(e

eneirCTA)zilorian

Article history: Active biolms of quinoa (Chenopodium quinoa, W.) starch were prepared by incorporating gold nanopar-

edible lms based on biopolymers, such as polysaccharides and

commonly used agricultural raw material for edible lm manufac-turing because it is inexpensive, relatively easy to handle, totallybiodegradable, and widely available in nature from varioussources, such as cereals, roots, tubers (Nascimento et al., 2012),and more recently rediscovered pseudocereals, such as amaranthand quinoa (Araujo-Farro, Podadera, Sobral, & Menegalli, 2010).

treatment.ecessary to selectes for productionthe develoof these baiming at p

ing specic properties; as a result, the resulting biolms areactive biolms or active packaging. The promising biolms ibiolms with antimicrobial activity (Kechichian, Ditcheld,Santos, & Tadini, 2010).

Nanocomposites with antimicrobial function are highly usefulfor the minimisation of the growth of contaminant microorgan-isms during the processing or storage of food and thereby theextension of shelf life and improvement of food safety (Rhim,Wang, & Hong, 2013). One of the most widely studied nanocom-posites used in antimicrobial food packaging is based on the

Corresponding author. Tel.: +55 051 3308 9789.E-mail address: simone.ores@ufrgs.br (S.H. Flres).

Food Chemistry 173 (2015) 755762

Contents lists availab

Food Che

lseproteins (Andreuccetti, Carvalho, Galicia-Garca, Martnez-Bustos,& Grosso, 2011; Kanmani & Rhim, 2014; Mei, Yuan, Wu, & Li,2013; Nascimento, Calado, & Carvalho, 2012; Souza et al., 2012;Souza, Goto, Mainardi, Coelho, & Tadini, 2013). Starch, a renewablebiopolymer consisting of amylose and amylopectin, is the most

In order to maintain the quality of foods, it is nthe correct materials and appropriate technologiof the packaging. Thus, current trends includeof packaging that interacts with food. Manymay be incorporated with different compoundshttp://dx.doi.org/10.1016/j.foodchem.2014.10.0680308-8146/ 2014 Elsevier Ltd. All rights reserved.pmentiolmsrovid-calledncludeVeiga-1. Introduction

Interest in the maintenance and/or improvement of the qualityof packaged products and the reduction of waste packaging hasencouraged the exploration of new packaging materials, such asbiodegradable lms formulated with raw materials derived fromrenewable sources, called biolms.

In recent years, many researchers focused on the production of

The quinoa seed (Chenopodium quinoa, Willdenow) is a graintypically found in the South American Andean highlands. It is com-posed of signicant amounts of starch (up to 80%), has an amylosecontent of 1021% (depending on the variety), and a small starchgranule size (1 lm), which are characteristics that allow its easierdispersion and thus make this starch a promising material for lm(Araujo-Farro et al., 2010). This starch may be able to form trans-parent biodegradable edible lms without any prior chemicalReceived 4 June 2014Received in revised form 9 October 2014Accepted 14 October 2014Available online 28 October 2014

Keywords:NanocompositeAntibacterialFood packagingMetal nanoparticlesQuinoa starchticles stabilised by an ionic silsesquioxane that contains the 1,4-diazoniabicyclo[2.2.2]octane chloridegroup. The biolms were characterised and their antimicrobial activity was evaluated against Escherichiacoli and Staphylococcus aureus. The presence of gold nanoparticles produces an improvement in themechanical, optical and morphological properties, maintaining the thermal and barrier propertiesunchanged when compared to the standard biolm. The active biolms exhibited strong antibacterialactivity against food-borne pathogens with inhibition percentages of 99% against E. coli and 98% againstS. aureus. These quinoa starch biolms containing gold nanoparticles are very promising to be used asactive food packaging for the maintenance of food safety and extension of 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 tDevelopment of active biolms of quinoastarch containing gold nanoparticles andactivity

Carlos H. Pagno a, Tania M.H. Costa b, Eliana W. de MCarla R. Matte a, Juliano V. Tosati c, Alcilene R. Monta Laboratrio de Compostos Bioativos, Instituto de Cincia e Tecnologia dos Alimentos (Ib Instituto de Qumica, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, BracDepartamento de Engenharia de Alimentos, Universidade Federal de Santa Catarina, Fl

journal homepage: www.eChenopodium quinoa W.)valuation of antimicrobial

ezes b, Edilson V. Benvenutti b, Plinho F. Hertz a,o c, Alessandro O. Rios a, Simone H. Flres a,, Universidade Federal do Rio Grande do Sul (UFRGS), Porto Alegre, RS, Brazil

polis, SC, Brazil

le at ScienceDirect

mistry

vier .com/locate / foodchem

ever, as far as we know, the development of starch biolms incor-

The aqueous gold nanoparticles dispersion was obtained using

ence solution, without metal addition, was also prepared, and it

The biolms were dried in an oven with forced air circulation (DeL-

thickness was measured using a digital micrometer (DIGIMESS

emi2. Materials and methods

2.1. Quinoa starch production

Quinoa starch was extracted from quinoa seeds (C. quinoaWill-denow), purchased at the Municipal Market, Bolivia Glycerol(MARK), using the methodology developed by Araujo-Farro et al.(2010). The seeds were immersed in distilled water at a ratio of1:2 and refrigerated for 8 h. The softened grain suspension wasmilled in a kitchen blender (ARNO Mod.wwB3 400W, Brazil), andthe resulting suspension was removed by sieves (80, 150 mesh)and washed with distilled water. After, the material was centri-fuged (1000 g/15 min/4 C) (HITACHI High-Speed Refrigerated CR21GIII, Ibaraki, Japan), and the supernatant was removed; this pro-cedure was repeated three times. The starch was suspended inaqueous 0.20% (w/w) NaOH solution at an alkaline pH of 10.5,gently stirred for 5 min to solubilise the proteins, and centrifuged.The starch was redispersed in deionised water and neutralised bythe addition of 1 mol L1 HCl. Afterwards, the starch was centri-fuged and redispersed in water 5 times to remove residual proteinand salt. The puried starch was then frozen, lyophilised, andstored refrigerated until use.

2.2. Proximate analysis of the quinoa starch

The contents of protein, ash, ether extractable lipids, total bre,and water were determined using the standard AOAC methodsAOAC (1995). The total protein content was determined by theKjeldahl method using a correction factor of 6.25. The lipid contentwas determined using a Soxhlet extractor (Foss Soxtec, modelporated with gold nanoparticles as antimicrobial agent has notbeen previously described. The proposal for synthesis of gold nano-particles in the present work is based on the reduction of gold saltsin solution containing the ionic silsesquioxane as stabiliser. Thisstabiliser is essential for the nanoparticle synthesis additionally itcontains quaternary ammonium groups that promote water solu-bility and are known for their inhibitory and antimicrobial effectmaking this system very promising in the preparation of bioactivelms.

The mechanical properties of the lms are also very importantbecause these allow the handling, storage, and transport of thefood without damage. It was reported that the incorporation ofmetal nanoparticles can affect the mechanical properties of thebiolms, however a more precise approach of this inuence is nec-essary. This evaluation is of highest importance, since fragile pack-ages may be inappropriate for good food storage (Yoksan &Chirachanchai, 2010).

In this context, this work aims at developing a biodegradablebiolm of quinoa starch, with active packaging by the addition ofgold nanoparticles (AuNPs) stabilised with an ionic silsesquioxanecontaining the organic group 1,4-diazoniabicyclo[2.2.2]octanechloride, and evaluating their physical, mechanical and microbio-logical properties.incorporation of silver nanoparticles (AgNPs) into biolms(Incoronato, Buonocore, Conte, Lavorgna, & Nobile, 2010; Llorens,Lloret, Picouet, Trbojevich, & Fernandez, 2012; Rhim et al., 2013;Yoksan & Chirachanchai, 2010). Other metallic nanocomposites,such as zinc oxide (Emamifar, Kadivar, Shahedi, & Soleimanian-Zad, 2011), titanium oxide (Bodaghi et al., 2013), and copper(Ciof et al., 2005), have also been incorporated in biolms. How-

756 C.H. Pagno et al. / Food Ch2055, Denmark). The ash content was determined in a mufe fur-nace (Elektro Therm Linn, 312.6 SO LM 1729, Germany) set to550 C. The moisture content was determined by maintaining thePrecision 0.001 mm, resolution/0 mm 25 mm, Brazil) at ve ran-dom positions in each strip. The tensile strength (TS) [MPa] andpercent elongation at break (E) [%] were evaluated through a ten-sile test performed on a texture analyser (TA.XT2i e Stable MicroSystems, UK) with a load cell of 5 kg using the A/TGT self-tighten-ing roller grips xture, according to ASTM D882-09 (2009). Tenstrips were cut, and each one was mounted between the grips ofeo, model B5AFD, Brazil) at 35 C for 16 h.

2.5. Characterisation of quinoa starch biolms

The quinoa starch biolms were characterised by the determi-nation of their mechanical properties, solubility, optical properties(colour and opacity), gas permeability, and water vapour perme-ability, morphology and thermal stability. The biolms were condi-tioned in desiccators under 58% RH at 25 C for 48 h before beingsubjected to analyses (Pelissari, Andrade-Mahecha, Sobral, &Menegalli, 2013).

2.5.1. Mechanical propertiesThe biolms were cut into strips (100 mm25 mm), and theirwas designated as blank.

2.4. Preparation of starch quinoa biolms

The biolms were produced through the casting technique(Fakhouri et al., 2013). The lm-forming solution was preparedwith a suspension of 4% starch from quinoa (4 g/100 g of total lmsolution). The dispersion of quinoa starch was gelatinised at 82 Cfor 30 min with constant stirring in a water bath (DeLeo B450).Glycerol was then added to a concentration of 1%, and the resultingmixture was considered the standard formulation. Five types ofbiolms were produced: standard lm (lm with starch only), bio-lms containing 5% (v/v) of gold nanoparticles dispersion (AuNPs5)and 2.5% (v/v) of gold nanoparticles dispersion (AuNPs2.5), andtwo control biolms containing 5% (v/v) (C5) and 2.5% (v/v)(C2.5) of the blank solution. An amount of 0.24 g cm2 of thelm-forming solution was then poured evenly onto acrylic plates.as stabiliser a ionic silsesquioxane containing the divalent 1,4-diazoniabicyclo[2.2.2]octane chloride group, synthesised asdescribed by Arenas et al. (2008). For the preparation of gold nano-particles, aqueous solution (30 mL) containing 600 mg of ionic sils-esquioxane was mixed with 7.5 mL of 5 103 mol L1 HAuCl4. Tothis mixture, it was added 30 mL of 0.02 mol L1 NaBH4 freshlyprepared, always under stirring. The aqueous colloidal dispersionof gold nanoparticles, stabilised by ionic silsesquioxane was desig-nated by AuNPs. For comparison with AuNPs dispersion, a refer-lyophilised sample at 105 C (Oven DeLeo, model 48 TLK, Brazil)for approximately 4 h, and the measurement was determinedthrough the weight difference. The carbohydrate content was cal-culated by subtracting from 100 the sum of the percentages ofwater, protein, lipid, ash and dietary bre. All of the analyses wereperformed in triplicate. The results are expressed as grams per100 g of dry matter (DM).

2.3. Obtaining a gold nanoparticle dispersion

stry 173 (2015) 755762the equipment for testing; the initial distance between the gripsand the test speed were set to 50 mm and 0.8 mm s1, respectively(Souza et al., 2013).

emis2.5.2. Water vapour permeability (WVP)The WVP was determined gravimetrically according to the

method described by Mei et al. (2013) with some modications.The samples were placed in permeation cells (inner diame-ter = 63 mm, height = 25 mm) lled with granular anhydrous cal-cium chloride and hermetically sealed. The permeation cells wereplaced in a glass chamber with a saturated sodium chloride solu-tion to obtain RH gradients of 0%/75% at 25 C. The mass gainwas determined by weighing the cell permeation on an analyticalbalance (AY 220, Shimadzu) at intervals of 1 h during the rst24 h period and of 12 h after 24 h. The water vapour permeabilityof the samples was determined in triplicate using Eq. (1):

WVP w LA t Dp 1

whereW is the weight of the water that permeated through the lm(g), L is the thickness of the lm (m), A is the permeation area (m2), tis the time of permeation (h), and Dp is the water vapour pressuredifference between the two sides of the lm (Pa).

2.5.3. SolubilityThe solubility was calculated as the percentage of dry matter in

the lm solubilised after immersion for 24 h in water at 25 C.Discs of the lm (2 cm in diameter) were cut, weighed, immersedin 30 mL of distilled water, and slowly and periodically agitated.The amounts of dry matter in the initial and nal samples weredetermined by drying the samples at 105 C for 24 h. The solubilitywas calculated using Eq. (2) (Pelissari et al., 2013):

DE 100 wiwfwi

2

where wi is the initial dry weight of the sample (g), and wf is thenal dry weight of the sample (g).

2.5.4. Optical properties (colour and opacity)The opacity was determined by measuring the lm absorbance

at 210 and 500 nm using a UV spectrophotometer (Shimadzu UV-1800). The biolms were cut into a rectangle and directly placed ina spectrophotometer test cell. An empty test cell was used as thereference. The opacity of the biolms was calculated by dividingthe values of the absorbance (nm) by the thickness of the biolm(mm) (Wang, Dong, Men, Tong, & Zhou, 2013).

The colour of the biolms was determined with a colorimeter(Hunter Lab system, model Miniscan XE, USA) operated with D65(daylight) and using the CIELab colour parameters. The parametersL (luminosity), a (red-green), and b (yellow-blue) were deter-mined. A white disk (L0 = 97.5; a0 = 0.13 and b0 = 1.7) was used asthe standard. The colour difference (DE) compared with a whitestandard was calculated using Eq. (3) (Rotta et al., 2009):

DE DL2

q Da2 Db2 3

In Eq. (3), DL = L L0, Da = a a0, Db = b b0, where L0, a0,and b0 are the colour values of the standards, and L, a, and b arethe biolm colour values.

2.5.5. Gas permeabilityThe experimental setup used to measure the lm permeability

was designed and built in the lab scale. The apparatus is stainlessand composed of a cylinder, manual on/off valves, micrometricvalves, and two closed chambers, and the temperature and mea-sured pressure are controlled by transducers. The measurementswere performed using the method of the American Society for

C.H. Pagno et al. / Food ChTesting and Materials (ASTM D1434). The sample was mountedin a gas transmission cell to form a sealed semi-barrier betweenthe two chambers. One chamber contains the test gas at a specichigh pressure, and the other chamber, which is maintained at alower pressure, receives the permeating gas. Either of the followingprocedures was used: (1) the lower pressure chamber was initiallyevacuated, and the transmission of the gas through the lm wasindicated by an increase in the pressure; or (2) the high pressurechamber was maintained near atmospheric pressure, and thetransmission of the gas through the lm was indicated by a changethe pressure.

The coefcient of permeability was obtained from a mass bal-ance for O2 and CO2, in chamber the lower pressure measured bypressure transducer and using Eq. (4):

P2n1 P1 P1 P2i exp A R T t PeV l

4

where A is the permeable area (m2), V is the chamber volume (m3), tis the time (s), R is the gas constant (m3 Pa K1 mol1), T is the gastemperature inside the chamber (K), l is the lm thickness (lm), Peis the permeability (mol lmm2 s1 Pa1), P2(n + 1) is the pressureat time (Pa), P1 is the pressure under constant ux (Pa), and P2i is thepressure in the lower-pressure chamber (Pa). The routine programused the software Matlab (R2012a).

2.5.6. Scanning electron microscopy (SEM)SEM images were obtained for standard lm, AuNPs5 and AuN-

Ps2.5. The biolms were cut and pasted in a double sided conduct-ing tape on an aluminium support and coated with a thin lm ofplatinum using a Balted SCD 050 Sputter Coater apparatus (Scotia,New York, USA). The images were obtained using a scanning elec-tron microscope, model JSM 5800 LV, JEOL (Tokyo, Japan), con-nected to a secondary electron detector with energy dispersiveX-ray spectroscopy (EDS). The micrograph was obtained with amagnication of 1000 operating at an accelerating voltage of5 kV. Several images were taken from various parts of the solidphase, to assure the reproducibility of nal image taken as repre-sentative of the whole sample.

2.5.7. Transmission electron microscopy (TEM)TEM images of AuNPs5 and AuNPs2.5 were recorded using a

JEOL JEM-1220 microscope, operated at an acceleration voltage of80 kV. The biolms were cut into small pieces and placed in a ves-sel containing liquid N2. After evaporation of the liquid N2 the bio-lms were crushed, dispersed into isopropanol, using ultrasoundfor 60 min, and two drops of this mixture were used to place ontoa carbon-coated Cu grid, followed by drying at ambient conditions.

2.5.8. Thermogravimetric analysisThe thermogravimetric analysis of the biolms was performed

under argon ow on a Shimadzu Instrument model TGA-50 2, witha heating rate of 10 C min1, from room temperature up to 600 C.

2.6. Antimicrobial activity

The antimicrobial activity of the biolms with metal nanoparti-cles was tested against the Gram-negative bacteria Escherichia coli(ATCC 25972) and Gram-positive bacteria Staphylococcus aureus(ATCC 1901) according to the methodology proposed by Liao,Anchun, Zhu, and Quan (2010) with some modications. Firstly,the microorganism were inoculated in trypticase soy broth (TSB),and incubated at 37 C/24 h, until the exponential growth phasewas reached. After centrifuged and resuspended in saline solution(0.9%). For the assay, the antimicrobial activity of each bacterial theinoculum density was then adjusted to 1 106 colony-forming

1

try 173 (2015) 755762 757units (CFUs mL ) in saline solution (0.9%). The growth was moni-tored at 1 h intervals by O.D.600 nm and by viable cell counts. Thebiolms were cut into squares of 20 mm2 and accommodated in

clo[2.2.2]octane chloride bonded in a bridged way. This stabiliser

emiis a silica based hybrid polymer that presents appreciable water-solubility and, in previous reports, it showed ability to stabilisegold nanoparticles (Nunes et al., 2012). The concentration of ionicsilsesquioxane was 15.6 106 mol mL1, and the concentration ofgold was 18.5 106 g m L1, in the form of nanoparticles. TheAuNPs dispersion presents reddish colour and a UVVis absorptionspectrum is showed in Fig. 4 of Supplementary Material, showing aband with a maximum at 517 nm. The TEM image of nanoparticlesdispersion is also showed in the Supplementary Material Fig. 3. Thenanoparticles presented spherical form and the average diameterof AuNPs, determined by transmission electron microscopy(TEM), using Quantikov software, was of 5.5 nm with a standarddeviation of 1.4 nm.

3.3. Biolm characterisation

The Table 1 shows physical properties of the standard biolm,control biolms and biolms containing gold nanoparticleseppendorf tubes. Then, 500 lL of the bacterial suspensions wasapplied to the biolms with gold nanoparticles and to the controlbiolm. After incubation for 24 h with gentle agitation, an aliquotof 100 lL the bacterial suspensions were transferred separatelyinto tubes containing sterilised saline solution to obtain serial dilu-tions of the suspensions, and the resulting mixtures were vigor-ously mixed. Then, 10 lL of the bacterial solutions from themixtures were spread evenly on TSA (trypticase soy agar) mediumagar plates. The plates were incubated at 37 C in an aerobic Petridish for 10 h. The viable cells on each of the plates were counted byquantifying the CFUs, and the resulting counts were compared.Each test was performed in duplicate. The antibacterial effect ineach group was calculated as the percentage of inhibition, whichwas calculated using Eq. (5):

%inhibition CFU standard CFU experimental groupCFU standard

100

5

2.7. Statistical analysis

The results were evaluated by analysis of variance (ANOVA) andTukeys test at a signicance level of 0.05 using the Statistica 10.0software (STATSOFT Inc.).

3. Results and discussion

3.1. Composition of quinoa starch

The extraction of starch using the proposed methodologyyielded a material containing, approximately 11.39 0.04 mois-ture, and other constituents on a dry basis (db), 3 0.03% totalbre, 0.20 0.001% total ash, 0.02 0.001% total lipids,1.3 0.0002% total proteins.

3.2. AuNPs aqueous dispersion

Metal nanoparticles are thermodynamically unstable in solu-tion and have a tendency to coalesce via particleparticle interac-tions during the preparation process, which demands the use of astabiliser to avoid this interaction (de Menezes et al., 2012). Thenanoparticles used in this study were stabilised with an ionic sils-esquioxane containing the organic group 1,4-diazoniabicy-

758 C.H. Pagno et al. / Food Ch(AuNPs2.5 and AuNPs5). It is possible to observe that thicknessof all developed biolms, was not signicantly different(p < 0.05), ranging from 79.3 5 to 86.7 1.5 lm.3.3.1. Mechanical propertiesThe mechanical properties of standard biolm, control biolms

and biolms containing gold nanoparticles are presented onTable 1. It is possible to observe that the developed biolms showa signicant increase (p < 0.05) in the tensile strength in relation tothe standard biolm, with control biolms showing higher valuesthan the biolms containing AuNPs.

The higher values for TS in the control biolms can be attrib-uted to the presence of the silsesquioxane polymer chain that aredispersed in the lmogenic solution and during the drying processof the biolm they become randomly entangled with the quinoastarch chains, thus increasing the mechanical resistance of the bio-lms. This effect has been very explored in polymeric materialswhere silica based moieties were used as reinforcing llers(Xiong, Tang, Tang, & Zou, 2008). It was observed a decrease inthe TS for the biolms containing gold nanoparticles, AuNPs2.5and AuNPs5. This decrease was interpreted by taking into accountthat, in the presence of gold nanoparticles, the silsesquioxanepolymer acts as stabiliser of the nanoparticles, being partiallyconsumed in this task and consequently decreasing its availabilityin the initial lmogenic solution to perform the entanglementwith the starch chains. No signicant difference was observed(p > 0.05) between the AuNPs5 (9.9 0.7 MPa) and AuNPs2.5(11.0 0.9 MPa) biolms.

It was reported by Kanmani and Rhim (2014) that the mechan-ical properties of gelatin biolms containing silver nanoparticlesresulted in a signicant reduction (p < 0.05) in the TS from35.0 4.7 MPa for the biolm standard to 26.9 3.8 MPa. Accord-ing to these authors, this change may be due to a reduction inthe proteinprotein interactions in the biolms containing thenanoparticles. On the other hand, an inverse effect was reportedwhen silver nanoparticles were incorporated into biolms basedon starch and chitosan (p < 0.05) in the TS from 66.8 3.3 MPafor the standard biolm to 74.6 5.6 MPa for the biolms contain-ing the silver nanoparticles (Yoksan & Chirachanchai, 2010).

Therefore, the TS of the biolms seem to depend on the polymerused in the lmogenic solution, on the stabilisers used for themetal nanoparticles and also on their relative concentration. Thehighest value of TS was obtained for the C5 biolms(22.2 1.1 MPa), and these values are comparable to thosereported for other plastic biolms in the literature, such as high-density polyethylene (HDPE 22 to 23 MPa), low-density polyeth-ylene (LDPE 19 to 44 MPa), polypropylene (PP 31 to 38 MPa),and polylactic acid biolms (PLA 47.6 to 53.1 MPa) (Rhim et al.,2013).

The assessment of the elongation percent E (%) (Table 1)revealed that all of the prepared formulations showed a signicantdecrease (p < 0.05) in this value when compared with the standardbiolm (5.2 0.6%); however, the values obtained for these pre-pared formulations did not differ signicantly (p > 0.05), varyingfrom 2.7 to 3.1 0.6%.

3.3.2. Water vapour permeability (WVP)The standard biolm formulated with quinoa starch showed a

WVP of 0.393 0.054 g mmm2 h1 kPa1 (Table 1), which ishigher than that obtained by Araujo-Farro et al. (2010) for biolmsbased on quinoa starch (0.204 0.12 g mmm2 h1 kPa1). How-ever, the value obtained in the present study was similar to valuesobtained by other authors for biolms produced with cassavastarch: 0.307 0.00012 g mmm2 h1 kPa1 (Nascimento et al.,2012) and 0.283 0.013 g mmm2 h1 kPa1 (Souza et al., 2012).

The comparison of the WVP values of standard biolms withthose of biolms with AuNPs and control biolms revealed no sig-

stry 173 (2015) 755762nicant differences (p > 0.05), demonstrating that the presence ofsilsesquioxane and AuNPs do not interfere with the WVP of thebiolms. On the other hand, it was reported a signicant difference

Table 1Thickness, tensile strength (TS), elongation at break (E), water vapour permeability (WVP), and solubility of standard biolm, control biolms, and biolms containing goldnanoparticles.

Thickness (lm) (TS) [MPa] (E) [%] WVP (g mm/m2 h kPa) Solubility (%)

Standard 86. 7 1.5a 7.6 0.6d 5.2 0.6a 0.393 0.054a 21.51 0.59a

C5 88.3 6.1a 22.2 1.1a 2.7 0.3b 0.375 0.040a 8.22 0.39d

C2.5 85.3 2.3a 15.7 0.9b 2.7 0.2b 0.314 0.083a 14.76 0.31c

AuNPs5 80.3 3.2a 9.9 0.7c 3.0 0.6b 0.338 0.015a 18.75 0.43b

AuNPs2.5 79.3 5.8a 11.0 0.9c 3.1 0.4b 0.321 0.058a 17.90 0.83b

The results are represented as the means standard deviation of the three repetitions for thickness, WVP and solubility and ten for TS e E. Values with the same superscript inthe same column are not signicantly different (p < 0.05), for the standard biolm, control biolms with concentrations of stabiliser 2.5% (C2.5) and 5% (C5), and biolms

Ps5).

C.H. Pagno et al. / Food Chemistry 173 (2015) 755762 759in the WVP between agar standard biolm (1.97 0.31 109g mm2 s1 Pa1) and the formulation obtained with the additionof silver nanoparticles (1.47 0.01 109 g mm2 s1 Pa1) (Rhimet al., 2013). Similar results were found by Kanmani and Rhim(2014), where the control gelatin biolm showed WVP of3.02 0.14 109 g mm2 s1 Pa1, and, after the addition ofsilver nanoparticles, there was a slight decrease to 2.97 0.01 109 g mm2 s1 Pa1.

The WVP values are essential for dening the possible packag-ing applications of biolms. A material that is very permeable towater vapour may be suitable for the packaging of fresh products,whereas a slightly permeable biolmmay be useful for the packag-ing of dehydrated products. In their attempt to develop biolmscomposed of starch with different concentrations of potassiumsorbate for the conservation of fresh pasta, Andrade-Molina et al.(2013) obtained WVP values varying from 6 106 g m1 Pa1day1 to 14 106 g m1 Pa1 day1, which are similar to theaverage value obtained in this work after the appropriate unitconversion 8 0.3 106 g m1 Pa1 day1.

3.3.3. SolubilityThe biolms produced with the dispersion containing the nano-

particles and the control biolms exhibited a decrease in solubilitycompared with that of the standard biolm (21.51 0.59%)(Table 1). Lower solubility values were obtained for biolms C5(8.22 0.39%) and C2.5 (14.76 0.31%). This behavior may berelated to the highest TS values found for these formulations. Itis possible that the entanglement between silsesquioxane and qui-noa starch chains, which helped to improve the mechanical prop-erties, resulted also in a decrease in the biolm solubility.

The comparison of the solubility of biolms composed of ourand starch from different botanical sources with the addition of dif-ferent compounds is a challenge because the solubility is related tomany factors, including the type of material used to form the poly-mer matrix, the type of interaction that occurs in the matrix, theplasticiser used, and the process conditions (Pelissari et al., 2013).

3.3.4. Optical properties (colour and opacity)

containing gold nanoparticles with concentrations of 2.5% (AuNPs2.5) and 5% (AuNThe comparison of the optical properties of biolms (Table 2)revealed that the main difference is that the biolms prepared with

Table 2Optical properties of colour and opacity of standard biolms, control biolms, and biolm

Colour parameters

L a b

Standard 96.09 0.20a 4.93 0.05c 0.65 0.21C5 96.08 0.10a 4.95 0.04c 0.48 0.16C2.5 96.26 0.14a 4.97 0.02c 0.57 0.10AuNPs5 92.23 0.29c 9.29 0.40a 0.95 0.09AuNPs2.5 93.99 0.34b 7.43 0.18b 0.62 0.09

The results are represented as the means standard deviation. Values with the same lettwith concentrations of stabiliser 2.5% (C2.5) and 5% (C5), and biolms containing gold nAuNPs showed higher values of DE, 25.92 1.33 and 13.01 0.48for AuNPs5 and AuNPs2.5, respectively, which are signicantly dif-ferent (p < 0.05) from those obtained for the standard and controlbiolms. No differences were observed between the standard andcontrol biolms. Higher values of DE indicate biolms with highercolour intensity (Rotta et al., 2009). TheDE value obtained for stan-dard biolm (5.35 0.08) was higher than the DE value obtainedby Araujo-Farro et al. (2010) for a biolm composed of starch qui-noa (1.33 0.03). Rhim et al. (2013) developed biolms of agarwith silver nanoparticles and obtained a standard biolm withDE value of 6.42 0.49, similar to the values obtained in the pres-ent work. These authors observed a signicant increase (p < 0.05)inDEwith the increase in the concentration of silver nanoparticles.The AuNPs5 and AuNPs2.5 biolms also showed higher values ofthe parameter a (Table 2), indicating that they have a more red-dish colour compared with the standard and control biolms.

The UVVis absorption spectrum (400700 nm) of AuNPs5,AuNPs2.5 and showed absorption bands with maximum near520 nm for AuNPs5 and AuNPs2.5, similar to the maximumobtained for aqueous dispersion precursor (Section 3.2). These val-ues are typical of gold nanoparticles with diameter lower than10 nm (de Menezes et al., 2012). The characteristic colour of thesegold nanoparticles is due to collective oscillation of the electronsknown as the localised surface plasmon resonance absorption(Willets & Duyne, 2007).

In the analysis of opacity (Table 2), high absorbance values indi-cate less transparency and a high degree of opacity. The evaluationof the opacity in the visible region (500 nm) revealed that, in a gen-eral way, all of the tested biolms showed low absorbance values,indicating their high transparency and low opacity. However, theAuNPs5 and AuNPs2.5 biolms exhibited slightly higher absor-bance values, 2.44 0.19 A mm1 and 2.14 0.16 A mm1, respec-tively, demonstrating that the increase in the concentration ofAuNPs leads to an increase in opacity. Wang et al. (2013) devel-oped biolms of chitosan using polyphenols as antioxidants andnoted an increase in opacity with increasing concentrations ofthe antioxidants. These authors obtained values for the absorbanceat 600 nm ranging from 0.489 0.045 (standard biolm) to2.898 0.158 for the maximum concentration of the antioxidant.

In their evaluation of the opacity of biolms composed of gelatin

s containing gold nanoparticles.

Opacity (A mm1)

DE 210 nm 500 nm

a 5.35 0.08c 35.75 1.99d 1.86 0.16abca 5.37 0.07c 36.93 1.64d 1.48 0.05bca 5.43 0.03c 40.71 1.63c 1.43 0.10ca 25.92 1.33a 57.13 1.04a 2.44 0.19aa 13.01 0.48b 46.72 0.52b 2.14 0.16ab

er are not signicantly different (p > 0.05), for the standard biolm, control biolmsanoparticles with concentrations of 2.5% (AuNPs2.5) and 5% (AuNPs5).

with different concentrations of silver nanoparticles through theanalysis of the reading transmittance at 660 nm, Kanmani andRhim (2014) observed a similar effect: an increase in the concen-tration of silver nanoparticles caused a reduction in the transpar-ency of the biolm from 89.1 0.0% for the standard biolm to49.6 0.1% for the biolm with the maximum concentration of sil-ver nanoparticles.

As shown on the Table 2, the UV band at 210 nm exhibitedhigher absorbance values for biolms containing gold nanoparti-cles, AuNPs5 and AuNPs2.5 when compared to the standard andcontrol biolms (p < 0.05). This result indicates that the biolmscontaining the AuNPs are able to protect against UV radiation,which can accelerate the oxidation process of lipids.

Fig. 2 shows transmission electron microscopy (TEM) imagesobtained with magnication of 200.000, for AuNPs5 and AuN-Ps2.5 biolms. The gold nanoparticles exhibit contrast that allowstheir visualisation. It is possible to observe that they are smallerthan 10 nm and well dispersed, according to UVVis spectra.

3.3.7. Thermal stabilityThe biolms were submitted to thermogravimetric analysis.

The obtained curves are presented in the Fig. 3. It can be seen aweight loss from room temperature to 150 C, which is assignedto the water desorption. The estimated water amount was near10%. The major weight loss occurs between 250 and 350 C, forall samples, and it is ascribed to the organic phase desorption.

760 C.H. Pagno et al. / Food Chemistry 173 (2015) 7557623.3.5. Gas permeabilityThe gas permeabilities of the biolms developed in this work

were less than 50 cm3/dia m2 atm for CO2 and less than 30 cm3/dia m2 atm for O2, which is the minimum detection limit of theequipment used to quantify the permeation of gases. The quinoastarch biolms developed by Araujo-Farro et al. (2010) exhibitedan oxygen permeability value of 4.36 cm3/dia m2 atm, after theappropriate unit conversions. The presence of metal nanoparticlesor nanocomposites in biolms can lead to a greater reduction inpermeability because such elements can be considered an extrabarrier to the permeation of gases and thereby offer a delay in oxy-gen transport due to an increased tortuosity in the oxygen path-way. Biolms with low oxygen permeability are very useful forfood preservation because oxygen can trigger or accelerate oxida-tion and facilitate the growth of aerobic microorganisms, therebylowering food quality and shortening the shelf life. Strategies lead-ing to an increase in the gas barrier properties include the use ofactive oxygen scavengers in the packaging sachets, labels, or thepolymer layers and the use of passive nanocomposites (Llorenset al., 2012).

3.3.6. Morphological propertiesThe scanning electron microscopy (SEM) results (Fig. 1) show

that, in the studied magnication (1000), the biolms containinggold nanoparticles with concentration of 5% AuNPs5 (a), concen-tration of 2.5% AuNPs2.5 (b), standard biolm (c) and the controllms made with stabiliser only (Fig. 5 of Supplementary Material)present compact and uniform structure without the presence ofcracks or blistering. These results show that the addition of thegold nanoparticles and the stabiliser did not alter the morpholog-ical structure of the quinoa starch biolms as far as we couldobserve by SEM at the magnication used. Similar structure wasobserved by Araujo-Farro et al. (2010) for quinoa starch biolm.According to these authors, compact and uniform structure indi-cates a good interaction between the amylose, amylopectin, glyc-erol, and water in the biolm.Fig. 1. Scanning electron microscopy (SEM) images of the biolms containing gold nanobiolm (c).Therefore, all biolms show high thermal stability, at least, up to270 C. Similar results were obtained for gelatin biolms contain-ing silver nanoparticles (Kanmani & Rhim, 2014). It is also possibleto observe that the residual weight was higher for the AuNPs2.5and AuNPs5, since they should present silica inorganic moietydue to the residue of silsesquioxane decomposition.

3.4. Antimicrobial activity

The antimicrobial activity of the biolms is presented in Table 3.It was evaluated against the Gram-positive bacteria S. aureus ATCC1901 and the Gram-negative bacteria E. coli ATCC 25972. With theexception of the control biolm that showed no antimicrobialactivity and was considered as standard, all of the biolms testedin the present study showed an inhibitory effect against the testedbacteria; however, higher rates of inhibition were observed withthe biolms containing the gold nanoparticles (AuNPs5 and AuN-Ps2.5), which exhibited inhibition rates of 99% for E. coli and 98%for S. aureus.

The analysis of the inhibition of the growth of S. aureus by thedifferent biolms revealed that the biolms containing gold nano-particles (AuNPs5 and AuNPs2.5) exert an inhibitory superior effectof approximately 30% compared with that obtained for control bio-lms, C2.5 and C5. This nding demonstrates the efciency of themicrobial inhibition of the biolms containing gold nanoparticlesagainst this Gram-positive microorganism. However for the inhibi-tion grown for E. coli (Gram-negative), the differences are not toopronounced (Table 3). Therefore, the control biolms presentedan appreciable anti-microbial activity without the presence of goldnanoparticles. This activity was attributed to the stabiliser struc-ture that contains quaternary ammonium groups, which the activ-ity was already reported by Schneid et al. (2014). Despite the lowdifference in the microbial inhibition between biolms containinggold nanoparticles and the control biolms, it is worth noting thatthe concentration of nanoparticles in the biolms was lower than0.06% (w/w), demonstrating that low concentrations of gold nano-particles have antimicrobial activity, particularly against S. aureus.particles with concentrations of 5% AuNPs5 (a), 2.5% AuNPs2.5 (b) and standard

emisC.H. Pagno et al. / Food Ch4. Conclusions

In the present work, active biolms of quinoa starch, containinggold nanoparticles stabilised by an ionic silsesquioxane, were suc-cessfully obtained. The addition of gold nanoparticles dispersion tobiolms of quinoa starch was proven to be promising because itresulted in enhancements in the tensile strength, possibly due tothe presence of the silicon based polymer stabiliser of gold nano-particles. An increase in UV radiation absorption and a decreasein the solubility, which provides improved protection to packaged

Fig. 2. Transmission electron microscopy (TEM) images of the biolms containing gol

Fig. 3. Thermogravimetric Analysis (TGA) analyses of standard biolms andbiolms containing gold nanoparticles with concentrations of 2.5% (AuNPs2.5)and 5% (AuNPs5).

Table 3Antimicrobial activity of biolms containing gold nanoparticles and control biolmsagainst E. coli and S. aureus. The antimicrobial activity is represented by the percentgrowth inhibition.

S. aureus (%) E. coli (%)

Standard 0 0AuNPs5 98 99AuNPs2.5 98 99C5 71 93C2.5 70 96

Percentage inhibition of microbial for the standard biolm, control biolms withconcentrations of stabiliser 2.5% (C2.5) and 5% (C5), and biolms containing goldnanoparticles with concentrations of 2.5% (AuNPs2.5) and 5% (AuNPs5).food and enlarging the possibilities of applications. The thermalstability of all biolms is very impressive, near 270 C. Addition-ally, the biolms containing gold nanoparticles showed positiveresults concerning the antimicrobial activity, demonstrating theireffectiveness in the inhibition of the growth of pathogens, particu-larly S. aureus. Therefore, such biolms may be used as an activepackaging system for food that would prolong the shelf life andmaintain the quality of the packaged food during storage anddistribution.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.foodchem.2014.10.068.

References

Andrade-Molina, T. P., Shirai, M. A., Grossmann, M. V. E., & Yamashita, F. (2013).Active biodegradable packaging for fresh pasta. LWT Food Science andTechnology, 54(1), 2529.

Andreuccetti, C., Carvalho, R. A., Galicia-Garca, T., Martnez-Bustos, F., & Grosso, C.R. F. (2011). Effect of surfactants on the functional properties of gelatin-basededible lms. Journal of Food Engineering, 103(2), 129136.

AOAC. (1995). Association of ofcial analytical chemist: Ofcial methods of analysis(16th ed.). Washington, D.C.

d nanoparticles with concentrations of 5% AuNPs5 (a) and 2.5% AuNPs2.5 (b).

try 173 (2015) 755762 761Araujo-Farro, P. C., Podadera, G., Sobral, P. J. A., & Menegalli, F. C. (2010).Development of lms based on quinoa (Chenopodium quinoa, Willdenow)starch. Carbohydrate Polymers, 81(4), 839848.

Arenas, L. T., Pinheiro, A. C., Ferreira, J. D., Livotto, P. R., Pereira, V. P., Gallas, M. R.,et al. (2008). Anisotropic self-organization of hybrid silica based xerogelscontaining bridge positively charged 1,4-diazoniabicycle[2.2.2]octane chloridegroup. Journal of Colloid and Interface Science, 318, 96102.

ASTM D1434 (2009) Standard Test Method for Determining Gas PermeabilityCharacteristics of Plastic Film and Sheeting. D1434-82(2009) e1.

ASTM Standard E D 882 e 09. (2009). Standard test method for tensile properties ofthin plastic sheeting. West Conshohocken, PA: ASTM International. doi: http://dx.doi.org/10.1520/D0882e09. www.astm.org/Standards.

Bodaghi, H., Mosto, Y., Oromiehie, A., Zamani, Z., Ghanbarzadeh, B., Costa, C., et al.(2013). Evaluation of the photocatalytic antimicrobial effects of a TiO2nanocomposite food packaging lm by in vitro and in vivo tests. LWT FoodScience and Technology, 50(2), 702706.

Ciof, N., Torsi, L., Ditaranto, N., Tantillo, G., Ghibelli, L., Sabbatini, L., et al. (2005).Copper nanoparticle/polymer composites with antifungal and bacteriostaticproperties. Chemistry of Materials, 17(21), 52555262.

de Menezes, E., Nunes, M. R., Arenas, L. T., Dias, S. L. P., Garcia, I. S., Gushikem, Y.,et al. (2012). Gold nanoparticle/charged silsesquioxane lms immobilized ontoAl/SiO2 surface applied on the electrooxidation of nitrite. Journal of Solid StateElectrochemistry, 16(12), 37033713.

Emamifar, A., Kadivar, M., Shahedi, M., & Soleimanian-Zad, S. (2011). Effect ofnanocomposite packaging containing Ag and ZnO on inactivation ofLactobacillus plantarum in orange juice. Food Control, 22(34), 408413.

Fakhouri, F. M., Costa, D., Yamashita, F., Martelli, S. M., Jesus, R. C., Alganer, K., et al.(2013). Comparative study of processing methods for starch/gelatin lms.Carbohydrate Polymers, 95(2), 681689.

Incoronato, A. L., Buonocore, G. G., Conte, A., Lavorgna, M., & Nobile, M. A. (2010).Active systems based on silver-montmorillonite nanoparticles embedded intobio-based polymer matrices for packaging applications. Journal of FoodProtection, 73(12), 22562262.

Kanmani, P., & Rhim, J.-W. (2014). Physicochemical properties of gelatin/silver nanoparticle antimicrobial composite lms. Food Chemistry, 148,162169.

Kechichian, V., Ditcheld, C., Veiga-Santos, P., & Tadini, C. C. (2010). Naturalantimicrobial ingredients incorporated in biodegradable lms based on cassavastarch. LWT Food Science and Technology, 43(7), 10881094.

Liao, J., Anchun, M., Zhu, Z., & Quan, Y. (2010). Antibacterial titanium platedeposited by silver nanoparticles exhibits cell compatibility. InternationalJournal of Nanomedicine, 5, 337342.

Llorens, A., Lloret, E., Picouet, P. A., Trbojevich, R., & Fernandez, A. (2012). Metallic-based micro and nanocomposites in food contact materials and active foodpackaging. Trends in Food Science & Technology, 24(1), 1929.

Mei, J., Yuan, Y., Wu, Y., & Li, Y. (2013). Characterization of edible starchchitosanlm and its application in the storage of Mongolian cheese. International Journalof Biological Macromolecules, 57, 1721.

Nascimento, T. A., Calado, V., & Carvalho, C. W. P. (2012). Development andcharacterization of exible lm based on starch and passion fruit mesocarpour with nanoparticles. Food Research International, 49(1), 588595.

Nunes, M. R., Gushikem, Y., Landers, R., Dupont, J., Costa, T. M. H., & Benvenutti, E. V.(2012). Charged silsesquioxane used as a vehicle for gold nanoparticles toperform the synthesis of catalyst xerogels. Journal of Sol-Gel Science andTechnology, 63(2), 258265.

Pelissari, F. M., Andrade-Mahecha, M. M., Sobral, P. J. d. A., & Menegalli, F. C. (2013).Comparative study on the properties of our and starch lms of plantainbananas (Musa paradisiaca). Food Hydrocolloids, 30(2), 681690.

Rhim, J. W., Wang, L. F., & Hong, S. I. (2013). Preparation and characterization ofagar/silver nanoparticles composite lms with antimicrobial activity. FoodHydrocolloids, 33(2), 327335.

Rotta, J., Ozrio, R. ., Kehrwald, A. M., de Oliveira Barra, G. M., de Melo CastanhoAmboni, R. D., & Barreto, P. L. M. (2009). Parameters of color, transparency,water solubility, wettability and surface free energy of chitosan/hydroxypropylmethylcellulose (HPMC) lms plasticized with sorbitol.Materials Science and Engineering C, 29(2), 619623.

Schneid, A. C., Roesch, E. W., Sperb, F., Matte, U., da Silveira, N. P., Costa, T. M. H.,et al. (2014). Silver nanoparticleionic silsesquioxane: A new system proposedas an antibacterial agent. Journal of Materials Chemistry B, 2, 10791086.

Souza, A. C., Benze, R., Ferro, E. S., Ditcheld, C., Coelho, A. C. V., & Tadini, C. C.(2012). Cassava starch biodegradable lms: Inuence of glycerol and claynanoparticles content on tensile and barrier properties and glass transitiontemperature. LWT Food Science and Technology, 46(1), 110117.

Souza, A. C., Goto, G. E. O., Mainardi, J. A., Coelho, A. C. V., & Tadini, C. C. (2013).Cassava starch composite lms incorporated with cinnamon essential oil:Antimicrobial activity, microstructure, mechanical and barrier properties. LWT Food Science and Technology., 54(2), 346352.

Wang, L., Dong, Y., Men, H., Tong, J., & Zhou, J. (2013). Preparation andcharacterization of active lms based on chitosan incorporated teapolyphenols. Food Hydrocolloids, 32(1), 3541.

Willets, K. A., & Duyne, R. P. V. (2007). Localized surface plasmon resonancespectroscopy and sensing. Annual Review of Physical Chemistry, 58, 267297.

Xiong, H., Tang, S., Tang, H., & Zou, P. (2008). The structure and properties of astarch-based biodegradable lm. Carbohydrate Polymers, 71(2), 263268.

Yoksan, R., & Chirachanchai, S. (2010). Silver nanoparticle-loaded chitosanstarchbased lms: Fabrication and evaluation of tensile, barrier and antimicrobialproperties. Materials Science and Engineering C, 30(6), 891897.

762 C.H. Pagno et al. / Food Chemistry 173 (2015) 755762

Development of active biofilms of quinoa (Chenopodium quinoa W.) starch containing gold nanoparticles and evaluation of antimicrobial activity1 Introduction2 Materials and methods2.1 Quinoa starch production2.2 Proximate analysis of the quinoa starch2.3 Obtaining a gold nanoparticle dispersion2.4 Preparation of starch quinoa biofilms2.5 Characterisation of quinoa starch biofilms2.5.1 Mechanical properties2.5.2 Water vapour permeability (WVP)2.5.3 Solubility2.5.4 Optical properties (colour and opacity)2.5.5 Gas permeability2.5.6 Scanning electron microscopy (SEM)2.5.7 Transmission electron microscopy (TEM)2.5.8 Thermogravimetric analysis

2.6 Antimicrobial activity2.7 Statistical analysis

3 Results and discussion3.1 Composition of quinoa starch3.2 AuNPs aqueous dispersion3.3 Biofilm characterisation3.3.1 Mechanical properties3.3.2 Water vapour permeability (WVP)3.3.3 Solubility3.3.4 Optical properties (colour and opacity)3.3.5 Gas permeability3.3.6 Morphological properties3.3.7 Thermal stability

3.4 Antimicrobial activity

4 ConclusionsAppendix A Supplementary dataReferences