sal

n P

, Hala U

Article history:Received 4 February 2014Accepted 17 April 2014

1. Introduction

(Gennadios, Weller, Hanna, & Froning, 1996; Prodpran, Benjakul, &Artharn, 2007).

Fish protein isolate (FPI) prepared by alkaline solubilisation wasreported as the promising starting material for preparation of lmswith negligible yellow discolouration (Tongnuanchan, Benjakul,

Prodpran, & Songtipya, 2011). However, FPI based lm is rigid,t of hydrophiliced-FPI lms haveo hydrophilicity oficant amounts ofty (Prodpran et al.,low stripe trevally,died by blendingith lower glycerolko, 2014). FPI/FSGand water vapourHowever, FPI/FSG

blend lms still have poorer mechanical as well as water vapourbarrier properties, in comparison with synthetic lms.

Nanoparticles with the ller property has been implemented inbiopolymer lms for improving mechanical, thermal and watervapour barrier properties (Kovacevic, Vrsaljko, Blagojevic, &Leskovac, 2008). Amongst nano-llers, ZnO nanoparticles(ZnONP) have the excellent ability for nano-scale dispersion andinterfacial interactions in protein matrix due to their large specicsurface area and high surface energy (Rouhi, Mahmud, Naderi, Ooi,

* Corresponding author. Tel.: 66 7428 6334; fax: 66 7455 8866.

Contents lists availab

Food Hydr

els

Food Hydrocolloids 41 (2014) 265e273E-mail address: soottawat.b@psu.ac.th (S. Benjakul).Biopolymers have been widely paid attention over the last de-cades due to their advantages and potential applications in foodindustries (Kanmani & Rhim, 2014; Rhim & Ng, 2007). Biopolymerlms are the excellent vehicles for incorporating a wide variety ofadditives, such as antimicrobials, antioxidants, antifungal agents,colourants, and other nutrients, thus improving food quality andextending shelf-life of foods and products (Rhim & Ng, 2007).Amongst biopolymers, proteins from different sources have beenimpressively used for the development of biodegradable lms dueto their relative abundance and good lm-forming ability

thus requiring the addition of high amounplasticiser (Tongnuanchan et al., 2011). Plasticisthe poor water vapour barrier property, owing tamino acids in protein molecules and to the signhydrophilic plasticisers required for lm exibili2007). Recently, properties of FPI lms from yelan abundant trash dark muscle sh, could be mowith sh skin gelatin (FSG) at a ratio of 5:5 wcontent (30%) (Arfat, Benjakul, Prodpran, & Osablend lms showed the improved mechanicalbarrier properties, compared with FPI lms.Available online 26 April 2014

Keywords:BionanocompositeFish protein isolateFish skin gelatinBasil essential oilZnO nanoparticlesAntimicrobial packaginghttp://dx.doi.org/10.1016/j.foodhyd.2014.04.0230268-005X/ 2014 Elsevier Ltd. All rights reserved.Composite lms based on sh protein isolate (FPI) and sh skin gelatin (FSG) blend incorporated with 50and 100% (w/w, protein) basil leaf essential oil (BEO) in the absence and presence of 3% (w/w, protein) ZnOnanoparticles (ZnONP) were prepared and characterised. Tensile strength (TS) decreased, whilst elonga-tion at break (EAB) increased as BEO level increased (p < 0.05). However, ZnONP addition resulted inhigher TS but lower EAB (p < 0.05). The lowest water vapour permeability (WVP) was observed for thelm incorporated with 100% BEO and 3% ZnONP (p < 0.05). BEO and ZnONP incorporation decreasedtransparency of FPI/FSG lms (p< 0.05). FTIR spectra indicated that lms added with BEO exhibited higherhydrophobicity. Both BEO and ZnONP had a marked impact on thermal stability of the lms. Micro-structural study revealed that the presence of ZnONP prevented bilayer formation of lm containing 100%BEO. FPI/FSG lms incorporated with 100% BEO, especially in combination with ZnONP, exhibited strongantibacterial activity against foodborne pathogenic and spoilage bacteria and thus could be used as anactive food packaging material to ensure safety and to extend the shelf-life of packaged foods.

2014 Elsevier Ltd. All rights reserved.a r t i c l e i n f o a b s t r a c tProperties and antimicrobial activity ofgelatin lm containing basil leaf essentinanoparticles

Yasir Ali Arfat a, Soottawat Benjakul a,*, ThummanooPonusa Songtipya b

aDepartment of Food Technology, Faculty of Agro-Industry, Prince of Songkla UniversitybDepartment of Material Product Technology, Faculty of Agro-Industry, Prince of Songk

journal homepage: www.h protein isolate/sh skinoil and zinc oxide

rodpran b, Punnanee Sumpavapol a,

t Yai, Songkhla 90112, Thailandniversity, Hat Yai, Songkhla 90112, Thailand

le at ScienceDirect

ocolloids

evier .com/locate/ foodhyd

roco& Mahmood, 2013). Recently, the incorporation of ZnONP as func-tional ller into the biopolymer lms such as starch based lms hasbeen reported to improve mechanical and water vapour barrierproperties (Alebooyeh, Nafchi, & Jokar, 2012; Yu, Yang, Liu, & Ma,2009). ZnO is currently listed as a generally recognised as safe(GRAS) material by the Food and Drug Administration(21CFR182.8991) and has previously shown strong in vitro antimi-crobial activity against foodborne pathogens and spoilage bacteria(Espitia et al., 2013; Zhang, Ding, Povey, & York, 2008).

Antimicrobial biodegradable lms have received increasingattention because of their potential to delay microbial spoilage offoods (Emiroglu, Yemis, Coskun, & Candogan, 2010; Rhim & Ng,2007). Essential oils from aromatic and medicinal plants havebeen known to be biologically active, mainly possessing antibac-terial and antioxidant properties (Ahmad, Benjakul, Prodpran, &Agustini, 2012). Although most essential oils are classied asGenerally Recognised as Safe, their use as food preservatives isoften limited due to negative organoleptic effects when added insufcient amounts to provide an antimicrobial effect. The advan-tage of applying such essential oils through the use of lms, insteadof applying them directly on foods, is that low diffusion rate of theactive compounds allows to attain the desired antimicrobial effectwith lower oil concentrations, thus limiting unwanted avours andodours to the food (Snchez-Gonzlez et al., 2011). Additionally,essential oils are hydrophobic in nature and the incorporation ofessential oils could improve the water vapour barrier property andimpart exibility of protein lms (Tongnuanchan, Benjakul, &Prodpran, 2013). The use of essential oil exhibited the increasedantimicrobial activity, when combined together with variousnanoparticles (Allahverdiyev, Kon, Abamor, Bagirova, & Rafailovich,2011). To the best of our knowledge, no information regarding thecombined effect of essential oils and ZnONP on properties of pro-tein based lms has been reported. The present study aimed toinvestigate the combined effects of ZnONP and basil leaf essentialoil (BEO) on physico-mechanical and thermal properties of FPI/FSGblend lms. Antimicrobial activity of lms against Listeria mono-cytogenes (foodborne pathogen) and Pseudomonas aureginosa (foodspoilage bacteria) was also examined.

2. Materials and methods

2.1. Chemicals

Zinc oxide nanoparticle (ZnONP) (particle size: 20e40 nm,specic surface area: 26.22 m2/g) was purchased from Nano ma-terials technology Co. Ltd. (Bangkok, Thailand). Sodium hydroxideand hydrochloric acid were obtained from Merck (Darmstadt,Germany). Commercial sh skin gelatin (FSG) from tilapia (w240bloom) was purchased from Lapi Gelatine S.p.A (Empoli, Italy).Essential oil from the leaves of basil (Ocimum basilicum) was ob-tained from Botanicessence (Bangkok, Thailand). All chemicalswere of analytical grade. L. monocytogenesDMST 1327was obtainedfrom Department of Medical Sciences, Ministry of Public Health,Nonthaburi, Thailand and P. aeruginosa TISTR 781 was obtainedfrom Thailand Institute of Scientic and Technological Research(TISTR), Thailand.

2.2. Collection and preparation of sh sample

Fresh yellow stripe trevally (Selaroides leptolepis) with anaverage weight of 90e100 g/sh were purchased from a localmarket in Hat Yai, Songkhla province, Thailand. Fish were kept inice with a sh/ice ratio of 1:2 (w/w) and transported to theDepartment of Food Technology, Prince of Songkla University

Y.A. Arfat et al. / Food Hyd266within 30 min. Upon the arrival, sh were immediately washed,lleted, and minced to uniformity by passing through a 5 mmscreen using a Model HC 5000 mincer (Microuidics, Massachu-setts, USA).

2.3. Preparation of sh protein isolate

Prior to the isolation of sh protein, the prepared mince wassubjected to washing as per the method of Toyohara, Sakata,Yamashita, and Shimizu (1990) with slight modications. Fishmince was homogenised with 5 volumes of cold 0.05 M NaCl (2e4 C) at a speed of 13,000 rpm for 2 min, using an IKA Labortechnikhomogeniser (Selangor, Malaysia). The washed mince was lteredthrough two layers of cheese-cloth. The washing process wasrepeated twice. Washed mince obtained was stored on ice untilused.

Washed mince was added with cold distilled water at the ratioof 1:9 (w/v), followed by homogenisation for 1 min at a speed of13,000 rpm. The pH of homogenate was then adjusted to 11 using2 M NaOH. The resulting mixture was centrifuged at 10,000 g for20 min at 4 C using a refrigerated centrifuge (Avanti-JE Centrifuge,Beckman 163 Coulter Inc., Fullerton, CA, USA). The supernatant wascollected and the pH was adjusted to 5.5 using 2 M HCl. The pre-cipitate was then ltered through 4 layers of cheese-cloth. Theretentate was dewatered by centrifugation at 12,000 g for 20 minat 4 C. The nal pH of the sample was adjusted to pH 7.0 using2 M NaOH. The sample was referred to as sh protein isolate; FPI.FPI was used for lm preparation.

2.4. Preparation of sh protein isolate/sh skin gelatin lm addedwith BEO and ZnONP

Firstly, lm-forming solution was prepared according to themethod of Chinabhark, Benjakul, and Prodpran (2007). FPI wasadded with 3 volumes of distilled water and homogenised at13,000 rpm for 1 min using a homogeniser. Subsequently, the pH ofthe mixture was adjusted to 3 using 1 N HCl, to solubilise the pro-tein. The obtained solution was ltered through 2 layers of cheese-cloth to remove undissolved debris. The protein concentration of theltrate determined by the Kjeldahl method (AOAC, 2000) wasadjusted to 3% (w/v). Glycerol at 30% (w/w) of proteinwas used as aplasticiser. The mixtures were stirred gently for 30 min at roomtemperature and were used for preparing blend FFS.

Prior to blending, FSG powder was dissolved in distilled water toobtain the protein concentration of 3% (w/v). The pH of the mixturewas adjusted to 3 using 1 N HCl. The solution was heated at 70 Cfor 30 min. Then, glycerol (30%, w/w) was added as mentionedabove. Thereafter, both FPI and FSG solutions were mixed at a ratioof 1:1 (v/v). The obtained solution was added without and with 3%ZnONP (w/w, protein content) in droplets. Before addition ofZnONP, ZnONP was suspended in distilled water and homogenisedfor 1 min at 5000 rpm (IKA Labortechnik homogeniser, Selangor,Malaysia). The obtained FPI/FSG/ZnONP suspension was stirred for5 min and then homogenised for 30 s at the speed of 5000 rpm.

BEO previously mixed with Tween 20 at 25% (w/w, based onessential oil) was added to the FPI/FSG/ZnONP suspension at levelsof 50% and 100% (w/w, protein content). Final volumewas made upto 100 ml using distilled water previously adjusted to pH 3. Toobtain the uniform distribution of BEO and ZnONP, the suspensionswere homogenised with three passes through a high pressurehomogeniser (Microuidizer M-110EH, Microuidics Corp.,Newton, MA, USA) with an operating pressure of 1500 bars. Sus-pensions were gently stirred for 30 min at room temperature andwere referred to as a lm-forming suspension (FFS). Prior to cast-ing, FFS samples were degassed for 10 min using the sonicating

lloids 41 (2014) 265e273bath (Elmasonic S 30 H, Singen, Germany). To prepare the lm, 4 g

rocoof FFS was cast onto a rimmed silicone resin plate (5 cm 5 cm),air-blown for 12 h at 25 C, followed by drying in an environmentalchamber (Binder GmbH, Tuttlingen, Germany) at 25 0.5 C and50 5% relative humidity (RH) for 24 h. Dried lm samples weremanually peeled off and subjected to analyses.

2.5. Determination of properties and antimicrobial activity of lms

Prior to testing, lms were conditioned for 48 h at 50 5%relative humidity (RH) at 25 0.5 C. For ATR-FTIR, TGA and SEM,lms were dried in a desiccator containing dried silica gel for 1week and in a desiccator containing P2O5 for 2 weeks at roomtemperature (28e30 C) to obtain themost dehydrated lms and tominimise the plasticising effect due to water.

2.5.1. Film thicknessThe thickness of lm was measured using a digital micrometer

(Mitutoyo, Model ID-C112PM, Serial No. 00320, Mituyoto Corp.,Kawasaki-shi, Japan). Ten random locations around each lmsample were used for thickness determination.

2.5.2. Mechanical propertiesTensile strength (TS) and elongation at break (EAB) of lms

were determined as described by Iwata, Ishizaki, Handa, andTanaka (2000) using the Universal Testing Machine (Lloyd In-struments, Hampshire, UK). Ten samples (2 cm 5 cm) with initialgrip length of 3 cm were used for testing. The samples wereclamped and deformed under tensile loading using a 100 N loadcell with the cross-head speed of 30 mm/min until the sampleswere broken. The maximum load and the nal extension at breakwere used for calculation of TS and EAB, respectively(Tongnuanchan et al., 2011).

2.5.3. Water vapour permeability (WVP)WVP was measured using a modied ASTM method (American

Society for Testing and Materials, 1989) as described by Shiku,Hamaguchi, Benjakul, Visessanguan, and Tanaka (2004). The lmwas sealed on an aluminium permeation cup containing driedsilica gel (0% RH) with silicone vacuum grease and a rubber gasketto hold the lm in place. The cups were placed in a desiccatorcontaining the distilled water at 30 C. The cups were weighed at1 h intervals over a 10 h period. WVP of the lm was calculated asfollows:

WVPg m1 s1 Pa1

wlA1 t1P2 P11

wherew is theweight gain of the cup (g); l is the lm thickness (m);A is the exposed area of lm (m2); t is the time of gain (s); (P2 P1)is the vapour pressure difference across the lm (4244.9 Pa at30 C).

2.5.4. ColourColour of lmwas determined using a CIE colourimeter (Hunter

associates laboratory, Inc., Reston, VA, USA). Colour of the lm wasexpressed as L* e (lightness), a* e (redness/greenness) and b*e(yellowness/blueness) values. Total difference in colour (DE*) wascalculated according to the following equation (Gennadios et al.,1996):

DE* DL* 2 Da* 2 Db* 2

q

where DL*, Da* and Db* are the differences between the corre-sponding colour parameter of the samples and that of white stan-

Y.A. Arfat et al. / Food Hyddard (L* 92.65, a* 0.92, b* 0.49).2.5.5. Light transmittance and transparency valueThe light transmittance of lms was measured at the ultraviolet

and visible range (200e800 nm) using a UVevisible spectropho-tometer (Shimadzu UV-1800, Kyoto, Japan) according to themethod of Shiku et al. (2004). The transparency value of lm wascalculated using the following equation (Han & Floros, 1997):

Transparency value log T600x

where T600 is the fractional transmittance at 600 nm and x is thelm thickness (mm). The greater transparency value represents thelower transparency of lm.

2.5.6. Antimicrobial propertiesAntimicrobial activity of lm samples was assessed using the

agar diffusion method according to Ponce, Roura, Del Valle, andMoreira (2008). The zone of inhibition on solid media was usedfor determining antimicrobial effects of lms against typical foodpathogenic and spoilage bacteria including Gram-positive bacteria(L. monocytogenes) and Gram-negative bacteria (P. aeruginosa). Filmsamples were cut into a disc shape of 10 mm diameter using acircular knife and then placed on Mueller Hinton (Merck, Darm-stadt, Germany) agar plates, which had been previously smearedwith 100 ml of inoculum containing approximately 106e107 CFU/mlof tested bacteria. The plates were then incubated at 37 C for 24 h.The antimicrobial activity of nanocomposite lms was determinedby measuring the diameter of the bacterial growth inhibition zonearound the lm.

2.6. Characterisation of the selected lms

The FPI/FSG lms incorporated without and with BEO at 100%(w/w protein), in the absence and presence of 3% ZnONP werecharacterised, in comparison with the control FPI/FSG lms(without BEO and ZnONP).

2.6.1. Attenuated total reectance-Fourier transform infrared (ATR-FTIR) spectroscopy

FTIR spectra of lm samples were determined using a BrukerModel Equinox 55 FTIR spectrometer (Bruker Co., Ettlingen, Ger-many) equippedwith a horizontal ATR Trough plate crystal cell (45

ZnSe; 80 mm long, 10 mmwide and 4 mm thick) (PIKE TechnologyInc., Madison, WI, USA) at 25 C as described by Nuthong, Benjakul,and Prodpran (2009). Samples were placed onto the crystal cellsand the cells were clamped into the mount of FTIR spectrometer.The spectra in the range of 650e4000 cm1 with automatic signalgain were collected in 32 scans at a resolution of 4 cm1 and wereratioed against a background spectrum recorded from the cleanempty cell at 25 C.

2.6.2. Thermo-gravimetric analysis (TGA)Films were scanned using a thermo-gravimetric analyser (TGA-

7, Perkin Elmer, Norwalk, CT, USA) from 40 to 600 C at a rate of10 C/min (Nuthong et al., 2009). Nitrogen was used as the purgegas at a ow rate of 20 mL/min.

2.6.3. Scanning electron microscopy (SEM)Morphology of surface and freeze-fractured cross-section of lm

samples were visualised using a scanning electron microscope(SEM) (Quanta400, FEI, Eindhoven, the Netherlands) at an acceler-ating voltage of 15 kV. Prior to visualisation, the lm samples weremounted on brass stub and sputteredwith gold in order tomake thesample conductive, and photographs were taken at 7000 magni-

lloids 41 (2014) 265e273 267cation for surface visualisation. For cross-section, freeze-fractured

lmsweremounted around stubs using double sided adhesive tape,coated with gold and observed at the 3500magnication.

2.7. Statistical analysis

Experiments were run in triplicate using different three lots ofsamples. Data were subjected to analysis of variance (ANOVA) andmean comparisons were carried out by Duncans multiple rangetest (Steel & Torrie, 1980). Statistical analysis was performed using

bonding and hydrophobic interaction were the major bonds sta-bilising FPI and FSG lms network (Arfat et al., 2014). Addition oflipid or oil in protein-based lms may hinder proteineproteininteraction and provide exible domains within the lm(Limpisophon, Tanaka, & Osako, 2010). The results suggested thatessential oil functioned as plasticiser in resulting lm. Plasticisersare molecules, which penetrate into biopolymer matrix, therebyincreasing the volume of empty spaces between the chains ofmolecules and causing a decrease in the strength of intermolecular

ZnONP added as shown in Table 1. FPI/FSG lm is known to have

affe

Y.A. Arfat et al. / Food Hydrocolloids 41 (2014) 265e273268the Statistical Package for Social Science (SPSS 17.0 for windows,SPSS Inc., Chicago, IL, U.S.A.).

3. Results and discussion

3.1. Properties of FPI/FSG lms incorporated with BEO and ZnONP

FPI/FSG lms incorporated with BEO at 50 and 100% (w/w) inthe absence and presence of 3% ZnONP, exhibited the differentproperties and antimicrobial activity.

3.1.1. ThicknessThickness of FPI/FSG lms incorporated with BEO at different

levels in the absence and presence of 3% ZnONP is shown in Table 1.Thickness of lms increased when the levels of BEO increased(p < 0.05), regardless of ZnONP incorporated. It was plausibly dueto the protruded structures mediated by interaction betweenchemical components present in BEO and protein matrix. The re-sults suggested that protein chains could not form the compact andordered lm network in the presence of BEO as indicated byincreased thickness. Moreover, the increases in thickness with theaddition of BEO could be due to the increase in solid content inlms. Ahmad et al. (2012) reported the increase in thickness ofgelatin based lms incorporated with bergamot essential oil.

Films containing 3% ZnONP had higher thickness, comparedwith those without ZnONP (p< 0.05) at all BEO levels used. At 100%BEO level, lms incorporatedwith ZnONP had the highest thickness(p < 0.05). This was plausibly due to the combined effect of 3%ZnONP and 100% BEO, inserting between protein network as well asthe increased solid content of resulting lms. This behaviour wasalso observed by Sothornvit, Rhim, and Hong (2009) for wheyprotein isolate nanocomposite lms.

3.1.2. Mechanical propertiesMechanical properties of FPI/FSG lms containing BEO at

various levels without and with 3% ZnONP are shown in Table 1.Films incorporated with BEO showed the lower TS, but higher EAB,compared with the control (without essential oil), regardless ofZnONP addition (p < 0.05). Amongst all lms added with BEO, thelowest TS but the highest EAB were found in lms added with 100%BEO (p< 0.05). The incorporation of essential oil into FPI/FSG blendlm could enhance the development of heterogeneous lm matrix,leading to discontinuity of lm network. As a result, the lower ri-gidity and strength of lm matrix were obtained. Hydrogen

Table 1Thickness, mechanical properties and water vapour permeability of FPI/FSG lms as

ZnONP level (% w/w) Essential oil level (% w/w) Thickness (mm)

0 0 0.036 0.001d50 0.040 0.002c

100 0.045 0.002b3 0 0.041 0.002c

50 0.045 0.001b100 0.050 0.002aValues are given as mean SD (n 3).Different lowercase letters in the same column indicate signicant differences (p < 0.05poor water barrier properties due to the presence of polar aminoacids and hydroxyl group (eOH), mainly from glycerol (Arfat et al.,2014; Prodpran et al., 2007). WVP of FPI/FSG lms decreased withincreasing amount of BEO (p < 0.05). Incorporation of BEO,nonpolar or hydrophobic materials, increased the hydrophobicityof lms, thereby lowering the adsorptivity as well as diffusivity ofwater vapour through the lm as indicated by lower WVP. Mono-terpene hydrocarbon was generally found in essential oils (Turina,Nolan, Zygadlo, & Perillo, 2006) and contributed to hydrophobicityof lm. The presence of a hydrophobic disperse phase, even at smallratio, limits water vapour transfer since it interferes hydrophilicphase and increases the tortuosity factor of mass transfer (Atares,Bonilla, & Chiralt, 2010). Furthermore, water vapour transfer nor-mally occurs through the hydrophilic portion of lm network anddepends on hydrophilic/hydrophobic ratio of lm constituent(Tongnuanchan, Benjakul, & Prodpran, 2012). This result was inagreement with Pires et al. (2011) who reported that thymeessential oil incorporated into hake protein based lm decreasedWVP of resulting lms, especially with increasing amount ofessential oil.

cted by BEO and ZnONP incorporation.

TS (MPa) EAB (%) WVP (1011 gm1 s1 Pa1)11.94 0.82c 68.38 4.89c 3.62 0.17a9.52 0.72d 89.26 5.41b 3.06 0.14b7.23 0.52e 110.05 7.09a 2.67 0.13c

14.70 0.36a 56.39 5.67d 2.86 0.15bc13.05 0.58b 70.18 5.29c 2.36 0.16d11.17 0.80c 85.51 5.50b 1.63 0.10eforces along the matrix (Hoque, Benjakul, Prodpran, & Songtipya,2011). The addition of lipids, oils or fatty acids with increasinglevel decreased TS of protein-based lms e.g., lms from com-mercial sh skin gelatin (Tongnuanchan et al., 2013) and shmuscle protein (Tanaka, Ishizaki, Suzuki, & Takai, 2001).

At all BEO levels, FPI/FSG lms incorporated with 3% ZnONPshowed the higher TS and the lower EAB (p < 0.05), as comparedwith those without ZnONP. ZnONP more likely acted as the rein-forcing ller in the protein matrix, thereby enhancing the strengthof resulting lms. Reinforcing effect of nanoparticles in gelatinbased lm was reported by Rouhi et al. (2013). The ZnONP incor-porated generated an interactive force with protein molecules andlimited the molecular mobility of protein chain (Rouhi et al., 2013).The nanoparticles have high surface energy and large specic sur-face area, giving rise to strong interfacial interactions betweenpolymers and the nanoparticles. This brings about a signicantenhancement in mechanical properties of polymers (Kovacevicet al., 2008; Rouhi et al., 2013). Thus, both BEO and ZnONP incor-porated in FPI/FSG lm directly determined mechanical propertiesof resulting composite lms.

3.1.3. Water vapour permeability (WVP)WVP of FPI/FSG lms varied, depending on the levels of BEO and).

WVP of lms decreased when 3% ZnONP was incorporated(p < 0.05), regardless of BEO addition. The signicant decrease inWVP after the addition of ZnONP was mainly attributed to thehydrophobic nature of incorporated ZnONP. Moreover, the incor-poration of these nanoparticles to the protein matrix introduces atortuous pathway for water vapour molecules to pass through(Alebooyeh et al., 2012; Yu et al., 2009). The improvements of waterresistance in nanocomposite biopolymer lms, e.g. nanocompositestarch (Alebooyeh et al., 2012) and nanocomposite gelatin (Rouhiet al., 2013) have been reported. Amongst all lms, those contain-ing 100% BEO and 3% ZnONP had the lowest WVP (p < 0.05). Filmsincorporated with 100% BEO without and with 3% ZnONP had the

Y.A. Arfat et al. / Food Hydrocodecreases in WVP by 26.2, and 54.9%, respectively, in comparisonwith the control lm (without essential oils and ZnONP) (p < 0.05).Thus, both BEO level and ZnONP had the profound impact on WVPof FPI/FSG lms.

3.1.4. ColourThe colour expressed as L*, a*, b* and DE* values of FPI/FSG lms

added with BEO at various levels in the absence and presence of 3%ZnONP is shown in Table 2. As the levels of BEO increased, higher b*and DE* values were observed in resulting lms with theconcomitant lower L* and a* values, regardless of ZnONP incorpo-ration (p < 0.05) (Table 2). The changes in colour of resulting lmswere most likely attributed to the colouring components inessential oil. This result was in agreement with Tongnuanchan et al.(2013) who reported that increasing amount of root essential oilmarkedly increased b* and DE* values with the concomitant de-creases in L* and a* values of sh skin gelatin lms. Nevertheless,the addition of thyme oil (0e0.25 ml oil/g protein) into hake pro-tein lm did not have a signicant impact on colour attributes(Pires et al., 2011).

At all levels of BEO incorporation, the addition of ZnONP did nothave a marked impact on L*- and a*-values of resulting lms(p > 0.05). However, the decreases in b*- and DE*-values wereobtained as ZnONP was incorporated (p < 0.05). These results werein accordance with Espitia et al. (2013) who reported thatincreasing amount of ZnONP markedly decreased b*- and DE*-values, whilst it had no signicant effect on L*- and a*-values ofmethyl cellulose (MC)eZnO nanocomposites. Such changes incolour of resulting lms were most likely caused by the whiteningeffect of ZnONP (Espitia et al., 2013). The results suggested thatZnONP decreased yellowish colour in BEO incorporated FPI/FSGlms.

3.1.5. Light transmittance and transparency valueTransmission of UV and visible light at wavelength range of

200e800 nm of FPI/FSG lms incorporated with 50 and 100% BEOwithout and with 3% ZnONP are shown in Table 3. Control FPI/FSGlms (without BEO and ZnONP) had a good barrier property in theUV range (200e280 nm). Protein based lms are generally known

Table 2Colour of FPI/FSG lms as affected by BEO and ZnONP incorporation.

ZnONPlevel(% w/w)

Essentialoil level(% w/w)

Colour

L* a* b* DE*

0 0 90.54 0.13a 1.37 0.04a 2.13 0.14e 2.71 0.08e50 90.11 0.07b 1.68 0.03b 3.46 0.04c 3.98 0.10c

100 89.87 0.06c 1.93 0.13c 4.17 0.10a 4.72 0.13a3 0 90.65 0.11a 1.41 0.06a 1.81 0.08f 2.45 0.12f

50 90.24 0.06b 1.66 0.04b 3.06 0.03d 3.60 0.08d100 89.93 0.09c 1.90 0.09c 3.69 0.11b 4.31 0.09b

Values are given as mean SD (n 3).

Different lowercase letters in the same column indicate signicant differences(p < 0.05).to have the excellent UV barrier properties due to the presence ofhigh content of aromatic amino acids which can absorb UV light(Arfat et al., 2014; Hoque et al., 2011).

Light transmittance of lms at wavelength of 280 nm decreasedmarkedly from 7.81% (control lm) to 0.08% and 0% upon theincorporation of BEO at levels of 50 and 100%, respectively. Thus,the incorporation of BEO to lm could improve the UV-barrierproperty, especially for lms added with higher level of BEO. Thetransmission of visible light in the range of 350e800 nm of lmsincorporated with BEO decreased with increasing levels of BEO. Thedecrease in light transmittance was possibly caused by light scat-tering of lipid droplets distributed throughout the protein network(Tongnuanchan et al., 2013). The barrier property towards UV andvisible light became more pronounced when ZnONP was incorpo-rated. This was plausibly due to the hindrance of light passage orlight scattering by the nanoparticles dispersed in the lm matrix(Kanmani & Rhim, 2014). Rouhi et al. (2013) reported that thegelatin lms added with ZnO nanorods had lower transmission forUV light. The FPI/FSG lms incorporated with BEO and reinforcedwith ZnONP could be used as potential packaging lms to preventUV light, which can induce lipid oxidation of various foods.

The transparency values of all lm samples are presented inTable 3. The control lm had the lowest transparency value,compared with those lms added with BEO (p < 0.05). The lowertransparency value represents the greater transparency of lm. Thetransparency value of lms increased as the level of BEO incorpo-rated increased (p < 0.05). The result indicated that lms con-taining BEO became less transparent. The decreases intransparency of lms were related with the decrease in lighttransmittance when BEO was incorporated. Moreover, the colour-ing components in BEO as well as the heterogeneous lm networkwere contributable to the decreased transparency of lms. Thisresult was in agreement with Tongnuanchan et al. (2012) who re-ported that the addition of citrus essential oils lowered the trans-parency of sh skin gelatin lm. Addition of thyme essential oil alsodecreased the transparency of hake protein lm (Pires et al., 2011).The lms added with ZnONP had higher transparency value(p < 0.05), regardless of essential oil incorporation. This indicatedthe decrease in transparency of resulting lms. Similarly,Sothornvit, Hong, An, and Rhim (2010) reported that the addition ofnanoparticles to whey protein lms resulted in the reducedtransparency of the lms. Moreover, transparency value of lm alsovaried with the thickness of lm. Thus, the incorporation of BEOand ZnONP had the profound impact on light transmittance andtransparency of lms.

3.1.6. Antimicrobial activity of FPI/FSG lms incorporated with BEOand ZnONP

Antibacterial activity of the FPI/FSG lms incorporated with BEOat various levels in the absence and presence of 3% ZnONP towardGram-positive food pathogenic bacteria (L. monocytogenes) andGram-negative food spoilage bacteria (P. aeruginosa) in comparisonwith the control lms is shown in Fig. 1. The inhibition zone of BEOincorporated lms increased with increasing BEO contents(p < 0.05), regardless of ZnONP. At the same BEO level, the lmsshowed the higher inhibition toward Gram-positive bacteria(L. monocytogenes), compared with Gram-negative bacteria(P. aeruginosa). Essential oils have been employed as food additivefor the preservation of a wide range of foods, including vegetables,rice, fruit, dairy products, sh, meat and poultry (Burt, 2004). Theantimicrobial activity of the BEO could be explained by the signif-icant amount of linalool, an oxygenated monoterpenoid. Thiscompound was shown to have antimicrobial activity (Hussain,Anwar, Hussain-Sherazi, & Przybylski, 2008). Linalool has the po-

lloids 41 (2014) 265e273 269tential to act as either a protein denaturing agent or as a solvent

dehydrating agent, contributing to its antimicrobial activity(Suppakul, Miltz, Sonneveld, & Bigger, 2003). The mechanism ofantimicrobial activity of essential oils is related to its ability topenetrate the cell wall of a bacterium and attack on the phospho-lipid bilayer present in cell membranes. This causes increasedpermeability and leakage of cytoplasm, or in their interaction withenzymes located on the cell wall (Emiroglu et al., 2010). Gram-negative bacteria have an additional external membrane sur-rounding the cell wall, which restricts diffusion of hydrophobiccompounds through its lipopolysaccharide covering (Sharma, Jain,Yadav, & Sharma, 2010). Suppakul et al. (2003) reported that basilessential oils exhibited powerful antimicrobial activity against awide range of microorganisms. Bozin, Mimica-Dukic, Simin, andAnackov (2006) showed that the Gram-positive strains of bacteriashowed higher sensitivity to O. basilicum essential oils.

At the same BEO level incorporated, antibacterial activity of FPI/FSG lms containing 3% ZnONP was higher than that without 3%ZnONP (p < 0.05). This was governed by the strong antimicrobial

Table 3Light transmittance and transparency values of FPI/FSG lms as affected by BEO and ZnO

ZnONP level (% w/w) Essential oil level (% w/w) Light transmittance (%)

200 280 350

0 0 0.00 7.81 68.3250 0.00 0.08 35.78

100 0.00 0.00 17.893 0 0.00 0.45 47.15

50 0.00 0.00 29.71100 0.00 0.00 13.68

Values are given as mean SD (n 3).Different lowercase letters in the same column indicate signicant differences (p < 0.05

Y.A. Arfat et al. / Food Hydroco270Fig. 1. Antimicrobial activity of FPI/FSG lms against L. monocytogenes (A) andP. aeruginosa (B) as inuenced by BEO and ZnONP. Bars represent the standard devi-ation (n 3). Different lowercase letters on the bars indicate signicant differences(P < 0.05). NI indicates the absence of inhibitory zone.activity of ZnONP against both microorganisms. Amongst all lms,that incorporated with 100% BEO and 3% ZnONP had the highestantimicrobial activity (p < 0.05). This was more likely due tocombined antimicrobial effect of BEO and ZnONP distributedthroughout the lms matrix. Antimicrobial activity of ZnO nano-particles and the mechanism of inhibition against the microor-ganisms was demonstrated (Espitia et al., 2013; Zhang et al. 2010).The release of Zn2 ions from the powder could penetrate throughthe cell wall of microorganism and react with interior componentsthat nally affect the viability of the cells. ZnONP has been knownto mediate the generation of hydrogen peroxide (H2O2), a powerfuloxidising agent causing damage to the cell membrane of bacteria(Tayel et al., 2011). At the same BEO level, lm containing 3% ZnONPshowed higher antibacterial effect on L. monocytogenes thanP. aeruginosa (p < 0.05). Tam et al. (2008) demonstrated that ZnOwas more effective for killing Gram-positive than Gram-negativebacteria. Gram-positive bacteria typically have one cytoplasmicmembrane and cell wall is composed of peptidoglycan (Ma &Zhang, 2009). More complex cell wall structure of gram-negativebacteria, with a layer of peptidoglycan between outer membraneand cytoplasmic membrane might prevent ZnO penetration intothe cells (Ma & Zhang, 2009). Antibacterial action of ZnO is sug-gested to occur through its interactionwith specic cell compoundslike surface proteins (e.g., adhesions) and teichoic acids plus lipoids(forming lipoteichoic acids), which may be found in Gram-positiverather than in Gram-negative bacteria (Tayel et al., 2011). However,further studies are required to clarify the exact antibacterialmechanism of ZnONP. Thus, FPI/FSG lms incorporated with 100%BEO and 3% ZnONP had the potential to control bothL. monocytogenes and P. aeruginosa in foods.

3.2. Characteristics of FPI/FSG lms incorporated with BEO andZnONP

The FPI/FSG lms containing 100% BEO in the absence andpresence of 3% ZnONP were characterised, in comparison with thecontrol lms (without BEO and ZnONP).

NP incorporation.

Transparency values

400 500 600 700 800

71.76 75.04 77.34 85.32 87.69 3.10 0.05e62.01 66.27 71.68 83.66 86.08 3.62 0.04c50.88 59.87 66.13 81.98 85.75 3.99 0.06b56.12 68.04 71.87 77.23 80.17 3.51 0.04d43.13 59.71 66.32 76.87 78.21 3.96 0.04b36.00 53.84 61.44 74.58 77.68 4.23 0.07a

).

lloids 41 (2014) 265e2733.2.1. FTIR spectroscopyFTIR spectra of FPI/FSG lms as inuenced by BEO and ZnONP

incorporation are illustrated in Fig. 2. All lms exhibited the majorbands at wavenumber of 1631 cm1 (amide-I, representing C]Ostretching/hydrogen bonding coupled with COO), 1537 cm1

(amide-II, arising from bending vibration of NeH groups andstretching vibrations of CeN groups) and 1238 cm1 (amide-III,illustrating the vibrations in plane of CeN and NeH groups ofbound amide or vibrations of CH2 groups of glycine) (Muyonga,Cole, & Duodu, 2004). Yakimets et al. (2005) reported the similarresult for bovine skin gelatin lm where amide-I, amide-II andamide-III peaks were found at the wavenumbers of 1633, 1536 and1240 cm1, respectively. The band situated at the wavenumber of1036e1039 cm1 was found in all lm samples, corresponding to

higher weight loss were found in lms incorporated with BEO,comparedwith the control lms, regardless of ZnONP addition. Thisresult was in agreement with the lowered TS and higher EAB oflms incorporated with BEO (Table 1). This might be owing to the

20

30

40

50

60

70

80

90

100

110

Wei

ght (

%)

3% ZnONP

100% BEO + 3% ZnONP

Control

100% BEO

16252862

3275 1745

rocolloids 41 (2014) 265e273 271the interactions arising between plasticiser (OH group of glycerol)and lm structure (Hoque et al., 2011). An amide-A band wasobserved at the wavenumber of 3275e3280 cm1 and amide-Bband was found at 2922e2924 cm1 for all lm samples. Theamide-A band represents the NH-stretching coupled withhydrogen bonding and amide-B band represents the asymmetricstretching vibration of CH as well as NH3 (Muyonga et al., 2004).

Amplitude of peaks at wavenumbers 2862 cm1 and 2922 cm1

increased when the lms were incorporated with BEO. Those peaksrepresent the methylene asymmetrical and symmetrical stretchingvibration of the aliphatic CeH in CH2 and CH3 groups, respectively(Guillen & Cabo, 2004; Muik, Lendl, Molina-Diaz, Valcarcel, & Ayora-Canada, 2007). Both themethylene asymmetrical stretching bands atapproximately 2853 cm1 and methylene symmetrical stretchingbandnear2924cm1wereobviouslypresent inmost lipids (Guillen&Cabo, 2004). Thehigheramplitudeof thesepeakswasobtained inlmincorporated with BEO, indicating the presence of BEO containinghydrocarbon in lm matrix. Similar results for lemon essential oilincorporatedgelatinlmswere reported (Tongnuanchanet al., 2012).Furthermore, thepeakwasobserved atwavenumberof1745cm1 forlm incorporated with BEO, regardless of ZnONP addition. It morelikely represented the C]O stretching vibration of aldehyde or estercarbonyl groups (Guillen & Cabo, 2004; Muik et al., 2007). Aldehyde,ketone and ester are the main chemical groups in essential oils(Mohamed, El-Emary, & Ali, 2010). It was noted that no peak at1745 cm1 was found in the control lm.

The slight shift of FTIR spectra of all lms incorporated with 3%ZnONP in the amide-I region to the lower wavenumber was mainlyassociated with interaction of carbonyl of polypeptide chains withZnONP. The shifts of the amide-I peak to a lower wavenumber wererelated to a decrease in the molecular order (Rouhi et al., 2013).

5001000150020002500300035004000

Abs

orba

nce (

A.U

.)

Wavenumber (cm-1)

100% BEO + 3% ZnONP

3% ZnONP

100% BEO

Control

1632

1630

1627

3280

3279

3275

Fig. 2. ATR-FTIR spectra of FPI/FSG lms as inuenced by BEO and ZnONPincorporation.Amide- A(2922)Amide- B

Amide- IAmide- II

Amide- III

Y.A. Arfat et al. / Food HydFurthermore, the amide-A band from the NeH stretching vibrationof thehydrogenbondedNeHgroupbecamevisible atwavenumbers3280 and3275 cm1 for the controllm and3% ZnONP incorporatedlm, respectively. Amide-A band of the lm incorporated with 3%ZnONP and 100% BEO was found at 3275 cm1. This result clearlyshowed that amide-A band shifted to lower wavenumber whenZnONPwas added. The shift to the lowerwavenumber in the amide-A region suggested thatNeHgroups of protein chain interactedwithZnONP,mainly via hydrogen bonding (Nikoo et al., 2011). Rouhi et al.(2013) reported similar results from gelatin-based nanocompositelms incorporated with ZnO nanorods. Therefore, incorporation ofBEO and ZnONP altered the molecular organisation and intermo-lecular interaction in the lm matrix.

3.2.2. Thermo-gravimetric analysis (TGA)TGA thermograms revealing thermal degradation behaviour of

FPI/FSG lms as affected by BEO and ZnONP incorporation areshown in Fig. 3. Their corresponding degradation temperatures(Td), weight loss (Dw) and residue are presented in Table 4. Threemain stages of weight loss were observed for all lms. The rststage weight loss (Dw1 2.53e3.22%) was observed over thetemperature (Td1) ranging from 47.88 to 50.86 C, possibly associ-atedwith the loss of freewater adsorbed in the lm, as well as othervolatile compounds absorbed in the lm. The similar result wasfound in cuttlesh skin gelatin lm (Hoque et al., 2011) and porcineplasma protein lm added with different cross-linking agents(Nuthong et al., 2009). The second stage weight loss (Dw2 24.39e28.61%) appeared at the onset temperature of 192.47e202.45 C(Td2), depending on the lm samples. This transition revealed thedegradation of lm matrix. This was most likely due to the degra-dation or decomposition of lower molecular weight protein frac-tions and glycerol compounds (Tongnuanchan et al., 2012). For thethird stage of weight loss (Dw3 43.76e46.35%), Td3 of 293.5e307.34 C were observed for all lms but varied with lm samples.This was possibly caused by the loss or decomposition of larger-sizeor highly interacted proteins and high temperature stable compo-nents in lm matrix.

It was noted that slightly lower rst stage weight loss was foundin lm incorporated with BEO. Lower Td2 and Td3 with coincidental

0

10

50 100 150 200 250 300 350 400 450 500 550 600

Temperature ( C)

Fig. 3. TGA curves of FPI/FSG lms as inuenced by BEO and ZnONP incorporation.plasticising effect of BEO, which might impede interaction betweenprotein molecules in the lm network. A reduction in thermalstability can be promoted by changes in the protein structure andprovoked by the rupture of low energy intermolecular bonds which

Table 4Thermal degradation temperature (Td, C) and weight loss (Dw, %) of FPI/FSG lms asaffected by 100% BEO and 3% ZnONP incorporation.

ZnONPlevel(% w/w)

Essentialoil level(% w/w)

D1 D2 D3 Residue(%)

Td1,onset

Dw1 Td2,onset

Dw2 Td3,onset

Dw3

0 0 47.88 3.22 197.04 26.23 301.65 45.72 24.83100 48.39 3.03 192.47 28.61 293.50 46.35 22.01

3 0 50.06 2.73 202.45 24.39 307.34 43.76 29.12100 50.86 2.53 198.50 26.07 303.93 44.24 27.16

D1, D2, and D3 denote the rst, second and third stage weight loss, respectively, oflm during heating scan.

y BE

rocomaintain the protein conformation (Kaminska & Sionkowska,1999). Increase in weight loss on the second and third stages oflm incorporated with all essential oils, especially at 100% level,was plausibly associated with the lower protein interaction(Tongnuanchan et al., 2013). However, the lms with 3% ZnONPshowed enhanced thermal stability in comparison with the lmswithout ZnONP, irrespective of BEO incorporation, as evidenced bythe higher Td2 and Td3 and lower weight loss for Dw2 and Dw3.ZnONP in the protein matrix could insert within lm matrix oracted as thermal insulator or mass transport barrier to the volatileproducts generated during thermal decomposition. This resulted inthe improved thermal stability of the composites. Carboxymeth-ylcellulose/ZnO nanocomposite lm with a high concentration ofZnONP showed enhanced thermal stability in comparison withcontrol lm (Espitia et al., 2013).

Additionally, all lms had residual mass (representing charcontent) at 600 C in the range of 22.01e29.12%. The highest charcontent observed in nanocomposite lm without BEO was mostlikely ascribed to the high thermal stability of the ZnO nano-particles and the highest bondings formed in the proteinnetwork. In general, the lower residue (or char content) fromthermal degradation was found in lms incorporated with BEO,compared with control lm, regardless of ZnONP addition. Thus,both BEO and ZnONP had the marked impact on thermal stabilityof FPI/FSG lms.

3.2.3. Film morphologySEM micrographs of the surface and freeze-fractured cross-

Fig. 4. SEM micrographs of surface and cross-section of FPI/FSG lms as inuenced bsection, respectively.

Y.A. Arfat et al. / Food Hyd272section of FPI/FSG lms without and with 100% BEO in the absenceand presence of 3% ZnONP are illustrated in Fig. 4. The slightlyrough surface was noticeable in the control FPI/FSG blend lm(without BEO and ZnONP). No obvious separation was observed inthe matrix of FPI/FSG blend lm, indicating the compatibility be-tween FPI and FSG. Roughness of surface structure was more pro-nounced in lms incorporated with BEO and ZnONP than thatfound in the control lm. The increases in the roughness could bedue to the distribution of ZnONP as well as BEO droplet throughoutthe lmmatrix. This might be also associated with the coexisting ofprotruded lm structure as indicated by the increased thickness ofresulting lms (Table 1). Furthermore, nanocomposite lmsshowed distinctive ZnONP images on their lms surfaces as evi-denced by the appearance of white particles on the surface andcross-section of lms. Those ZnONP were uniformly dispersed inthe nanocomposite lms, leading to effective force transfer fromproteinmatrix to ZnONP reinforcing phase. This could be associatedwith the higher TS of FPI/FSG nanocomposite lm with 3% ZnONP,compared with other lms (Table 1).FPI/FSG lms incorporated with 100% BEO showed the phaseseparation between oil phase (upper) and protein phase (lower).The higher content of essential oil droplets could enhance creamingand phase separation. BEO with low density at high content mightbe separated and localised at the upper surface of lm, therebyforming the bilayermicrostructure (Tongnuanchan et al., 2013). Thehydrophobic oil droplet could hinder the water molecules totransfer through the lm (Karbowiak, Debeaufort, & Voilley, 2007),thus contributing to lower WVP of BEO incorporated lms,compared with the control lm, as shown in Table 1. However, nophase separation was observed in FPI/FSG lms incorporated withBEO in the presence of ZnONP. Nanoparticles are shown to behighly effective emulsifying agents due to their strong adsorption atoil-water and airewater interface (Larson-Smith & Pozzo, 2012).

4. Conclusion

The active nanocomposite lms based on FPI/FSG blend incor-porated with BEO and ZnONPwere developed. The developed lmsshowed the improved mechanical and water vapour barrier prop-erties as well as thermal stability, compared with control lms(without BEO and ZnONP). Moreover, FPI/FSG lms incorporatedwith BEO greatly inhibited the growth of Gram-positive and Gram-negative foodborne pathogenic and spoilage bacteria. Antimicro-bial activities of BEO incorporated FPI/FSG lms were more pro-nounced in the presence of ZnONP. Thus, FPI/FSG lmsincorporated with BEO and ZnONP could be used as an active foodpackaging to prevent growth of pathogen and spoilage bacteria in

O and ZnONP incorporation. Magnication: 7000 and 3500 for surface and cross-

lloids 41 (2014) 265e273foods.

Acknowledgements

The authors would like to express their sincere thanks to Princeof Songkla University and the TRF Senior Research Scholar pro-gramme for the nancial support.

References

Ahmad, M., Benjakul, S., Prodpran, T., & Agustini, T. W. (2012). Physico-mechanicaland antimicrobial properties of gelatin lm from the skin of unicorn leath-erjacket incorporated with essential oils. Food Hydrocolloids, 28, 189e199.

Alebooyeh, R., Nafchi, A. M., & Jokar, M. (2012). The effects of ZnO nanorods on thecharacteristics of sago starch biodegradable lms. Journal of Chemical HealthRisks, 2, 13e16.

Allahverdiyev, A. M., Kon, K. V., Abamor, E. S., Bagirova, M., & Rafailovich, M. (2011).Coping with antibiotic resistance: combining nanoparticles with antibiotics andother microbial agents. Expert Review of Anti Infective Therapy, 9, 1035e1052.

American Society for Testing and Materials. (1989). Annual book of ASTM standards.Philadelphia, PA, USA: ASTM.

AOAC. (2000). Ofcial methods of analysis. Washington, DC: Association of OfcialAnalysis Chemists.

Arfat, Y. A., Benjakul, S., Prodpran, T., & Osako, K. (2014). Development and char-acterisation of blend lms based on sh protein isolate and sh skin gelatin.Food Hydrocolloids, 39, 58e67.

Atares, L., Bonilla, J., & Chiralt, A. (2010). Characterization of sodium caseinate-basededible lms incorporated with cinnamon or ginger essential oils. Journal of FoodEngineering, 100, 678e687.

Nuthong, P., Benjakul, S., & Prodpran, T. (2009). Characterization of porcine plasmaprotein-based lms as affected by pretreatment and cross-linking agents. In-ternational Journal of Biological Macromolecules, 44, 143e148.

Pires, C., Ramos, C., Teixeira, G., Batista, I., Mendes, R., Nunes, L., et al. (2011).Characterization of biodegradable lms prepared with hake proteins and thymeoil. Journal of Food Engineering, 105, 422e428.

Ponce, A. G., Roura, S. I., del-Valle, C. E., & Moreira, M. R. (2008). Antimicrobial andantioxidant activities of edible coatings enriched with natural plant extracts:in vitro and in vivo studies. Postharvest Biology and Technology, 49, 294e300.

Prodpran, T., Benjakul, S., & Artharn, A. (2007). Properties and microstructure of

Y.A. Arfat et al. / Food Hydrocolloids 41 (2014) 265e273 273Bozin, B., Mimica-Dukic, N., Simin, N., & Anackov, G. (2006). Characterization of thevolatile composition of essential oil of some lamiaceae species and the anti-microbial and antioxidant activities of the entire oils. Journal of Agriculture andFood Chemistry, 54, 1822e1828.

Burt, S. (2004). Essential oils: their antibacterial properties and potential applica-tions in foods-a review. International Journal Food Microbiology, 94, 223e253.

Chinabhark, K., Benjakul, S., & Prodpran, T. (2007). Effect of pH on the properties ofprotein-based lm from bigeye snapper (Priacanthus tayenus) surimi. Bio-resource Technology, 98, 221e225.

Emiroglu, Z. K., Yemis, G. P., Coskun, B. K., & Candogan, K. (2010). Antimicrobialactivity of soy edible lms incorporated with thyme and oregano essential oilson fresh ground beef patties. Meat Science, 86, 283e288.

Espitia, P. J. P., Soares, N. F. F., Telo, R. F., Coimbra, J. S. R., Vitor, D. M., Batista, R. A.,et al. (2013). Physical-mechanical and antimicrobial properties of nano-composite lms with pediocin and ZnO nanoparticles. Carbohydrate Polymers,94, 199e208.

Gennadios, A., Weller, C. L., Hanna, M. A., & Froning, G. W. (1996). Mechanical andbarrier properties of egg albumin lms. Journal of Food Science, 61, 585e589.

Guillen, M. D., & Cabo, N. (2004). Study of the effects of smoke avourings on theoxidative stability of the lipids of pork adipose tissue by means of Fouriertransform infrared spectroscopy. Meat Science, 66, 647e657.

Han, J. H., & Floros, J. D. (1997). Casting antimicrobial packaging lms andmeasuring their physical properties and antimicrobial activity. Journal of PlasticFilm and Sheeting, 13, 287e298.

Hoque, M. S., Benjakul, S., Prodpran, T., & Songtipya, P. (2011). Properties of blendlm based on cuttlesh (Sepia pharaonis) skin gelatin and mungbean proteinisolate. International Journal of Biological Macromolecules, 49, 663e673.

Hussain, A. I., Anwar, F., Hussain-Sherazi, S. T., & Przybylski, R. (2008). Chemicalcomposition, antioxidant and antimicrobial activities of basil (Ocimum basili-cum) essential oils depends on seasonal variations. Food Chemistry, 108, 986e995.

Iwata, K., Ishizaki, S., Handa, A., & Tanaka, M. (2000). Preparation and character-ization of edible lms from sh water-soluble proteins. Fisheries Science, 66,372e378.

Kaminska, A., & Sionkowska, A. (1999). The effect of UV radiation on the values ofthermal parameters of collagen containing b-carotene. Polymer Degradation andStability, 65, 87e90.

Kanmani, P., & Rhim, J. W. (2014). Physical, mechanical and antimicrobial propertiesof gelatin based active nanocomposite lms containing AgNPs and nanoclay.Food Hydrocolloids, 35, 644e652.

Karbowiak, T., Debeaufort, F., & Voilley, A. (2007). Inuence of thermal process onstructure and functional properties of emulsion-based edible lms. Food Hy-drocolloids, 21, 879e888.

Kovacevic, V., Vrsaljko, D., Blagojevic, S. L., & Leskovac, M. (2008). Adhesion pa-rameters at the interface in nanoparticulate lled polymer systems. PolymerEngineering and Science, 48, 1994e2002.

Larson-Smith, K., & Pozzo, D. C. (2012). Pickering emulsions stabilized by nano-particle surfactants. Langmuir, 28, 11725e11732.

Limpisophon, K., Tanaka, M., & Osako, K. (2010). Characterisation of gelatin-fattyacid emulsion lms based on blue shark (Prionace glauca) skin gelatin. FoodChemistry, 122, 1095e1101.

Ma, X. Y., & Zhang, W. D. (2009). Effects of ower-like ZnO nanowhiskers on themechanical, thermal and antibacterial properties of waterborne polyurethane.Polymer Degradation and Stability, 94, 1103e1109.

Mohamed, A. A., El-Emary, G. A., & Ali, H. F. (2010). Inuence of some citrus essentialoils on cell viability, glutathione-s-transferase and lipid peroxidation in ehrli-chascites carcinoma cells. Journal of American Science, 6, 820e826.

Muik, B., Lendl, B., Molina-Diaz, A., Valcarcel, M., & Ayora-Canada, M. J. (2007). Twodimensional correlation spectroscopy and multivariate curve resolution for thestudy of lipid oxidation in edible oils monitored by FTIR and FT-Raman spec-troscopy. Analytica Chimica Acta, 593, 54e67.

Muyonga, J. H., Cole, C. G. B., & Duodu, K. G. (2004). Characterisation of acid solublecollagen from skins of young and adult Nile perch (Lates niloticus). FoodChemistry, 85, 81e89.

Nikoo, M., Xu, X., Benjakul, S., Xu, G., Ramirez-Suarez, J. C., Ehsani, A., et al. (2011).Characterization of gelatin from the skin of farmed Amur sturgeon (Acipenserschrenckii). International Aquatic Research, 3, 135e145.protein-based lm from round scad (Decapterus maruadsi) muscle as affectedby palm oil and chitosan incorporation. International Journal of BiologicalMacromolecules, 41, 605e614.

Rhim, J. W., & Ng, P. K. W. (2007). Natural biopolymer-based nanocomposite lmsfor packaging applications. Critical Reviews in Food Science and Nutrition, 47,411e433.

Rouhi, J., Mahmud, S., Naderi, N., Ooi, C. H. R., & Mahmood, M. R. (2013). Physicalproperties of sh gelatin-based bio-nanocomposite lms incorporated withZnO nanorods. Nanoscale Research Letters, 8, 364e368.

Snchez-Gonzlez, L., Pastor, C., Vargas, M., Chiralt, A., Gonzlez-Martnez, C., &Chfer, M. (2011). Effect of hydroxypropylmethylcellulose and chitosan coatingswith and without bergamot essential oil on quality and safety of cold-storedgrapes. Postharvest Biology and Technology, 60, 57e63.

Sharma, P. C., Jain, S., Yadav, G., & Sharma, A. (2010). Natural preservatives: currentinsights and applications. Der Pharmacia Sinica, 1, 95e108.

Shiku, Y., Hamaguchi, P. Y., Benjakul, S., Visessanguan, W., & Tanaka, M. (2004).Effect of surimi quality on properties of edible lms based on Alaska pollack.Food Chemistry, 86, 493e499.

Sothornvit, R., Hong, S. I., An, D. J., & Rhim, J. W. (2010). Effect of clay content on thephysical and antimicrobial properties of whey protein isolate/organo-claycomposite lms. LWT-Food Science and Technology, 43, 279e284.

Sothornvit, R., Rhim, J. W., & Hong, S. I. (2009). Effect of nano-clay type on thephysical and antimicrobial properties of whey protein isolate/clay compositelms. Journal of Food Engineering, 91, 468e473.

Steel, R. G. D., & Torrie, J. H. (1980). Principle and procedure of statistics (2nd ed.).New York: McGraw-Hill.

Suppakul, P., Miltz, J., Sonneveld, K., & Bigger, S. W. (2003). Antimicrobial propertiesof basil and its possible application in food packing. Journal of Agriculture andFood Chemistry, 51, 3197e3207.

Tam, K. H., Djurisic, A. B., Chan, C. M. N., Xi, Y. Y., Tse, C. W., Leung, Y. H., et al. (2008).Antibacterial activity of ZnO nanorods prepared by a hydrothermal method.Thin Solid Films, 516, 6167e6174.

Tanaka, M., Ishizaki, S., Suzuki, T., & Takai, R. (2001). Water vapor permeability ofedible lms prepared from sh water soluble proteins as affected by lipid type.Journal of the Tokyo University of Fisheries, 87, 31e37.

Tayel, A. A., El-Tras, W. F., Moussa, S., El-Baz, A. F., Mahrous, H., Salem, M. F., et al.(2011). Antibacterial action of zinc oxide nanoparticles against foodbornepathogens. Journal of Food Safety, 31, 211e218.

Tongnuanchan, P., Benjakul, S., & Prodpran, T. (2012). Properties and antioxidantactivity of sh skin gelatin lm incorporated with citrus essential oils. FoodChemistry, 134, 1571e1579.

Tongnuanchan, P., Benjakul, S., & Prodpran, T. (2013). Physico-chemical properties,morphology and antioxidant activity of lm from sh skin gelatin incorporatedwith root essential oils. Journal of Food Engineering, 117, 350e360.

Tongnuanchan, P., Benjakul, S., Prodpran, T., & Songtipya, P. (2011). Characteristics oflm based on protein isolate from red tilapia muscle with negligible yellowdiscoloration. International Journal of Biological Macromolecules, 48, 758e767.

Toyohara, H., Sakata, T., Yamashita, M., & Shimizu, Y. (1990). Degradation of ovalleshmeat gel caused bymyobrillar proteinase(s). Journal of Food Science, 55, 364e368.

Turina, A. D. V., Nolan, M. V., Zygadlo, J. A., & Perillo, M. A. (2006). Natural terpenes:self-assembly and membrane partitioning. Biophysical Chemistry, 122, 101e113.

Yakimets, I., Wellner, N., Smith, A. C., Wilson, R. H., Farhat, I., & Mitchell, J. (2005).Mechanical properties with respect to water content of gelatin lms in glassystate. Polymer, 46, 12577e12585.

Yu, J., Yang, J., Liu, B., & Ma, X. (2009). Preparation and characterization of glycerolplasticized-pea starch/ZnO-carboxymethylcellulose sodium nanocomposites.Bioresource Technology, 100, 2832e2841.

Zhang, L., Ding, Y., Povey, M., & York, D. (2008). ZnO nanouids e a potentialantibacterial agent. Progress in Natural Science, 18, 939e944.

Zhang, L., Jiang, Y., Ding, Y., Daskalakis, N., Jeuken, L., Povey, M., et al. (2010).Mechanistic investigation into antibacterial behaviour of suspensions ofZnO nanoparticles against E. coli. Journal of Nanoparticle Research, 12, 1625e1636.

Properties and antimicrobial activity of fish protein isolate/fish skin gelatin film containing basil leaf essential oil an ...1 Introduction2 Materials and methods2.1 Chemicals2.2 Collection and preparation of fish sample2.3 Preparation of fish protein isolate2.4 Preparation of fish protein isolate/fish skin gelatin film added with BEO and ZnONP2.5 Determination of properties and antimicrobial activity of films2.5.1 Film thickness2.5.2 Mechanical properties2.5.3 Water vapour permeability (WVP)2.5.4 Colour2.5.5 Light transmittance and transparency value2.5.6 Antimicrobial properties

2.6 Characterisation of the selected films2.6.1 Attenuated total reflectance-Fourier transform infrared (ATR-FTIR) spectroscopy2.6.2 Thermo-gravimetric analysis (TGA)2.6.3 Scanning electron microscopy (SEM)

2.7 Statistical analysis

3 Results and discussion3.1 Properties of FPI/FSG films incorporated with BEO and ZnONP3.1.1 Thickness3.1.2 Mechanical properties3.1.3 Water vapour permeability (WVP)3.1.4 Colour3.1.5 Light transmittance and transparency value3.1.6 Antimicrobial activity of FPI/FSG films incorporated with BEO and ZnONP

3.2 Characteristics of FPI/FSG films incorporated with BEO and ZnONP3.2.1 FTIR spectroscopy3.2.2 Thermo-gravimetric analysis (TGA)3.2.3 Film morphology

4 ConclusionAcknowledgementsReferences