). Moreover, thecrystalline form of caroteniod greatly limits their bioavailabilitywith low uptake and absorption in body (Yuan et al., 2008).

Different strategies have been reported to solve the problemsabove by encapsulating carotenoids into an appropriate deliverysystem, such as nanoemulsion (Kim, Hyun, Yun, Lee, & Byun, 2012),liposome (Lee et al., 2002), microemulsion (Amar, Aserin, & Garti,2003), solid lipid nanoparticles (Cornacchia & Roos, 2011), nano-dispersion (Anarjan, Mirhosseini, Baharin, & Tan, 2011) and com-plex assemblies with macromolecules (Gouranton et al., 2008; Pan,Yao, & Jiang, 2007). Among these methods, oil-in-water nano-emulsions are considered to be an efficient, low cost and conve-nient way to increase the dispersibility, stability and bioavailability

R. Liang et al. / Food Hydrocolloids 33 (2013) 225e233226

of nutraceuticals (Huang et al., 2010). Nanoemulsions are colloidalsystems with droplet sizes smaller than 200 nm (Tadros, Izquierdo,Esquena, & Solans, 2004). With the small size, nanoemulsions showhighly kinetic and thermodynamic stability by resisting to thecreaming, flocculation and coalescence (Devarajan & Ravichandran,2011). Nanoemulsions are also claimed to exhibit highly absorptionefficacy which improves the oral bioavailability of many functionalfood ingredients (Li, Zheng, Xiao, &McClements, 2012;McClements& Xiao, 2012). There have been reports of successfully incorporatingbeta-carotene in nanoemulsions. For example, Qian C. used highpressure homogenization to prepare beta-carotene nanoemulsionsstabilized by b-lactoglobulin with particle size of 78 nm (Qian,Decker, Xiao, & McClements, 2012). And Mao L. reported the ef-fect of homogenization models on the particle size of beta-carotenenanoemulsions (Mao, Yang, Xu, Yuan, & Gao, 2010).

Although with the improved thermodynamically stability,nanoemulsions are still sensitive to the environment stresses, suchas heat, freezing and thawing, and altering pH which would furtherdestabilizes the encapsulated compound (McClements, 1999). Andthe liquid-based state of nanoemulsion hinders their applicationinto the dry food system. To overcome these limitations, a dryingprocess is recommended following the emulsification process toremove the water from the system and simultaneous trap the oilphasewithin a glassy matrix of wall materials to produce a flowableand dispersible powder (Gharsallaoui, Roudaut, Chambin, Voilley, &Saurel, 2007).

Among various drying technologies, spray drying is the mostpopular and widely studied encapsulation technology due to thehigh production capacity and minimal operational costs. The keysteps for spray drying is the choice of an appropriate wall materialwhich possess good emulsifying properties and good film formingproperties during dehydration (Madene, Jacquot, Scher, & Desobry,2006). However, not all surfactants commonly used in nano-emulsions are fulfilled to these requirements. For example, smallmolecular emulsifiers, such as Tween, Span, lecithin or decaglycerolmonolaurate, are benefit to produce a small particle size but notsuitable to form a good film around the core materials during thespray drying. As for the biopolymeric emulsifiers, such as gumarabic andwhey protein, are always limited by their high viscositiesat high concentrations which would further lower the efficiency ofdrying process. Comparatively, modified starches are the potentialones to meet the requirements with good emulsifying and filmforming properties, low viscosities, high oil loading capacities andoxygen barrier properties (Frascareli, Silva, Tonon, & Hubinger,2012; Murugesan & Orsat, 2012).

On the basis of our newest published work, stable beta-carotenenanoemulsions were successfully prepared by using a high con-centration ofmodified starches, whichmeans it is possible to spray-dry the emulsions directly (Liang, Shoemaker, Yang, Zhong, &Huang, 2013). And the properties of the emulsions are closelyrelated to the interfacial layers of modified starchmolecules aroundthe oil surface. In this investigation, spray drying process wasconducted following the emulsification process to produce beta-carotene powders. And the filming forming properties of modi-fied starches were first studies to predict the storage stability ofencapsulated beta-carotene. Although few researches are found inliterature on the preparation of nanoemulsion powders by spraydrying, some benefit results from studies of the freeze-drying beta-carotene emulsion powder and D-limonene spray drying powderrevealed that RH and physical state of wall matrix are mainparameters to determine the stability of core materials (Ramoneda,Ponce-Cevallos, del Pilar Buera, & Elizalde, 2011; Soottitantawatet al., 2004). So a further attention was paid to investigate theeffect of water activity on beta-carotene retentions and glasstransition of powders. At the same time the relationship between

degradation kinetics of beta-carotene and Tg of powders wasinvestigated.

2. Materials and methods

2.1. Materials

Beta-carotene (�97.0%, UV) was purchased from Fluka (St. Louis,MO, USA). Neobee 1053 (medium-chain triacylglycerol, MCT oil)with 44% C-10 and 56% C-8 was donated by Stepan Company(Maywood, New Jersey). N-octenyl succinate anhydride (OSA)modified starches: HI-CAP (OSA 1), CAPSUL (OSA 2) and CAPSUL TA(OSA 3) were used as encapsulating materials obtained from Na-tional Starch Group (Bridgewater, New Jersey). All other chemicalswere purchased from SigmaeAldrich (St. Louis, MO, USA) and usedwithout further purification. Deionized water obtained from Milli-Q water purification system (Millipore Co., Bedford, MA, USA) wasused in all experiments.

2.2. Film forming properties of modified starches

2.2.1. Film formationModified starch solutions of 10% (w/w) were prepared using

deionized water and mixed overnight. Glycerol was added to allsolutions at concentration of 15% (w/w) to avoid brittleness of thefinal films. Film solutions with 3 g total solids per plate were cast bypipetting onto 14.7 cm internal diameter, rimmed, smooth high-density polyethylene plates placed on a smooth level granite slab.The solution was spread evenly with a glass rod and allowed to dryovernight at room temperature until dried films could be releasedintact from plates.

2.2.2. Oxygen permeability (OP) testAn Ox-Tran 2/20 modular system (Modern Controls Inc., Min-

neapolis, MN, USA) was used to measure oxygen transmission ratesthrough the films, according to the ASTM standard method D3985(ASTM,1995). Oxygen transmission rates were determinedwith theOx-Tran controlled at 25 �C and 50 � 1% RH. A filmwas placed on astainless steel mask with an open testing area of 5 cm2. OP wascalculated via dividing the oxygen transmission rate by the differ-ence between the oxygen partial pressures for sides of the film(1 atm) and multiplying by the average film thickness.

2.2.3. Water vapor permeability (WVP) measurementThe measurement of WVP was determined gravimetrically ac-

cording to ASTM E96-00 method with some modifications (ASTM,1993). The film was sealed on the top of a permeation cell con-taining distilled water (100% RH), placed in a desiccator at 25 �C and0% RH containing silica. The cells were weighed at intervals of 2 hfor 24 h. Water vapor transmission rate (WVTR) of the films wasobtained from the slope of weight loss versus time by a linearregression. Then WVP was calculated by the equation:

WVP ¼ WVTR � thicknesswater vapor partial pressure

(1)

where WVTR is in g/h m2, thickness is in millimeters and partialpressure is in kilopascals (McHugh, Avena-Bustillos, & Krochta,1993).

2.3. Preparation of beta-carotene nanoemulsions

The water phase was prepared by dispersing modified starchpowder in deionized water at concentration of 30% (w/w) andstirring overnight at room temperature to enhance the hydration.

R. Liang et al. / Food Hydrocolloids 33 (2013) 225e233 227

Then an oil phase containing 0.3% (w/w) of beta-carotene inMCToilwas emulsified into the hydrated water phase in a mass ratio of 3:7.A primary emulsionwas produced using a high speed homogenizer(Ultra-Turrax T25 basic, IKAWorks Inc., Willington, USA) equippedwith a S25 N18 G rotor working at 24,000 rpm for 1 min. Next, highpressure homogenization (EmulsiFlex-C3, Avestin Inc., Ottawa,Ontario, Canada) was conducted for 10 cycles at 150 MPa to pro-duce a fine emulsion. The temperature of the inlet reservoir andhomogenization valve was kept at 15 �C by a heat exchanger. Lightexposure of the samples was avoided during the process.

2.4. Spray drying

The nanoemulsion was spray-dried in a Mobile Minor� spraydryer (GEA Niro, Soeborg, Denmark), equipped with a co-currentnozzle atomizer. The size of drying chamber is Ø 800 � 620 mm,60� cone. The operational conditions of the spray drying were: airinlet temperature of 190 �C, air outlet temperature of 85 �C, feedrate of 20 mL/min.

2.5. Particle size determination

The average particle size of the droplets was determined bydynamic light scattering (Nano-ZS90, Malvern Instruments, Wor-cestershire, UK). Samples were diluted 1:100 using deionized waterto prevent multiple scattering effects. The instrument reports Z-average diameters for droplet size and polydispersity index (PDI)for size distribution.

2.6. The storage stability of spray-dried beta-carotenenanoemulsions

Around 0.2 g of the spray-dried nanoemulsion powder wasweighed and spread in a 10 mL glass bottle with a thin layer, andplaced in a desiccator with constant RH of 11, 33, 52, 75, and 97%using saturated salt solutions of lithium chloride, Magnesiumchloride, magnesium nitrate, sodium chloride and potassium sul-fate at 25 �C in the absence of light (Greenspan, 1977). At fixed timeintervals, the sample bottles were removed from the desiccator tomeasure the concentration of beta-carotene as the methoddescribed below for 30 days. The results are expressed as the beta-carotene retentions which are defined as Ct/C0, where C0 is theinitial beta-carotene concentration measured immediately afterspray drying, Ct is the beta-carotene concentration at storage time t.

2.7. Quantification of beta-carotene by HPLC

The concentration of beta-carotene in dried nanoemulsionpowders was determined as following: around 0.2 g of the storagepowder was first dispersed in 1 g of water by stirring until a ho-mogenous dispersion was produced. Then 100 mL of the solutionwas added to 900 mL dimethyl sulfoxide in a test tube and mixed ina vortex mixer for 3 min. After that hexane/dichloromethane (4:1)was added to the tube to extract the beta-carotene. After a wellshaking, the tube was centrifuged at 3000 g/min for 10 min toseparate the organic phase from water phase. At last, the beta-carotene concentration in the organic phase was quantified by aHPLC system (Dionex Ultimate 3000) with a Dionex Bio LC AD25UV/Vis Detector (Dionex Corp., Sunnyvale, CA, USA) at wavelengthof 450 nm. Twenty microliters of samples were injected onto a C30column (250 � 4.6 mm i.d., 5 mm, YMC, Inc., Wilmington, NC) atroom temperature. Themobile phase consisted of methanol (A) andmethyl tert-butyl ether (B) at a flow rate of 1.0 mL/min. Thesolvent gradient was: 0e4min, linear gradient from90% A to 70% A;

4e7 min, linear gradient from 70% A to 15% A; 7e15 min, held at15% A; 15e17 min, A went back to 90% linearly.

2.8. Moisture sorption isotherms

Moisture sorption isotherms of spray-dried powders weredetermined by static gravimetric method. The samples wereweighted in glass bottles and placed into desiccators with constantRHs inside from 11 to 90% as described above at 25 �C. The sampleswereweighed periodically until equilibriumwas attained (10 days).Once equilibrium was reached, water content was determined asthe difference inweight before and after drying in an oven at 105 �Cfor 3 h. The samples were measured 5 times and the averages wereused.

2.9. Glass transition determination

According to the method from Soottitantawat et al. (2004), glasstransition temperatures (Tg) were determined by a Q2000 (TA In-struments, USA) differential scanning calorimeter equipped with arefrigerated cooling system. Approximately 5 mg of samples wasweighed in an aluminum pan and sealed hermetically. The thermalanalysis was performed using a two-cycle scan model, with thetemperature ranging from �100 �C to 130 �C depending on thewater content. Heating and cooling rates were 10 �C/min. An emptypan was used as reference. Glass transition temperature was takenas the midpoint of the baseline shift in the second scan.

2.10. Scanning electron microscopy (SEM)

A scanning electron microscope (Quanta-200, FEI, Holland) wasused to observe the outer structure changes of the spray-driedpowders under different RH during storage. Samples wereattached to double-sided adhesive tape and then mounted on thespecimen holder. The samples were sputter coated with 10 mmthickness of gold under vacuum. The sputtered coated sampleswere scanned with an accelerating beam voltage of 10 kV.

2.11. Statistical analysis

The whole experiment was conducted in duplicate and all an-alyses were carried out at least in triplicate and expressed asmean � standard deviation (SD). The OneWay Analysis of Variance(ANOVA) test was analyzed using the SPSS 17.0 package. Duncan’smultiple range test was used to determine the significant differ-ences of mean value (p < 0.05).

3. Results and discussion

3.1. Film forming properties of the modified starches

An important functionality property of wall materials formicroencapsulation produced by spray drying is their film-formingcapacity during the drying process (Madene et al., 2006). Thepermeability of the film as an indirect nature in the real dried foodmodel is important for the storage stability of encapsulated corematerials fromoxygen and other possible deteriorations (Andersen,Risbo, Andersen, & Skibsted, 2000; Arvanitoyannis, Kalichevsky,Blanshard, & Psomiadou, 1994). In order to predict the storagestability of the spray-dried beta-carotene nanoemulsion powders,film forming properties of the modified starches were first inves-tigated by casting method andmeasuring their OP andWVP values.The OP values for three different modified starch (MS) films weremainly around 1e4 cm3 mm/m2 day kPa (Fig. 1A) which was inagreement with the results from Jimenez, Jose Fabra, Talens, and

Fig. 2. Particle size distributions of fresh (F) and reconstituted (R) beta-carotenenanoemulsions stabilized by OSA 1, OSA 2 and OSA 3.

Table 1Particle size and size distribution of fresh and reconstituted beta-carotenenanoemulsions.a

Wall Fresh beta-carotene Reconstituted beta-carotene

Fig. 1. Oxygen permeability (A) and water vapor permeability (B) of films prepared byOSA 1, OSA 2 and OSA 3. Values bearing different letters are significantly different(P < 0.05).

R. Liang et al. / Food Hydrocolloids 33 (2013) 225e233228

Chiralt (2012). And they were increased in the order OSA 1 < OSA 3< OSA 2 which could be explained by the rule of like-dissolves-like(Yoo & Krochta, 2011). Oxygen is a non-polar molecule which tendsto dissolve in less polar polymers.Whereas with the high amount ofhydroxyl groups, starch molecules have a high polarity whichresulted in a low OP value for the starch films. After modificationwith n-OSA, the MS molecules become less polarity because of thesubstitution ofeOH groups with the less polareOSA groups. In thisstudy, the degree of substitution for threeMS increased in the orderof OSA 1 < OSA 3 < OSA 2 (data were not shown here) which wereconsistent with the increase of OP values for MS films. WVP valuesfor the MS films were measured, Fig. 1B. OSA 1 films showed thehighest WVP value of 3.82 � 0.09 g mm/m2 h kPa and films madefrom OSA 2 and OSA 3 had a similar WVP value around 3.1 g mm/m2 h kPa. These results were lower than the WVP values of nativestarch films reported by Jimenez et al. (2012) but agreed with thereport of Zavareze et al. (2012) that the modification lowered thehydrophilic properties of starch and further decreased their WVPvalues. Thewater barrier properties of theMS films could be relatedto their hygroscopicity of the spray-dried powders which affectedthe storage stability of the powders in different RH.

materials nanoemulsion emulsion

OSA 1 Z-average(nm)

159.3 � 0.141 151.6 � 1.556

PDI 0.321 � 0.021 0.324 � 0.002OSA 2 Z-average

(nm)114.6 � 0.989 102.3 � 0.989

PDI 0.234 � 0.013 0.237 � 0.011OSA 3 Z-average

(nm)118.3 � 1.344 115.1 � 1.202

PDI 0.238 � 0.002 0.244 � 0.004

a Data expressed as mean � standard deviation (n ¼ 3).

3.2. The effect of spray drying on the nanoemulsion particles

After preparing nanoemulsions usingMS, theywere spray-driedto get the powder directly and then re-suspended in water. Thepowders showed a good dissolution in water. As showed in Fig. 2,unimodal and narrow particle size distributions were obtained forfresh and reconstituted beta-carotene emulsions. The Z-averagediameters for three emulsions stabilized by OSA 1, OSA 2 and OSA 3

were 159, 114 and 118 nm, respectively (Table 1). And the recon-stituted emulsions also showed the similar particle sizes as originalemulsion with a good dispersity (PDI < 0.4) which suggested thatthe spray drying process did not affect the characteristics ofnanoemulsions (r < 200 nm).

3.3. Influence of RH on the retention of beta-carotene in spray-dried nanoemulsion powder

In order to investigate the effect of RH on the retention of beta-carotene in the spray-dried nanoemulsions, a 30 days storage testwas carried out with various RHs (11, 33, 52, 75 and 97%) at 25 �C,Fig. 3AeC. In general, for eachMS and RH it was showed a decline inthe level of beta-carotenewith respect to storage time. At the end of30 days for each RH, there was a consistent pattern with regard tothe retention of beta-carotene, the rankings were OSA 1 > OSA3 > OSA 2, as the results summarized in Table 2. It also can be seenfrom the results that with the increasing of RH the degradation ofbeta-carotene was first accelerated until RH reaching to 52% andthen sharply decreased with the higher RH for OSA 1 as the wallmaterial. Also, for powders stabilized by OSA 2 and OSA 3, the re-tentions of beta-carotene showed the lowest values (23.67 and26.71%) at RH 75%. Similar results were obtained by Serris et al.which related to the degradation of beetroot pigment encapsulatedby maltodextrin (Serris & Biliaderis, 2001). These results suggestedthat the RH showed a significant effect on the storage stability ofbeta-carotene powders, but the relationship between the RH andbeta-carotene retentions was not simple.

Fig. 3. Retention of beta-carotene in powder encapsulated by (A) OSA 1; (B) OSA 2 and(C) OSA 3 over time with different store RH: (-) 11% RH; (C) 32% RH; (:) 52% RH;(;) 75% RH; (A) 97% RH. The solid lines showed the fitting data by Weibull model.

Table 2Beta-carotene retention levels in nanoemulsion powder stabilized by modified starches

Wall materials Beta-carotene retention (%)

11% 32%

OSA 1 71.87 � 0.38d,E 69.58 � 0.61g,D

OSA 2 35.06 � 0.39b,I 30.62 � 0.41e,H

OSA 3 36.18 � 0.13c,N 35.47 � 0.01f.M

bep Values of means bearing different letters in the same column are significantly differeAeO Values of means bearing different letters in the same rows are significantly differen

a Data expressed as mean � standard deviation (n ¼ 3).

Table 3Kinetic parametersa and correlation coefficients of determination of the degradationof beta-carotene powder encapsulated by different MS stored under different RHat 25 �C.

Wall materials % RH Weibull equation

k ( � 10�3/day) n R2

OSA 1 11 13.36 � 0.50 1.263 � 0.040 0.998833 14.16 � 0.87 1.158 � 0.063 0.995852 33.92 � 0.50 1.478 � 0.060 0.997275 24.90 � 0.65 1.362 � 0.060 0.996897 18.36 � 0.51 0.997 � 0.024 0.9986

OSA 2 11 34.40 � 0.23 1.230 � 0.020 0.999633 38.92 � 0.37 1.243 � 0.028 0.999252 40.97 � 0.25 1.204 � 0.017 0.999775 44.06 � 0.98 1.222 � 0.063 0.996097 21.26 � 0.71 0.812 � 0.031 0.9977

OSA 3 11 33.2 � 0.42 1.237 � 0.036 0.998833 35.4 � 0.84 1.330 � 0.065 0.996552 38.44 � 0.83 1.300 � 0.062 0.995075 41.60 � 0.73 1.197 � 0.047 0.996097 19.94 � 0.38 1.009 � 0.022 0.9992

a The range indicates 95% confidence levels of the fitting values.

R. Liang et al. / Food Hydrocolloids 33 (2013) 225e233 229

3.4. Degradation kinetics of beta-carotene at different RH

In order to study thedegradation characteristics of beta-carotenenanoemulsionpowders in response toRHvariation, the degradationkinetic modeling was carried out by fitting the time-course degra-dation profile with the Weibull distribution function as reported inthe previous work (Rawson, Brunton, & Tuohy, 2012; Syamaladevi,Sablani, Tang, Powers, & Swanson, 2011). The equation is:

R ¼ exp½�ðktÞn� (2)

where R is the retention of beta-carotene, t is the storage time, k isthe degradation rate constant, n is a shape factor. The time-coursesdegradation curves of beta-carotene correlated with Weibullequation were shown as the solid lines in Fig. 3. And the kineticparameters and regression coefficients were listed in Table 3 forpowders encapsulated by OSA 1, OSA 2 and OSA 3 under differentRHs. According to the results, high regression coefficients (R2>0.99)were shown for all sampleswhichwas conformedwith the researchby Spada et al. (2012) that theWeibull model was suitable to modelthe degradation of beta-carotene by yielding a good fit of theexperimental data. The obtained k values from the function describethe general trends for beta-carotene degradation in relation to theRH. For the samples examined, the k values first increased with theincrease of RH, followed by a decrease at an RH around 75, 97 and97% for powders stabilized by OSA 1, OSA 2 and OSA 3, respectively.These results were corresponding to the results of beta-caroteneretentions discussed above which may be relevant to the physicalstructure change of the systems at high RH. Several literature ob-tained the similar results and investigated that the structuralcollapse affected the degradation kinetics with increasing RH until avalue at which the samples collapsed; then the rate constants

after 30 days storagea at 25 C.

52% 75% 97%

36.51 � 0.51j,A 50.93 � 0.53m,B 58.04 � 0.64p,C

27.67 � 0.53h,G 23.67 � 0.01k,F 52.54 � 0.57n,J

29.53 � 0.13i,L 26.71 � 0.14l,K 55.61 � 0.28� ,O

nt (P < 0.05).t (P < 0.05).

Fig. 4. Outer structure changes of the spray-dried powder encapsulated by OSA 1 (A), OSA 2 (B) and OSA 3 (C) stored at 25 �C. A-1, B-1 and C-1, storage at 33% RH for 10 days; A-2,B-2 and C-2, storage at 51% RH for 10 days; A-3, B-3 and C-3, storage at 75% RH for 10 days; B-4 and C-4, storage at 97% RH for 10 days.

Fig. 5. Water adsorption isotherms of beta-carotene nanoemulsion powders. The solidlines showed the fitting data by GAB model.

R. Liang et al. / Food Hydrocolloids 33 (2013) 225e233230

diminished (Rodriguez-Huezo, Pedroza-Islas, Prado-Barragan,Beristain, & Vernon-Carter, 2004; Selim, Tsimidou, & Biliaderis,2000).

The loss of beta-carotene during storage was maybe resulted inthe oxidation reaction by oxygen diffusing through the wall ma-terials (Xie, Zhou, & Zhang, 2007). It can be clearly seen from theresults in Table 3 that k values were increased for the beta-carotenepowders in the order OSA 1 < OSA 3 < OSA 2 which was corre-sponding with the increasing of the OP values for MS films in theorder OSA 1 < OSA 3 < OSA 2 (Fig. 1A). This result was consistentwith reports from Moreau and Rosenberg (1998) that the perme-ability of the wall matrix to oxygen determines the stability of thecore material.

Weibull parameter n related to the shape of degradation curvewas summarized in Table 3. For the samples at 11, 33, 52 and 75%RH, n values were greater than 1 which indicated the general shapeof the degradation curve is convex with a fast degradation rate ofbeta-carotene during the storage. When RH reaching to 97%, nvalues were decreased less than 1 (OSA 1 and OSA 2) or close to 1(OSA 3) which represented the degradation curve is concavewith the loss of beta-carotene during the initial stage of storage.

Table 4GAB equation parametersa and correlation coefficients of beta-carotene nano-emulsion powder after equilibration at 25 �C.

Wallmaterials

m0 (g H2O/100 g dry basis) C k1 R2

OSA 1 3.93 � 0.38 10.21 � 5.25 0.99 � 0.10 0.9924OSA 2 4.53 � 0.43 34.04 � 15.02 0.87 � 0.03 0.9876OSA 3 5.05 � 0.31 39.53 � 12.05 0.83 � 0.02 0.9944

a The range indicates 95% confidence levels of the fitting values.

Table 6GordoneTaylor equation parametersa and correlation coefficients of beta-carotenepowder encapsulated by MS.

Wall materials Tg1 z R2

OSA 1 88.61 � 6.35 7.30 � 0.59 0.9959OSA 2 109.20 � 3.21 2.21 � 0.18 0.9914OSA 3 119.50 � 2.93 2.56 � 0.15 0.9955

a The range indicates 95% confidence levels of the fitting values.

R. Liang et al. / Food Hydrocolloids 33 (2013) 225e233 231

The same behavior was found by Prado, Buera, and Elizalde (2005)to describe the loss of beta-carotene in a supercooled polymericmatrix. As explained by Soottitantawat et al. (2004), these phe-nomena may be attributed to the fact that at storage temperaturethe higher RH would induce the structural collapse of the samplesfrom a glass to a rubbery state by decreasing the surface area to arather sticky or compacted appearance.

3.5. External structure of beta-carotene nanoemulsion powders

To explain the degradation results obtained above, SEM wasutilized to observe the external structure change during the storageat different RHs. As shown in Fig. 4, rounded surface with grooveswere observed for the powder encapsulated by OSA 1, OSA 2 andOSA 3, respectively. However, a smoother powder surface wasobserved in the OSA 1 which was in agreement with the previousresearch (Soottitantawat et al., 2005). At a low RH (33%) the spray-dried particles retained their original shape (Fig. 4A-1, B-1 and C-1).With the increasing level of RH the OSA 1 capsules appeared partlyrehydrated and adhered together at RH 52% (Fig. 4A-2), anddestroyed totally at RH 75% (Fig. 4A-3). As for the powders of OSA 2and OSA 3, no obvious changes of the structure were found even atRH 75%, just with some fractures of the particles (Fig. 4B-3 and C-3)which were corresponded to the increasing of the degradation ratewith the RH (Table 3). A collapsed structure appeared at the RH 97%for OSA 2 and OSA 3 (Fig. 4B-4 and C-4). These results confirmed theexplanation for the change of the n values which the emulsiondroplet of beta-carotene in the powder crushed due to thedestruction of the capsule matrixes and forming a rather sticky orcompacted appearance at higher RH. And the compactness struc-tures of different OSAs increased in the order OSA 2 < OSA 3 < OSA1 (Fig. 4A-3, B-4 and C-4). These results were consistent with thebeta-carotene retentions after 30 days storage (Table 2).

3.6. Water sorption isotherms of beta-carotene nanoemulsionpowders

The sorption of moisture caused a change of matrix structureand was considered to be an important factor which affects thestorage stability of the powder product under different RHs. Thewater sorption isotherms of the spray-dried beta-carotene powderswere measured in Fig. 5. The experimental water contents in driedbasis as a function of water activity (aw ¼ RH/100) was shownwiththe symbols. Moreover, GAB (Guggenheim, Anderson and de Boer)

Table 5Glass transition temperature (Tg)a of beta-carotene powder after equilibration at25 �C as a function of water activity (aw).

Wall materials Tg (�C)

aw ¼ 0.33 aw ¼ 0.52 aw ¼ 0.75

OSA 1 31.36 � 0.12 9.06 � 0.12 �26.01 � 0.09OSA 2 80.32 � 0.02 73.6 � 0.16 55.17 � 0.30OSA 3 82.75 � 0.02 74.56 � 0.10 55.54 � 0.28

a Data expressed as mean � standard deviation (n ¼ 3).

model which is the most extensively preferred for powder productwas used to describe the sorption properties of beta-carotenepowder. The equation as follows:

mm0

¼ Ck1aw1� k1awð Þ 1þ C � 1ð Þk1aw½ � (3)

where m is the equilibrium water content (% dried basis), m0 is themonolayer water content (% dried basis), C and k1 are the Gug-genheim constant related to the heat of sorption. The predictedwater sorption isotherms conducted at 25 �C for all systems wereshown with the solid lines in Fig. 5. According to the correlationcoefficients (R2) given in Table 4, GAB model described reasonablywell with the equilibrium water content for all systems studied(R2 > 0.99 in all samples). The moisture sorption profiles of beta-carotene powders exhibited a typical sigmoid shape which wasconsistent with the beta-carotene encapsulating systems reportedby Ramoneda et al. (2011). At a higher aw (aw > 0.5), OSA 1 systemshowed a higher water uptake than those of the OSA 2 and OSA 3.These results were also represented by bigger k1 values for OSA 1powder (Table 4). However, no significant difference was observedfor the equilibrium moisture contents of OSA 2 and OSA 3 systems.These phenomenawere in good agreement with theWVP results ofthe MS films which showed higher WVP for OSA 1 and no differ-ences for OSA 2 and OSA 3. It can be explained that the higher WVPof the film would accelerate the water transport to the capsulessystem further increase the hygroscopicity of the powder.

3.7. Glass transition temperatures of beta-carotene nanoemulsionpowders

The storage test results (retention, k and n values) of the beta-carotene nanoemulsion powders discussed above revealed thatthe structural state of the powder under different RH was vital todetermine the level of degradation of beta-carotene. Glass transi-tion temperature (Tg) is an index which expresses the critical

Fig. 6. Prediction of glass transition temperature of powders at 25 �C under variablewater activity with the GordoneTaylor equation and the GAB sorption equation.

Fig. 7. Influence of phase transition parameter T � Tg of powder on the degradation rate constant of encapsulated beta-carotene at 25 �C. (A) OSA 1, (B) OSA 2, (C) OSA 3. The errorbars indicate 95% confidence levels.

R. Liang et al. / Food Hydrocolloids 33 (2013) 225e233232

temperature at which thematerials begin to soften due to the long-range coordinated molecular motion with the phase transitionfrom a rigid glassy to a viscous rubbery state (Li, Taylor, & Mauer,2011; Soottitantawat et al., 2004). Changes in the storage RHexpressed as aw values of beta-carotene powder imply changes inthe water content according to sorption isotherm which furtherinduce the glass transition of the amorphous phase. As the resultsshown in Table 5, the increased aw lowered the Tg values of the solidpowder which means it would be easier to the phase transition at alower temperature by plasticization effect of moisture. More in-formation related to the relationship between the water contentwith the Tg was accurately predicted by the Gordon and Taylormodel as following:

Tg ¼ x1Tg1 þ zx2Tg2x1 þ zx2

(4)

where x1 is the mass fraction of capsule matrixes; Tg1 is glasstransition temperature of the matrix at zero water content; x2 ismass fraction of water; Tg2 is glass transition temperature ofamorphous water (�135 �C) and z is a constant related to thestrength of the matrixewater interaction.

The parameters were obtained by nonlinear regression of theexperimental Tg values at different RH were shown in Table 6 andthe corresponding profiles of Tg against aw at 25 �C for threedifferent wall materials were given in Fig. 6. The beta-carotenepowders stabilized by OSA 2 and OSA 3 exhibited a higher Tgvalues than that of OSA 1 which were also represented by a smallerz values. The z value represents the effectiveness of water indecreasing Tg in a given matrix (Ramoneda et al., 2011). Thus beta-carotene powder stabilized by OSA 1 with the highest capacity forwater sorption (as discussed above) exhibited a higher z valueswith highest susceptibility to water.

3.8. Relationship between beta-carotene degradation constantswith the glass transition temperature

In order to demonstrate the effect of the physical state on thestorage stability of the beta-carotene powder clearly, a relationshipbetween degradation constant (k) and glass transition temperature(Tg) was analyzed by plotting k values as a function of the storagetemperature (T)minus the Tg of the powder, T� Tg. The Tg values usedhere were derived from the GordoneTaylor and GAB equations(Fig. 6). Several authors have suggested an analogous correlation toexplain the storage stability related with the structural changearound Tg (Selimet al., 2000; Serris &Biliaderis, 2001; Soottitantawatet al., 2004). As shown in Fig. 7, the k values increased with theincreasing value of T � Tg, over a point, followed by a sharplydecrease. These observations could be presumably explained by thediffusion and mobility constrains which are determined by the

macrostructure and porosity of thematrix (Serris & Biliaderis, 2001).With the increasing of T � Tg, the mobility of the reactants wasincreased including oxygen which diffuses through the porous wallmaterials and beta-carotene molecules in core materials. As thetemperature approached to Tg, the transformation from glassy torubbery state occurs. Beta-carotene powder stabilized by OSA 1 wasplasticizedbyabsorbedwaterandshowed thehighestkvaluearoundRH 53%. But for powders encapsulated by OSA 2 and OSA 3, compactstructures were exhibited even at 75% RH (as evidenced by SEMpictures in Fig. 4). According to the predicted Tg values in Fig. 6,around 93% RH would induce a Tg value around 25 �C. It could beassumed that in the region of T� Tg around zero, higher k valuesmayexist for OSA 2 and OSA 3 capsules. However, after fully collapseddeclined k valueswere shown for T� Tg> 0 regionwhichmay bedueto the disappearance ofmicropore spaces of the surface by forming arather stickyorcompactedappearance further inhibit thediffusionofoxygen to the system (as shown in Fig. 4A-3, B-4 and C-4). Whortonand Reineccius (1995) also showed similar results for flavor encap-sulations by defining the state as the re-encapsulation process.

4. Conclusion

In summary, beta-carotene nanoemulsion powder encapsulatedby MS was prepared by high pressure homogenization and spraydrying processes. The powder has a good dissolution in water andthe reconstituted emulsions maintain in nanosize. From the storagetest at 25 �C under different RH, WVP and OP properties of MS filmwould affect the water sorption and oxygen transport of powderswhich further affect the beta-carotene retentions in the system. Atlast, the correlation of glass transition temperature and degradationconstant was analyzed to discover that the degradation of beta-carotene was closely related with the structure state of the ma-trixes. The knowledge obtained from this study could be importantto design new strategies to improve the stability of active ingredient.

Acknowledgement

We thank Dr. JohnMKrochta of Department of Food Science andTechnology, University of California at Davis for the assistant of filmforming experiment. This research was supported by the NationalNatural Science Foundation of China (30901000 and 31171686),BK2012556, 2011BAD23B02, 2013AA102207 and Funds for theCentral Universities JUSRP11015.

References

Amar, I., Aserin, A., & Garti, N. (2003). Solubilization patterns of lutein and luteinesters in food grade nonionic microemulsions. Journal of Agricultural and FoodChemistry, 51(16), 4775e4781.

Anarjan, N., Mirhosseini, H., Baharin, B. S., & Tan, C. P. (2011). Effect of processingconditions on physicochemical properties of sodium caseinate-stabilized

R. Liang et al. / Food Hydrocolloids 33 (2013) 225e233 233

astaxanthin nanodispersions. LWT e Food Science and Technology, 44(7), 1658e1665.

Andersen, A. B., Risbo, J., Andersen, M. L., & Skibsted, L. H. (2000). Oxygen perme-ation through an oil-encapsulating glassy food matrix studied by ESR linebroadening using a nitroxyl spin probe. Food Chemistry, 70(4), 499e508.

Arvanitoyannis, I., Kalichevsky, M., Blanshard, J. M. V., & Psomiadou, E. (1994). Studyof diffusion and permeation of gases in undrawn and uniaxially drawn filmsmade from potato and rice starch conditioned at different relative humidities.Carbohydrate Polymers, 24(1), 1e15.

ASTM. (1993). Standard methods for water vapor transmission of materials. InAnnual book of American (pp. 629e636). Philadelphia: American Society forTesting and Materials.

ASTM. (1995). Standard test methods for oxygen gas transmission rate throughplastic films and sheeting using a coulometric sensor. In Annual book of Amer-ican (pp. 472e477). Philadelphia: American Society for Testing and Materials.

Boon, C. S., McClements, D. J., Weiss, J., & Decker, E. A. (2010). Factors influencing thechemical stability of carotenoids in foods. Critical Reviews in Food Science andNutrition, 50(6), 515e532.

Cornacchia, L., & Roos, Y. H. (2011). Stability of beta-carotene in protein-stabilizedoil-in-water delivery systems. Journal of Agricultural and Food Chemistry,59(13), 7013e7020.

Devarajan, V., & Ravichandran, V. (2011). Nanoemulsions: as modified drug deliverytool. International Journal of Comprehensive Pharmacy, 4(1), 1e6.

Ferreira, J. E. M., & Rodriguez-Amaya, D. B. (2008). Degradation of lycopene andb-carotene in model systems and in lyophilized Guava during ambient storage:kinetics, structure, and matrix effects. Journal of Food Science, 73(8), C589eC594.

Frascareli, E. C., Silva, V. M., Tonon, R. V., & Hubinger, M. D. (2012). Determination ofcritical storage conditions of coffee oil microcapsules by coupling water sorp-tion isotherms and glass transition temperature. International Journal of FoodScience and Technology, 47(5), 1044e1054.

Gharsallaoui, A., Roudaut, G., Chambin, O., Voilley, A., & Saurel, R. (2007). Appli-cations of spray-drying in microencapsulation of food ingredients: an overview.Food Research International, 40(9), 1107e1121.

Gouranton, E., El Yazidi, C., Cardinault, N., Amiot, M. J., Borel, P., & Landrier, J.-F. (2008).Purified low-density lipoprotein and bovine serum albumin efficiency to inter-nalise lycopene into adipocytes. Food and Chemical Toxicology, 46(12), 3832e3836.

Greenspan, L. (1977). Humidity fixed points of binary saturated aqueous solutions.Journal of Research of The National Bureau of StandardseA Physics and Chemistry,81, 89e96.

Handelman, G. J. (2001). The evolving role of carotenoids in human biochemistry.Nutrition, 17(10), 818e822.

Hou, Z. Q., Zhang, M., Liu, B., Yan, Q. L., Yuan, F., Xu, D. X., et al. (2012). Effect ofchitosan molecular weight on the stability and rheological properties of beta-carotene emulsions stabilized by soybean soluble polysaccharides. Food Hy-drocolloids, 26(1), 205e211.

Huang, Q. R., Yu, H. L., & Ru, Q. M. (2010). Bioavailability and delivery of nutra-ceuticals using Nanotechnology. Journal of Food Science, 75(1), R50eR57.

Jimenez, A., Jose Fabra, M., Talens, P., & Chiralt, A. (2012). Effect of sodium caseinateon properties and ageing behaviour of corn starch based films. Food Hydrocol-loids, 29(2), 265e271.

Kandlakunta, B., Rajendran, A., & Thingnganing, L. (2008). Carotene content of somecommon (cereals, pulses, vegetables, spices and condiments) and unconven-tional sources of plant origin. Food Chemistry, 106(1), 85e89.

Kim, D. M., Hyun, S. S., Yun, P., Lee, C. H., & Byun, S. Y. (2012). Identification of anemulsifier and conditions for preparing stable nanoemulsions containing theantioxidant astaxanthin. International Journal of Cosmetic Science, 34(1), 64e73.

Lee, S. C., Yuk, H. G., Lee, D. H., Lee, K. E., Hwang, Y. I., & Ludescher, R. D. (2002).Stabilization of retinol through incorporation into liposomes. Journal ofBiochemistry and Molecular Biology, 35(4), 358e363.

Li, N., Taylor, L. S., & Mauer, L. J. (2011). Degradation kinetics of catechins in greentea powder: effects of temperature and relative humidity. Journal of Agriculturaland Food Chemistry, 59(11), 6082e6090.

Li, Y., Zheng, J. K., Xiao, H., & McClements, D. J. (2012). Nanoemulsion-based deliverysystems for poorly water-soluble bioactive compounds: influence of formula-tion parameters on polymethoxyflavone crystallization. Food Hydrocolloids,27(2), 517e528.

Liang, R., Shoemaker, C. F., Yang, X., Zhong, F., & Huang, Q. (2013). Stability andbioaccessibility of b-carotene in nanoemulsions stabilized by modified starches.Journal of Agricultural and Food Chemistry, 61(6), 1249e1257.

von Lintig, J. (2010). Colors with functions: elucidating the biochemical and mo-lecular basis of carotenoid metabolism. In R. J. Cousins (Ed.). Annual review ofnutrition, Vol. 30, (pp. 35e56).

McClements, D. J. (1999). Food emulsions: Principles, practices and techniques (2nded.). Florida: CRC Press.

McClements, D. J., & Xiao, H. (2012). Potential biological fate of ingested nano-emulsions: influence of particle characteristics. Food & Function, 3(3), 202e220.

McHugh, T. H., Avena-Bustillos, R., & Krochta, J. M. (1993). Hydrophilic edible films:modified procedure for water vapor permeability and explanation of thicknesseffects. Journal of Food Science, 58(4), 899e903.

Madene, A., Jacquot, M., Scher, J., & Desobry, S. (2006). Flavour encapsulation andcontrolled release e a review. International Journal of Food Science and Tech-nology, 41(1), 1e21.

Mao, L., Yang, J., Xu, D., Yuan, F., & Gao, Y. (2010). Effects of homogenization modelsand emulsifiers on the physicochemical properties of beta-carotene nano-emulsions. Journal of Dispersion Science and Technology, 31(7), 986e993.

Mattea, F., Martin, A., Matias-Gago, A., & Cocero, M. J. (2009). Supercritical anti-solvent precipitation from an emulsion: beta-carotene nanoparticle formation.Journal of Supercritical Fluids, 51(2), 238e247.

Moreau, D. L., & Rosenberg, M. (1998). Porosity of whey protein-based microcap-sules containing anhydrous milkfat measured by gas displacement pycnometry.Journal of Food Science, 63(5), 819e823.

Murugesan, R., & Orsat, V. (2012). Spray drying for the production of nutraceuticalingredientsda review. Food and Bioprocess Technology, 5(1), 3e14.

Pan, X., Yao, P., & Jiang, M. (2007). Simultaneous nanoparticle formation andencapsulation driven by hydrophobic interaction of casein-graft-dextran and b-carotene. Journal of Colloid and Interface Science, 315(2), 456e463.

de Paz, E., Martin, A., Estrella, A., Rodriguez-Rojo, S., Matias, A. A., Duarte, C. M. M.,et al. (2012). Formulation of beta-carotene by precipitation from pressurizedethyl acetate-on-water emulsions for application as natural colorant. Food Hy-drocolloids, 26(1), 17e27.

Prado, S. M., Buera, M. P., & Elizalde, B. E. (2005). Structural collapse prevents b-carotene loss in a supercooled polymeric matrix. Journal of Agricultural and FoodChemistry, 54(1), 79e85.

Qian, C., Decker, E. A., Xiao, H., & McClements, D. J. (2012). Physical and chemicalstability of beta-carotene-enriched nanoemulsions: Influence of pH, ionicstrength, temperature, and emulsifier type. Food Chemistry, 132(3), 1221e1229.

Ramoneda, X. A., Ponce-Cevallos, P. A., del Pilar Buera, M., & Elizalde, B. E. (2011).Degradation of beta-carotene in amorphous polymer matrices. Effect of watersorption properties and physical state. Journal of the Science of Food and Agri-culture, 91(14), 2587e2593.

Rawson, A., Brunton, N., & Tuohy, M. (2012). High pressure-temperature degrada-tion kinetics of polyacetylenes in carrots. Food Chemistry, 133(1), 15e20.

Ribeiro, H. S., & Cruz, R. C. D. (2005). Biliquid foams containing carotenoids. Engi-neering in Life Sciences, 5(1), 84e88.

Rodriguez-Huezo, M. E., Pedroza-Islas, R., Prado-Barragan, L. A., Beristain, C. I., &Vernon-Carter, E. J. (2004). Microencapsulation by spray drying of multipleemulsions containing carotenoids. Journal of Food Science, 69(7), E351eE359.

Selim, K., Tsimidou, M., & Biliaderis, C. G. (2000). Kinetic studies of degradation ofsaffron carotenoids encapsulated in amorphous polymer matrices. FoodChemistry, 71(2), 199e206.

Serris, G. S., & Biliaderis, C. G. (2001). Degradation kinetics of beetroot pigmentencapsulated in polymeric matrices. Journal of the Science of Food and Agricul-ture, 81(8), 691e700.

Singh, P., & Goyal, G. K. (2008). Dietary lycopene: its properties and anticarcinogeniceffects. Comprehensive Reviews in Food Science and Food Safety, 7(3), 255e270.

Soottitantawat, A., Takayama, K., Okamura, K., Muranaka, D., Yoshii, H., Furuta, T.,et al. (2005). Microencapsulation of L-menthol by spray drying and its releasecharacteristics. Innovative Food Science & Emerging Technologies, 6(2), 163e170.

Soottitantawat, A., Yoshii, H., Furuta, T., Ohgawara, M., Forssell, P., Partanen, R., et al.(2004). Effect of water activity on the release characteristics and oxidativestability of D-limonene encapsulated by spray drying. Journal of Agricultural andFood Chemistry, 52(5), 1269e1276.

Spada, J. C., Zapata Norena, C. P., Ferreira Marczak, L. D., & Tessaro, I. C. (2012). Studyon the stability of beta-carotene microencapsulated with pinhao (Araucariaangustifolia seeds) starch. Carbohydrate Polymers, 89(4), 1166e1173.

Syamaladevi, R. M., Sablani, S. S., Tang, J., Powers, J., & Swanson, B. G. (2011). Sta-bility of anthocyanins in frozen and freeze-dried raspberries during long-termstorage: in relation to glass transition. Journal of Food Science, 76(6), E414eE421.

Tadros, T., Izquierdo, R., Esquena, J., & Solans, C. (2004). Formation and stability ofnano-emulsions. Advances in Colloid and Interface Science, 108, 303e318.

Whorton, C., & Reineccius, G. A. (1995). Evaluation of the mechanisms associ-ated with the release of encapsulated materals from maltodextrin matrices.In S. J. Risch, & G. A. Reineccius (Eds.). Encapsulation and controlled release offood ingredients, Vol. 590, (pp. 143e160).

Xie, Y. L., Zhou, H. M., & Zhang, Z. R. (2007). Effect of relative humidity on retentionand stability of vitamin A microencapsulated by spray drying. Journal of FoodBiochemistry, 31(1), 68e80.

Yoo, S., & Krochta, J. M. (2011). Whey proteinepolysaccharide blended edible filmformation and barrier, tensile, thermal and transparency properties. Journal ofthe Science of Food and Agriculture, 91(14), 2628e2636.

Yuan, Y., Gao, Y., Zhao, J., & Mao, L. (2008). Characterization and stability evaluationof [beta]-carotene nanoemulsions prepared by high pressure homogenizationunder various emulsifying conditions. Food Research International, 41(1), 61e68.

Zavareze, E. D., Pinto, V. Z., Klein, B., El Halal, S. L. M., Elias, M. C., Prentice-Hernandez, C., et al. (2012). Development of oxidised and heat-moisture treatedpotato starch film. Food Chemistry, 132(1), 344e350.

Zeb, A., & Murkovic, M. (2011). Determination of thermal oxidation and oxidationproducts of b-carotene in corn oil triacylglycerols. Food Research International.

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具