optimization of process conditions for the production of films based

7/27/2019 Optimization of Process Conditions for the Production of Films Based

1/11

Optimization of process conditions for the production offilms basedon the flour from plantain bananas (Musa paradisiaca)

Franciele Maria Pelissari a, Margarita Mara Andrade-Mahecha a, Paulo Jos do Amaral Sobral b,Florencia Cecilia Menegalli a,*a Department of Food Engineering, School of Food Engineering, University of Campinas, CEP 13083-862 Campinas, SP, Brazilb Department of Food Engineering, School of Animal Science and Food Engineering, University of So Paulo, CEP 13635-900 Pirassununga, SP, Brazil

a r t i c l e i n f o

Article history:

Received 29 August 2012Received in revised form27 December 2012Accepted 10 January 2013

Keywords:

Banana flourBiodegradable filmsExperimental designMechanical propertiesBarrier properties

a b s t r a c t

In this work, the casting process has been employed for the production offlour films from plantainbananas (Musa paradisiaca); glycerol has been used as plasticizer. The influence of process conditionssuch as the glycerol concentration (Cg), the process temperature (Tp), the drying temperature (Td), andthe relative humidity (RH) on the mechanical, barrier, and optical properties of banana flour films hasbeen evaluated by means of a central composite design. The results have been statistically analyzed bythe response surface methodology and desirability function, and the optimal process conditions for filmformation have been determined. The process variables have a significant impact on the mechanicalproperties, water vapor permeability (WVP), and opacity of the films, but these features are mostlyaffected by the Cg parameter. Compared to other biodegradable films, the banana flour film displays highopacity, low solubility in water, good WVP and flexibility, and excellent mechanical strength and rigidity.The desirability function employed here has allowed for the establishment of the optimal process con-ditions, which are Cg 19 g of glycerol/100 g offlour, Tp 81 C, Td 54 C, and RH 48%, proving to bean effective tool for this type of study.

Published by Elsevier Ltd.

1. Introduction

Polysaccharides and proteins of animal and vegetable origin arenatural biodegradable polymers that have traditionally been usedto produce environmentally-friendlyfilms(Chandra & Rustgi, 1998;Krochta & De Mulder-Johnston, 1997). Starch is the most widelyemployed polysaccharide for film production, because it is natu-rally abundant and inexpensive (Alves, Mali, Belia, & Grossmann,2007; Mali, Grossmann, Garca, Martino, & Zaritzky, 2004). Starchfilms present good mechanical and oxygen barrier properties, buttheir sensitivity to moisture is a major drawback. To improve the

characteristics of these materials, some authors have designedfilms based on starch and protein mixtures (Coughlan, Shaw, Kerry,& Kerry, 2004; Jagannath, Nanjappa, Das Gupta, & Bawa, 2003).Researchers have also added lipids to the formulation of some films,to enhance the water vapor barrier (Bravin, Peressini, & Sensidoni,2004; Garca, Martino, & Zaritzky, 2000).

When developing films, another alternative is to use flours,which are naturally occurring complex blends of starch, protein,

lipids, and fibers. Some authors have reported on the potentialapplication offlours obtained from whole materials such as amar-anth, soy, and wheat for film production (Mariniello et al., 2003;Rayas, Hernandez, & Ng, 1997; Tapia-Blcido, Sobral, & Menegalli,2005a). The excellent characteristics of these films stem from thenatural and intrinsic molecular interactions taking place betweentheir starch, protein, lipid, and fiber components.

An interesting renewable raw material for the preparation ofedible and biodegradable films is the unripe banana fruit. Origi-nating in Southeast Asia, bananas (genus Musa) are an importantfood crop that is extensively grown in tropical and subtropical re-

gions. Unfortunately, post-production losses are huge due to thehighly perishable nature and inadequate post-harvest handling ofbanana fruit. Processing both the surplus fruit and the fruit that isinappropriate for fresh consumption reduces these losses. Starch isthe major constituent of unripe bananas and comprises over 70 g/100 g of their dry weight. Moreover, this fruit contains a significantamount of protein (1.0e2.5 g/100 g), lipids (0.2e0.5 g/100 g)and fiber (1.5e2.5 g/100 g) (Zhang, Whistler, BeMiller, &Hamaker, 2005), which could be interesting for the production ofbiodegradable films. In this sense, the flour of unripe bananas maybe an attractive alternative for the attainment of a continuousmatrix.

* Corresponding author. Tel.: 55 19 3521 4039; fax: 55 19 3521 4027.E-mail addresses: [email protected], [email protected] (F.C. Menegalli).

Contents lists available at SciVerse ScienceDirect

LWT - Food Science and Technology

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / l w t

0023-6438/$ e see front matter Published by Elsevier Ltd.

http://dx.doi.org/10.1016/j.lwt.2013.01.011

LWT - Food Science and Technology 52 (2013) 1e11
mailto:[email protected]:[email protected]:[email protected]://www.sciencedirect.com/science/journal/00236438http://www.elsevier.com/locate/lwthttp://dx.doi.org/10.1016/j.lwt.2013.01.011http://dx.doi.org/10.1016/j.lwt.2013.01.011http://dx.doi.org/10.1016/j.lwt.2013.01.011http://dx.doi.org/10.1016/j.lwt.2013.01.011http://dx.doi.org/10.1016/j.lwt.2013.01.011http://dx.doi.org/10.1016/j.lwt.2013.01.011http://www.elsevier.com/locate/lwthttp://www.sciencedirect.com/science/journal/00236438mailto:[email protected]:[email protected]


2/11

Romero-Bastida et al. (2005) recently used starches isolatedfrom banana, okenia, and mango to produce edible films. Okeniaand bananafilms were the most (32%) and the least (23%) soluble inwater, respectively. The bananafilm hadthe highest tensile strengthvalue (25MPa)comparedwith mango (18 MPa) andokenia (17 MPa)films. A group of researchers studied the effect of banana starchmodified by oxidation and acetylation on the properties of films(Zamudio-Flores, Bautista-Baos, Salgado-Delgado, & Bello-Prez,2009; Zamudio-Flores, Vargas-Torres, Prez-Gonzlez, Bosquez-Molina, & Bello-Prez, 2006). The oxidation level increased butacetylation decreased the solubility and WVP of thefilm. Oxidationenhanced the tensile strength of the film, and acetylation of theoxidized starch improved this property. Elongation at breakdiminished when the oxidation level rose.

More recently, Pitak and Rakshit (2011) employed banana flour/chitosan compositefilm bags to preserve freshly cut vegetables. Thecomposite yellowish film exhibited great water permeability of4.5e4.8 1010 g/m s Pa. Tensile strength and elongation were inthe range of 5.2e14.2 MPa and 1.6e2.6%, respectively, while solu-bility ranged between 40.9 and 64.2%. The presence of starch in thecompositefilm furnished water soluble and sealable bags or wraps,while the presence of chitosan provided them with the anti-

microbial property. In another work, Sothornvit and Pitak (2007)prepared films from the banana flour developed; however, theyevaluated the oxygen permeability and the mechanical propertiesof these films, but they did not determine solubility or the watervapor barrier and the optical properties.

Among the different varieties of banana, the cultivar Terra(Musa paradisiaca), a type of plantain, exhibits excellent features forthe preparation of biodegradable films (Pelissari, Andrade-Mahecha, Sobral, & Menegalli, 2012). Because a few studies repor-ted in the literature have dealt with this cultivar, we have selected itfor the present investigation.

Considering that the several parameters employed during filmproduction influence the film properties, a study was carried out toverify the effect of glycerol concentration (Cg), process temperature

(Tp), drying temperature (Td), and relative humidity (RH) on filmfeatures, so that banana flour films with the desired characteristicscan be obtained. In this context, the present study aimed todetermine the optimal formulation of banana flour film by usingresponse surface methodology and multi-response analysis, in or-der to obtain films with low water vapor permeability, moderateelongation, and high resistance to break.

2. Materials and methods

2.1. Materials

The flour was prepared from unripe plantain bananas of thevariety Terra (M. paradisiaca), according to the methodology

described by Pelissari et al. (2012). The fruit was obtained from theharvest occurred in March 2010 in the state of Esprito Santo, Brazil,but it was not subjected to any postharvest treatment. All thechemicals used in this work were reagent grade.

2.2. Physicochemical analysis of the banana flour

The size of flour particles was determined in triplicate witha laser diffraction analyzer (Laser Scattering Spectrometer Mas-tersizer S, model MAM 5005 e Malvern Instruments Ltd., Surrey,England) using ethanol as solvent. The moisture, ash, protein, andcrude fiber contents were analyzed by the AOAC standard methods(2005). Lipids were assayed by the method of Bligh and Dyer(1959), as described by Cecchi (1999). The amylose content was

obtained according to the methodology reported in ISO 6647

(1987), and total starch was identified by using the method pro-posed by Diemair (1963). All the experiments were performed atleast in triplicate, and the results are presented as mean values.

2.3. Film production

Films were produced by the casting method. This process con-sists in drying a film-forming suspension (FFS) that has beenapplied onto a support. The procedure developed herein involvedthe homogenization of a water solution of 4 g/100 g (d.b.) of bananaflour by mechanical stirring for 30 min, followed by heating to theprocess temperature (Tp: 75e95 C) under gentle stirring. Glycerol(Cg: 15e30 g of glycerol/100 g offlour) was added at this point, andthe solution was maintained at this temperature for 15 min. Next,the FFS was sonicated for 10 min, and 70 g of the solution waspoured onto acrylic plates (18 21 cm), so as to obtain a constantthickness. The films were dried in a chamber with air circulationunder controlled temperature (Td: 35 to 55 0.5 C) and relativehumidity (RH: 30 to 70 0.5%).

Before the characterization of the films in terms of moisturecontent, mechanical properties, and water vapor permeability, theywere conditioned in desiccators under 58% RH, at 25 C, for 48 h.

2.4. Film characterization

2.4.1. Thickness and densityThe thickness of the films was measured using a manual

micrometer (Fowler, model FOW52-229-001, Pennsylvania, USA)with an accuracy of 0.0001 mm. The mean thickness of each filmwas determined from an average of 10 random measurements.

To determine the density, samples of each film were cut into20 20 mm squares,and the thickness of thesefilms was measured(3 random measurements). The film samples were dried at 105 Cfor 24 h and weighed, and the density was calculated as the ratiobetween the weight and volume (thickness area) of the film. Thedensity experiments were accomplished in triplicate, and the data

are reported as mean values.

2.4.2. Moisture contentThe moisture content of the films was analyzed gravimetrically,

in triplicate, according to the standard method D644-99 (ASTM,1999), by drying the samples at 105 C for 24 h.

2.4.3. Mechanical propertiesThe tensile properties were investigated with the aid of a tex-

ture analyzer (Stable Micro Systems, model TA.TXplus, Surrey, En-gland) according to the standard method D882-02 (ASTM, 2002),by taking an average of six determinations in each case. The sam-ples were cut into 25-mm wide and 115-mm long strips by meansof a scalpel and mounted between the tensile grips. The initial grip

separation and the crosshead speed wereset at 80 mm and 1.0 mm/s, respectively. The tensile strength (force/initial cross-sectionalarea) and the elongation at break were computed directly fromthe strength elongation curves using the Texture Exponent 32software, and Youngs modulus was calculated as the slope of theinitial linear portion of this curve.

2.4.4. Solubility in waterThe solubility (S) values were determined by employing the

methodology described by Gontard, Guilbert, and Cuq (1992). Tothis end, three discs (diameter 20mm)of eachfilm were stored ina desiccator containing silica gel (w0% RH) for 48 h. The sampleswere weighed, to obtain the initial dry weight (Wi), and they werethen immersed into 50 mL water containing sodium azide (0.2 g/L)

at 25

C for 24 h, under sporadic agitation. After this period, the

F.M. Pelissari et al. / LWT - Food Science and Technology 52 (2013) 1e112


3/11

solution containing the film discs was filtered, the insoluble matterwas dried at 105 C for 24 h, and the resulting material wasweighed for determination of the final dry weight (Wf). Analyseswere carried out in quadruplicate, and the solubility in water (%) ofthe films was calculated according to Eq. (1):

SWi Wf

Wi 100 (1)

where Wi is the initial dry weight of the sample (g), and Wf is thefinal dry weight of the sample (g).

2.4.5. Water vapor permeability (WVP)The WVP was determined gravimetrically by following the

standard method E96-00 (ASTM, 2000) with modifications. For thispurpose, the film (diameter 0.06 m) was placed on the circularopening (area 0.00196 m2) of a permeation cell and was sealedwith sealant ring, to ensure that humidity migration would occurthrough thefilm only. The interior of thecell wasfilledwith silicagel(w0% RH), and the system was stored in a desiccator containingdistilledwater (100% RH)at 25 C. Aftersteady state conditions werereached (about2 h),the weight gain of thecells wasmonitoredevery

30 min, for 8 h. Analyses were conducted in triplicate, and the WVPwas calculated on the basis of Eq. (2) and expressed in g/m s Pa:

WVP w

t

d

A$DP(2)

where w/tis the slope of the line of weight gain (w) as a function oftime (t) graph (g/s), d is the mean sample thickness (m), A is thesample permeation area (m2), and DP is the difference in watervapor pressure through the sample for pure water at 25 C (Pa).

2.4.6. Optical propertiesBanana flour films were subjected to color and opacity analyses

using a colorimeter (UltraScan VIS, HunterLab, Virginia, EUA) in the

transmittance mode, with classification system of the CIELab andilluminant D65 (daylight) (HunterLab, 1996).

2.5. Experimental design

To determine the influence of the process conditions (Cg, Tp, Td,and RH) on the properties of the films, a central composite design24 was performed, with four replicates at the central point andeight axial points, totaling 28 experiments (Rodrigues & Iemma,2005). The dependent variables (responses) were the tensilestrength, the elongation at break, Youngs modulus, the solubility inwater, the WVP, the color difference (DE*), and the opacity. Theexperimental design and the coded and real values of the inde-pendent variables are given in Table 1. The study ranges had beendefined in preliminary tests.

A second order model (Eq. (3)) was adopted, in order to fit theresponse variables:

Y b0 X

biXi X

b2i X

2i

XbijXiXj (3)

where Y is the dependent variable, Xi and Xj are the coded inde-pendent variables,b0 is the constant,bi is the linear coefficient, b

2i is

the quadratic coefficient, and bij is the interaction coefficient.Statistical analysis of the experimental data and the response

surface methodology were performed using the Statistica 7.0 soft-ware (StatSoft Inc, Oklahoma, USA). The optimal values of theprocess conditions were obtained by the desirability function,a multi-response analysis, proposed by Derringer and Suich (1980).

3. Results and discussion

3.1. Physicochemical analysis of the banana flour

The flour prepared from Terra bananas (M. paradisiaca) had anaverage particle size of 31.7 mm, and the following chemical

Table 1Central composite design matrix and some characteristics of banana flour films.

Test Independent variablesa Drying time (h) Density (g/cm3) Moisture content (g/100 g)

X1 (Cg) X2 (Tp) X3 (Td) X4 (RH)

1 1 (18.75) 1 (80) 1 (40) 1 (40) 4.2 1.12 0.04 15.0 0.12 1 (26.25) 1 (80) 1 (40) 1 (40) 4.2 1.09 0.02 20.4 0.73 1 (18.75) 1 (90) 1 (40) 1 (40) 4.2 1.11 0.04 14.3 0.14 1 (26.25) 1 (90) 1 (40) 1 (40) 4.5 1.05 0.04 19.9 0.95 1 (18.75) 1 (80) 1 (50) 1 (40) 3.8 1.25 0.02 15.8 0.56 1 (26.25) 1 (80) 1 (50) 1 (40) 3.8 1.16 0.01 19.0 0.17 1 (18.75) 1 (90) 1 (50) 1 (40) 3.3 1.12 0.04 14.1 0.28 1 (26.25) 1 (90) 1 (50) 1 (40) 3.8 1.07 0.04 18.5 0.59 1 (18.75) 1 (80) 1 (40) 1 (60) 6.0 1.10 0.02 14.3 0.410 1 (26.25) 1 (80) 1 (40) 1 (60) 6.5 1.05 0.03 19.5 0.911 1 (18.75) 1 (90) 1 (40) 1 (60) 6.2 1.09 0.03 14.4 0.212 1 (26.25) 1 (90) 1 (40) 1 (60) 6.5 0.96 0.04 20.3 0.513 1 (18.75) 1 (80) 1 (50) 1 (60) 4.8 1.19 0.01 15.7 0.414 1 (26.25) 1 (80) 1 (50) 1 (60) 5.0 1.14 0.03 19.1 0.515 1 (18.75) 1 (90) 1 (50) 1 (60) 4.8 1.00 0.02 14.3 0.316 1 (26.25) 1 (90) 1 (50) 1 (60) 5.2 1.06 0.03 19.6 0.717 2 (15.00) 0 (85) 0 (45) 0 (50) 4.3 1.09 0.04 13.1 0.118 2 (30.00) 0 (85) 0 (45) 0 (50) 5.0 1.04 0.03 22.3 0.919 0 (22.50) 2 (75) 0 (45) 0 (50) 5.0 1.22 0.03 17.8 0.220 0 (22.50) 2 (95) 0 (45) 0 (50) 5.0 0.94 0.01 16.8 0.321 0 (22.50) 0 (85) 2 (35) 0 (50) 6.3 1.11 0.03 17.2 0.722 0 (22.50) 0 (85) 2 (55) 0 (50) 3.8 1.10 0.01 15.7 0.323 0 (22.50) 0 (85) 0 (45) 2 (30) 3.5 1.17 0.02 17.5 0.324 0 (22.50) 0 (85) 0 (45) 2 (70) 8.0 1.06 0.03 17.7 0.425 0 (22.50) 0 (85) 0 (45) 0 (50) 5.0 1.06 0.04 16.5 0.326 0 (22.50) 0 (85) 0 (45) 0 (50) 5.0 1.07 0.07 16.2 0.227 0 (22.50) 0 (85) 0 (45) 0 (50) 5.0 1.07 0.04 16.3 0.328 0 (22.50) 0 (85) 0 (45) 0 (50) 5.0 1.06 0.02 16.3 0.2

a

Cg

glycerol concentration (g/100 g offl

our), Tp

process temperature (

C), Td

drying temperature (

C), RH

relative humidity (%).

F.M. Pelissari et al. / LWT - Food Science and Technology 52 (2013) 1 e11 3


4/11

composition, on dry basis: 8.7 0.0 g of moisture/100 g; 1.9 0.1 gof ash/100 g; 3.2 0.1 g of protein/100 g; 0.6 0.0 g of lipids/100 g;1.2 0.1 g of crude fiber/100 g; and 83.2 0.2 g of starch/100 g(among which 27.8 0.1 g/100 g of dry starch was amylose). Thecomposition of this flour is similar to those obtained by DaMota, Lajolo, Cordenunsi, and Ciacco (2000), Kayisu, Hood, andVansoest (1981), and Suntharalingam and Ravindran (1993). Theprotein and lipids present in the banana flour exert an importantand well-known plasticizer effect, which reduces the mechanicalresistance and increases the flexibility of the biopolymer films.Also, the presence of hydrophilic components such as protein andfiber in the flour may give rise to a greater number of interactionswith water molecules. The amylose content is an important featurefor the film-forming capacity of the starch. Compared with otherraw materials used in the production of biodegradable films, thebanana flour presents higher amylose content than the amaranthflour (7.6 g/100 g of dry starch) (Tapia-Blcido et al., 2005a) and thecassava flour (17 g/100 g of dry starch) (Suppakul, Chalernsook,Ratisuthawat, Prapasitthi, & Munchukangwan, 2013), but lowerthan the achira flour (30.7 g/100 g of dry starch) (Andrade-Mahecha, Tapia-Blcido, & Menegalli, 2012).

3.2. Film characterization

All the prepared films displayed homogeneous surface with nobubbles or cracks, as well as good handling characteristics. Thismeans that the films could be easily detached from the plateswithout tearing, and that they were not sticky or too brittle. Thebanana flour films were transparent, with light yellow colorationand an average thickness of 87 3 mm.

Table 1 lists the drying time, density, and moisture content ofthe films obtained in the different process conditions tested in theexperimental design. The density of the films ranged from 0.94 to1.25 g/cm3. The drying conditions (drying temperature and relativehumidity) influenced the drying time. The combination of high Tdand low RH diminished the drying time of the films by about 50%

on average, as can be noticed when one goes from a less drastic(lower Td and higher RH) to a more drastic (higher Td and lowerRH) condition, considering the same Cg and Tp. On the other hand,the Cg variable affected the moisture content of the films after theconditioning period (58% RH, 48 h) the most. Indeed, films withlarger plasticizer content exhibited higher moisture values ascompared with those with lower Cg, a result of the high hygro-scopic character of glycerol.

3.3. Statistical analysis

Table 2 brings the film properties obtained according to theexperimental design. These data were submitted to statistical an-alyses, including fitting to Eq. (3), followed by an analysis of var-iance (ANOVA) at 90% confidence level. Only the statisticallysignificant parameters have been used for analysis of the behaviorof the fitted mathematical models:

Y1 6:85 2:71X1 0:40X4 0:94X21 0:51X1X2 0:46X1X3

0:67X3X4(4)

Y2 26:07 6:98X1 1:68X2 1:99X3 2:57X21 2:07X1X4

1:74X2X3 2:60X2X4(5)

Y3 334:95 324:52X1 124:27X21 57:44X1X2

45:34X2X3 (6)

Y4 26:69 1:33X21 1:64X

22 2:01X2X3 (7)

Y5 1:49 0:18X1 0:26X4 0:24X22 0:21X1X2 (8)

Table 2

Results of the central composite design.

Test Tensile strength (MPa) Elongation (%) Youngs modulus (MPa) Solubility (%) WVP (1010 g/m s Pa) DE* Opacity (%)

Y1 Y2 Y3 Y4 Y5 Y6 Y7

1 12.4 0.1 12.8 3.4 947.9 59.9 27.6 1.8 1.65 0.10 2.90 0.09 46.2 0.32 7.2 0.2 22.2 1.9 256.9 10.7 32.9 2.5 2.89 0.10 2.79 0.06 51.0 0.43 11.6 0.9 12.9 1.1 871.2 79.0 34.1 1.3 2.26 0.08 2.69 0.07 44.0 0.34 5.4 0.2 30.6 1.3 160.2 13.1 27.4 2.6 1.88 0.05 2.75 0.06 51.7 0.25 10.6 0.8 15.9 1.7 767.8 64.7 34.9 1.6 1.68 0.02 2.65 0.05 48.6 0.26 5.6 0.3 39.3 3.7 132.4 10.7 26.8 2.6 2.11 0.05 2.32 0.04 48.2 0.47 10.5 0.7 11.9 2.1 938.2 60.0 27.8 1.7 1.47 0.03 2.31 0.08 35.8 0.48 4.4 0.3 36.4 2.1 114.8 12.0 25.0 0.4 1.77 0.06 2.39 0.04 43.8 0.49 9.0 0.6 19.3 1.0 644.7 70.3 26.0 1.4 2.71 0.38 2.60 0.10 48.9 0.410 4.2 0.2 31.9 2.5 95.3 1.1 32.4 2.4 3.50 0.45 2.69 0.15 50.0 0.511 10.1 0.6 13.5 1.1 798.2 49.2 33.6 1.3 2.26 0.14 2.74 0.13 51.2 0.212 2.4 0.4 23.9 1.0 55.6 7.4 33.1 2.4 2.19 0.02 2.92 0.11 59.9 0.413 8.3 0.2 27.4 1.5 423.0 20.4 37.5 1.5 2.22 0.20 2.37 0.08 48.0 0.914 7.1 0.6 32.9 1.5 226.2 33.1 34.1 2.8 3.15 0.18 2.78 0.04 52.1 0.215 10.0 0.7 10.1 1.2 938.4 107.1 24.9 1.9 2.22 0.09 2.46 0.04 48.6 0.116 5.7 0.2 24.0 1.0 223.8 25.7 32.9 2.6 2.44 0.15 2.78 0.05 56.2 0.317 16.9 0.2 4.6 1.1 1482.2 54.3 25.1 2.3 1.48 0.10 2.43 0.07 47.5 0.218 4.5 0.2 29.7 1.7 120.1 15.4 34.7 1.7 1.89 0.05 2.75 0.07 52.9 0.119 7.3 0.7 27.8 2.7 309.1 67.1 28.9 0.7 1.98 0.04 3.19 0.27 70.0 0.420 5.6 0.5 26.8 2.0 253.9 16.7 33.5 0.9 2.22 0.03 2.62 0.08 50.9 0.421 8.1 0.9 26.5 2.1 439.3 71.5 24.8 0.3 1.67 0.03 2.74 0.04 48.1 0.222 5.4 0.4 35.0 2.0 176.5 41.8 28.9 2.2 1.53 0.03 2.71 0.09 52.3 0.423 6.4 0.2 27.6 2.3 247.2 22.0 27.1 1.4 1.36 0.07 2.75 0.15 39.3 0.224 6.9 0.3 22.3 3.0 293.4 15.9 26.0 0.2 2.02 0.13 2.68 0.07 52.9 0.325 7.1 0.7 24.1 1.6 367.1 22.3 24.8 0.8 1.51 0.06 2.57 0.02 50.4 0.226 7.3 0.6 25.4 1.6 361.1 12.0 24.4 0.9 1.46 0.01 2.53 0.06 50.2 0.327 7.3 0.4 28.0 2.9 359.5 27.5 24.6 1.0 1.52 0.05 2.54 0.04 50.8 0.328 7.4 0.4 25.6 3.3 356.9 30.4 24.7 2.6 1.48 0.05 2.58 0.06 50.4 0.4



5/11

Y6 2:60 0:06X1 0:09X3 0:06X22 0:08X1X4 (9)

Y7 49:20 2:18X1 1:66X2 3:04X4 2:26X22 1:33X

24

2:22X2X4(10)

where Y is the dependent variables (Y1 tensile strength,Y2 elongation at break, Y3 Youngs modulus, Y4 solubility,Y5 WVP, Y6 color difference, and Y7 opacity), and X are thecoded independent variables (X1 Cg, X2 Td, X3 Td, and

X4 RH).Table 3 summarizes the results of the ANOVA, including the

regression coefficients for the coded second order polynomialequation, the coefficients of determination (R2), and the F and pvalues. Finally, to determine whether the fitted equations are pre-dictive, they must satisfy a certain criterion based on the Ftest values(Fcalculated and Flisted), which in turn depends on the calculated Fratio value for the regressions related to the residuals (Fcalculated/Flisted). The criterion is that the value of this ratio must be higherthan that ofFlisted (Khuri & Cornell,1996), thereby enabling plottingof the response surfaces.

Except for solubility and color difference, the fitted equationsare predictive forall the studied properties, since their Fratio values(Fcalculated/Flisted) are higher than the corresponding Flisted values(Table 3). These results suggest that the models fitted for the me-chanical properties (tensile strength, elongation at break, andYoungs modulus), WVP, and opacity are suitable (significant andpredictive) and lead to significant regression, low residual values,no lack offit, and satisfactory coefficients of determination. As forsolubility and color difference, the coefficients of determination(R2) are 0.36 and 0.46, respectively, indicating that the modelsexplain only 36 and 46% of the observed data variation. The lowerFcalculated/Flisted value in relation to Flisted proves that the models arenot valid for these variables, so they cannot be considered predic-

tive in these cases (Table 3). Because these properties do not

correlate with the process conditions, it is not possible to generatethe response surfaces.

3.4. Response surface

3.4.1. Mechanical propertiesThe desired property of a food packaging material will depend

on the application. In general, food packaging must not bedeformed and must provide structural integrity to the food orreinforce food structure. In other situations, a deformable film isdesirable (Gontard et al., 1992).

Fig.1 depicts the response surfaces obtained for the mechanicalproperties of banana flour films. Cg affected the tensile strengthand Youngs modulus negatively and the elongation at break pos-itively (Eqs. (4)e(6); Fig. 1a, c, and e). This behavior of the glycerolcontent agrees with other studies and stems from its plasticizingeffect (Araujo-Farro, Podadera, Sobral, & Menegalli, 2010; Bergo,Sobral, Guevara, & Vadala, 2010; Sobral, Menegalli, Hubinger, &Roques, 2001; Tapia-Blcido, Sobral, & Menegalli, 2011). Accord-ing to Mali et al. (2004) and Mali, Sakanaka, Yamashita, andGrossmann (2005), films containing glycerol as plasticizer experi-ence weaker intermolecular forces or attractive forces between theadjacent polymer chains, which culminates in enhanced molecularmobility throughout the film structure. Consequently, the elonga-tion at break increases, reducing the tensile strength and Young smodulus (Fig. 1a, c, and e). In addition, as demonstrated byMyllrinen, Partanen, Seppl, and Forssell (2002) in their work onthe influence of glycerol on edible amylose and amylopectin films,the employed amylose/amylopectin ratio, plasticizer concentration,and/or the moisture content and storage conditions can affect themechanical properties of the film. Alves et al. (2007) reported thatcassava starch films with 20 g of glycerol/100 g of starch presenteda tensile strength of 21.7 MPa, elongation at break of 5.2%, andYoungs modulus of 40.5 MPa, while those with higher glycerolconcentration (45 g of glycerol/100 g of starch) had values of5.4 MPa, 153.2%, and 0.2 MPa, respectively. In their work on the

development of potato starch films, Talja, Heln, Roos, and Jouppila

Table 3

Results for regression coefficients and analysis of variance (ANOVA) for the dependent variables of the central composite design.

Parametera Coefficient Tensile strength (MPa) Elongation (%) Youngs modulus (MPa) Solubility (%) WVP (1010 g/m s Pa) DE* Opacity (%)

Y1 Y2 Y3 Y4 Y5 Y6 Y7

Mean bo 6.85 26.07 334.95 26.69 1.49 2.60 49.20Linear

X1 b1 2.71 6.98 324.52 e 0.18 0.06 2.18X2 b2 e 1.68 e e e e 1.66X3 b3 e 1.99 e e e 0.09 eX4 b4 0.40 e e e 0.26 e 3.04Quadratic

X1 b21 0.94 2.57 124.27 1.33 e e e

X2 b2

2

e e e 1.64 0.24 0.06 2.26X3 b23 e e e e e e eX4 b

24 e e e e e e 1.33

Interaction

X1X2 b12 0.51 e 57.44 e 0.21 e eX1X3 b13 0.46 e e e e e eX1X4 b14 e 2.07 e e e 0.08 eX2X3 b23 e 1.74 45.34 2.01 e e eX2X4 b24 e 2.60 e e e e 2.22X3X4 b34 0.67 e e e e e eFcalculated 40.17 21.44 70.94 4.45 13.40 5.09 8.09Flisted

b 2.08 2.04 2.21 2.33 2.21 2.33 2.08Fcalculated:Flisted 19.31 10.51 32.10 1.91 6.06 2.18 3.89p-Value


6/11

Fig. 1. Response surfaces of tensile strength as a function of (a) Cg and Tp (Td 45 C, RH 50%), (b) Td and RH (Cg 22.50 g/100 g offlour, Tp 85 C). Response surfaces of

elongation at break as a function of (c) Cg and RH (Tp 85 C, Td 45 C), (d) Tp and Td (Cg 22.50 g/100 g offlour, RH 50%). Response surfaces of Youngs modulus as a function

of (e) Cg and Tp (Td 45 C, RH 50%), (f) Tp and Td (Cg 22.50 g/100 g offlour, RH 50%).



7/11

(2008) also observed a rise in the tensile strength and Youngsmodulus values and a reduction in the elongation at break valueswhen they used a low plasticizer content (20 g of plasticizer/100 gof starch), due to the improved crystallinity of the films.

The RH (X4) process variable did not exert a remarkable effecton the elongation at break or Youngs modulus properties. How-ever, this parameter had an interesting negative effect on tensilestrength (Eq. (4)); that was, larger RH values resulted in lowertensile strength values (Fig. 1b). This behavior is probably due tochanges in the sample macrostructure as a function of the air RHduring drying. Transmission electron microscope (TEM) micro-graphs have attested to this behavior and suggested that it isrelated to the use of higher RH during film formation, which altersthe pore size distribution and the non-homogeneity of the amylosenetwork (Stading, Rindlav-Westling, & Gatenholm, 2001). There-fore, amylose generates homogenous films at lower RH values, buthigher RH affords denser polymer zones with larger pores, prob-ably causing the lower tensile strength and the higher elongationat break values obtained for the banana flour films (Fig. 1b and c).Tapia-Blcido, Sobral, and Menegalli (2005b) reported that severalphenomena may occur during the drying of the film-forming so-lution, because of its complex composition (biopolymers and

plasticizers). This in turn should influence the structure andproperties of the film by prompting changes in the microstructureor large-scale alterations in the macrostructure, thereby affectingthe entire network.

ANOVA (Table 3) confirmed that Tp (X2) and Td (X3) have sig-nificant negative and positive effects on the elongation at break,respectively (Eq. (5)). According to Fig.1d, high elongation at breakvalues correspond to the minimum Tp and maximum Td values.Thus, longer drying times favor the molecular interactions betweenthe compounds present in the film solution, contributing to thelower elongation of the banana flour films found in this research.

The effect of Tp on the mechanical properties of the bananaflourfilms is associated with the gelatinization process. During gelatini-zation, amylose leaches to the aqueous phase between the granules,

and the polymer chains approach each other more closely becauseof the high amylose content. This forms a denser polymer matrixwith greater mechanical strength and lower flexibility (Fig. 1d).High Tp may also favor denaturation of the proteins present in theflour and increase the number of interactions between the differentbiopolymers (Denavi et al., 2009). Some authors reported that the

high temperature of the prepared films should be understood interms of a closely packed state where extensive intermolecularbonding occurs, which inhibits further orientation and betteralignment of the protein and starch chains (Arvanitoyannis,Nakayama, & Aiba, 1998).

According to Eq. (6), the interaction between Tp and Td (X2X3)has a significant, positive and relevant effect on the Young s mod-ulus property. In the search for process conditions that allow for theachievement of rigid films, two combinations for these processvariables are possible: higher Tp and Td or lower Tp and Td. Fig. 1fillustrates the two regions, in black, where the highest Young smodulus values can be obtained. This behavior offers more possi-bilities to adjust process conditions that favor the production ofbanana flour films with the desired characteristics within thestudied limits.

3.4.2. Water vapor permeability (WVP)Since the main function of a food packaging material is to con-

trol the moisture transfer between the food and the surroundingatmosphere, or between two components of a heterogeneous foodproduct, the WVP should be as low as possible (Gontard et al.,1992). In general, the use of low Cg (15e22.5 g/100 g offlour), in-

termediate Tp (80e85 C), high Td (45e55 C), and low RH (30e40%) yields banana flour films with lower WVP values, as shownin Fig. 2.

Many authors have extensively reported the effect of Cg on theWVP offilms (Cuq, Gontard, Cuq, & Guilbert, 1997; Garca, Martino,& Zaritzky, 1999; Mali, Grossmann, Garca, Martino, & Zaritzky,2006). The increase in WVP values with rising plasticizer content(Fig. 2a) is due to structural modifications occurring in the starchnetwork in combination with the hygroscopic character of glycerol.Moreover, glycerol molecules aresmall and present high capacity tointeract with starch chains, thereby enhancing the molecularmobility and augmenting the free volume in the film matrix(Sothornvit & Krochta, 2001). Tapia-Blcido et al. (2011) carried outthe optimization of amaranth flour films plasticized with glycerol

and sorbitol by multi-response analysis. These authors reportedthat the film prepared with the optimal formulation and usingglycerol as plasticizer was more hygroscopic, less resistant to break,flexible, and more permeable to oxygen as compared with the filmcontaining sorbitol as plasticizer. These observations confirmed thestronger plasticizing effect of glycerol.

Fig. 2. Response surfaces of WVP as a function of (a) Cg and Tp (Td

45

C, RH

50%), (b) Td and RH (Cg

22.50 g/100 g offl

our, Tp

85

C).



8/11

Other components in the structure of the banana flour maybehave in different ways depending on the film processing condi-tions. Heating promotes the denaturation of proteins, and a moredrastic thermal treatment modifies the three-dimensional struc-ture of globular proteins. This exposes the sulfhydryl (SH) groups,which may culminate in hydrophilic interactions during dryingbecause of the hydrophilic nature of this group (Tapia-Blcido et al.,2005a). Thus, increasing the Tp from 75 to 85 C/15 min leads toreduced WVP values in these films. However, more intense tem-peratures (above 85 C) furnish more permeable films (Fig. 2a).

Fig. 2b evidences that drastic drying conditions (high temper-ature and low relative humidity) result in lower WVP values andthus less permeable films. As observed by ANOVA (Table 3), the RH(X4) process variable influences the WVP, exerting a positive effecton this property (Eq. (8)). The presence of higher RHs could con-tribute to greater moisture contents in thefilms, thereby improvingtheir hydrophilicity.

3.4.3. OpacityPackages are often the only way of exposing products to con-

sumers before purchase (Marsh & Bugusu, 2007). Thus, the mate-rials chosen to produce packages play a vital role in product

commercialization. The optical properties of polymer films arerelated to opacity (transparency), color, and brightness. Opacity hasan important part, since certain very attractive products, especiallyfoodstuffs, must be well visible across the film by consumers(Gontard et al., 1992).

The parameters Tp (X2) and RH (X4) as well as the interactionbetween them (X2X4) strongly influence opacity (Eq. (10)). Based onANOVA results (Table 3), Tp has a negative effect while RH exertsa positive effect on this property. According to Fig. 3, lower Cg (15e18.75 g/100 g offlour), higher Tp (85e95 C), and lower RH (30%)values give less opaque films.

A higher Tp condition is expected to suffice for the crystalline-to-amorphous transition of starch to happen. At the same time,a milder thermal treatment leads to incomplete material melting

and favors the presence of crystalline residues and the consequentformation of more opaque banana flour films (Fig. 3).

Concerning the RH variable, wetter conditions afford higheropacity values (Fig. 3b). More facile swelling of the starch granulesaccounts for this behavior, and higher gelatinization of thesegranules takes place in the presence of high RH, thereby culmi-nating in more compact structures and increased opacity values.

The experimental opacities of banana flour films range between35.8 and 70.0%, which are higher as compared with those of achiraflour films (17.5e44.1%) (Andrade-Mahecha et al., 2012), amaranthflour films (6.5%) (Tapia-Blcido et al., 2005a), and quinoa flourfilms (5.3%) (Araujo-Farro, 2008). These differences might be rela-ted to the different botanical sources used in the production of thefilms but will also depend on the method used to measure opacity.

3.5. Optimization and validation

The optimization of the process conditions for the production ofbanana flour films was accomplished using a multi-responsemethod called desirability (Derringer & Suich, 1980). This methodinvolves the transformation of each response variable (Yi) to anindividual function of desirability (di) (Eqs. (11) and (12)); thedesirability scale ranges from 0 to 1. If the response fell outside anacceptable region, it was set to di 0; if the response was fullydesirable (at its goal or target), it was set to di 1.

di Yi Ymin

Ymax Ymin(11)

where Ymin is the response minimum value, and Ymax is theresponse maximum value. In the case of WVP and opacity, Eq. (11)had to be redesigned, so that the minimum values for these re-sponses could be obtained (Eq. (12)).

di Ymax Yi

Ymax Ymin(12)

The individual desirability functions from the considered re-sponses were then combined, to obtain the overall desirability (D),defined as the geometric average of the individual desires (Eq. (13)).D was later maximized using the Statistica 7.0 software.

D

dn11 ;dn22 ;.; d

nkk

1.k(13)

where k is the number of considered responses and ni is the weightof each response. Thus, since 0D 1, a high D value shows that alldks are toward the target value, which is considered as the optimalsolution of the system.

Fig. 4 shows the multi-response optimization of the predictedprofiles for the response variables thatpresented valid mathematical

Fig. 3. Response surfaces of opacity as a function of (a) Cg and Tp (Td

45

C, RH

50%), (b) Tp and RH (Cg

22.50 g/100 g offl

our, Td

45

C).



9/11

models (tensile strength, elongation at break, Youngs modulus,WVP, and opacity) together with their respective desirability func-tionprofiles, foreach of theinvestigatedprocessvariables(Cg,Tp, Td,and RH). Therefore, the optimal process conditions leading to themaximum global desirability value (0.96) for the production of ba-nana flour films are Cg 19 g/100 g offlour, Tp 81 C, Td 54 C,and RH 48%, with predicted response variables of 8.7 MPa for thetensile strength, 25.0% for the elongation at break, 590.5 MPa forYoungs modulus, 2.01010 g/m s Pa forthe WVP, and50.0%for theopacity. Other authors also found the optimal process conditions fortheir films using the multi-response analysis (Andrade-Mahechaet al., 2012; Araujo-Farro et al., 2010; Tapia-Blcido et al., 2011).

Experimental validation has beenperformed in triplicate, and theresults are given as mean values in Table 4. Based on the relative de-viation values obtained for each response variable, the optimizationmethodology employed here is satisfactory. The experimental valueof the tensile strength variable is higher than the value predicted bythe desirability function, culminating in a film with increased me-chanical resistance. This is associated with the lower experimentalelongation at break as compared with the predicted value. Theexperimental values achieved for the bananaflourfilm propertiesarevery similar to the expected ones, which is highly satisfactory.

Since the banana flour is a natural blend composed of starch,protein, lipids, and fiber, the film produced from this raw materialhas mechanical, barrier, and optical properties comparable to those

of other films made from the flour of other plant species: achira(Andrade-Mahecha et al., 2012), amaranth (Tapia-Blcido et al.,2005a), quinoa (Araujo-Farro, 2008), and rice (Dias, Mller,Larotonda, & Laurindo, 2010) (Table 5). Compared with other bio-degradable films, the banana flour film exhibits excellent me-chanical properties; it has higher tensile strength than amaranth,achira, and quinoa flour films and similar tensile strength to the riceflour film. The film produced in this study also presents goodflexibility, as seen by the elongation at break value. The presenceof protein and lipids in the flour could collaborate with the plasti-cizing effect. According to Table 5, the higher Youngs modulus

Fig. 4. Simultaneous optimization of process conditions for the production of banana flour films as function of process variables, predicted response variables, and desirability

profi

les.

Table 4

Experimental validation under the optimized conditions for the production of ba-nana flour films.

Property Predictedvalue

Experimentalvaluea

Relativedeviationb (%)

Tensile strength (MPa) 8.7 9.2 0.2 5.9Elongation at break (%) 25.0 24.2 1.9 3.4Youngs modulus (MPa) 590.5 583.4 46.4 1.2

WVP (1010

g/m s Pa) 2.0 2.1 0.2 4.3Opacity (%) 50.0 51.3 0.3 2.5

a Values obtained in optimal conditions: Cg 19 g/100 g offlour, Tp 81 C,Td 54 C, and RH 48%.

b Relative deviation [(experimental value predicted value)/experimentalvalue] 100.



10/11

value attained for the banana flour film confirms the proximitybetween the polymer chains induced by the high amylose content(27.8 g/100 g of dry starch), which in turn facilitates the formationof a denser polymeric matrix with greater strength and rigidity.

Moreover, the polysaccharide and the protein present in the bananaflour provide additional hydrogen bonding interactions betweenthe polymer chains, accounting for the higher film strength(Sothornvit & Pitak, 2007).

The solubility in water of the banana flour film is relatively lowas compared with the other films (Table 5), a consequence of thehigh level of cohesion within the matrix, which results in a com-pact structure. This specific matrix structure is associated with thecomposition of the banana flour and also to the film-formingtechnique employed in this study. Regarding permeability, thebanana flour film has higher WVP than the amaranth, quinoa, andrice flour films (Table 5). However, its permeability is lower thanthose of other biodegradable films, such as the whey protein iso-late (13.4 1010 g/m s Pa) (Shaw, Monahan, ORiordan, &

OSullivan, 2002), amylose (11.9 1010 g/m s Pa), amylopectin(14.4 1010 g/m s Pa) (Rindlav-Westling, Stading, Hermansson, &Gatenholm, 1998), achira flour (5.3 1010 g/m s Pa) (Andrade-Mahecha et al., 2012), and corn starch (2.6 1010 g/m s Pa)(Garca et al., 2000).

With respect to the optical properties (Table 5), the bananaflourfilm displays a more uniform color, as observed by the lower colordifference value (DE). Furthermore, it is more opaque than theother films prepared from other types offlour.

4. Conclusion

The banana flour seems to be a very promising material to for-mulate coatings and edible films. The presence of native protein,

lipids, and fiber, allied with the high content of amylose in itscomposition, favors the production offilms with better propertiesthan those of other biodegradable films. For the first time, a centralcomposite design together with the desirability function (multi-response analysis) has been described for the optimization of theprocess conditions involved in the production of bananaflour films.This optimization technique has been very useful for the full un-derstanding of the process, since the models allowed for theexperimental design of the production offilms with the desiredcharacteristics within the studied limits.

Acknowledgments

The authors would like to acknowledge the financial support

provided by Coordenao de Aperfeioamento de Pessoal de Nvel

Superior (CAPES) and Conselho Nacional de DesenvolvimentoCientfico e Tecnolgico (CNPq).

References

Alves, V. D., Mali, S., Belia, A., & Grossmann, M. V. E. (2007). Effect of glycerol andamylose enrichment on cassava starch film properties. Journal of Food Engi-neering, 78(3), 941e946.

Andrade-Mahecha, M. M., Tapia-Blcido, D. R., & Menegalli, F. C. (2012). Develop-ment and optimization of biodegradable films based on achira flour. Carbohy-drate Polymers, 88(2), 449e458.

AOAC. (2005). Official methods of analysis (18th ed.). Washington: Association ofOfficial Analytical Chemists.

Araujo-Farro, P. C. (2008). Development and optimization of biodegradables filmsmade of products derivated from Royal variety quinoa (Chenopodium quinoaWilldenow) seeds. Brazil: Unicamp. PhD thesis.

Araujo-Farro, P. C., Podadera, G., Sobral, P. J. A., & Menegalli, F. C. (2010). Develop-ment offilms based on quinoa (Chenopodium quinoa, Willdenow) starch. Car-bohydrate Polymers, 81(4), 839e848.

Arvanitoyannis, I., Nakayama, A., & Aiba, S. (1998). Edible films made fromhydroxypropyl starch and gelatin and plasticized by polyols and water. Carbo-hydrate Polymers, 36(2e3), 105e119.

ASTM. (1999). Standard test method for moisture content of paper and paperboardby oven drying (D644-99). In Annual book of ASTM standards. Philadelphia:American Society for Testing and Materials.

ASTM. (2000). Standard test method of water vapor transmission of materials (E96-00). In Annual book of ASTM standards. Philadelphia: American Society forTesting and Materials.

ASTM. (2002). Standard test method for tensile properties of thin plastic sheeting(D882-02). In Annual book of ASTM standards. Philadelphia: American Societyfor Testing and Materials.

Bergo, P., Sobral, P. J. A., Guevara, V. C. I., & Vadala, A. C. S. (2010). Semi crystallinebehavior of cassava starch films plasticized with glycerol. Materials ScienceForum, 636e637, 745e752.

Bligh, E. G., & Dyer, W. J. (1959). A rapid method of total lipid extraction and pu-rification. Canadian Journal of Biochemistry and Physiology, 37, 911e917.

Bravin, B., Peressini, D., & Sensidoni, A. (2004). Influence of emulsifier type andcontent on functional properties of polysaccharide lipid-based edible films.Journal of Agricultural and Food Chemistry, 52(21), 6448e6455.

Cecchi, H. M. (1999). Fundamentos Tericos e Prticos em Anlise de Alimentos.Campinas: Editora Unicamp.

Chandra, R., & Rustgi, R. (1998). Biodegradable polymers. Progress in Polymer Sci-ence, 23(7), 1273e1335.

Coughlan, K., Shaw, N. B., Kerry, J. F., & Kerry, J. P. (2004). Combined effects ofproteins and polysaccharides on physical properties of whey proteinconcentrate-based edible films. Journal of Food Science, 69(6), E271eE275.

Cuq, B., Gontard, N., Cuq, J.-L., & Guilbert, S. (1997). Selected functional properties offish myofibrillar protein-based films as affected by hydrophilic plasticizers.Journal of Agricultural and Food Chemistry, 45(3), 622e626.

Da Mota, R. V., Lajolo, F. M., Cordenunsi, B. R., & Ciacco, C. (2000). Composition andfunctional properties of banana flour from different varieties. Starch e Strke,52(2e3), 63e68.

Denavi, G., Tapia-Blcido, D. R., An, M. C., Sobral, P. J. A., Mauri, A. N., &Menegalli, F. C. (2009). Effects of drying conditions on some physical propertiesof soy protein films. Journal of Food Engineering, 90(3), 341e349.

Derringer, G., & Suich, R. (1980). Simultaneous optimization of several responsevariables. Journal of Quality Technology, 12(4), 214e219.

Dias, A. B., Mller, C. M. O., Larotonda, F. D. S., & Laurindo, J. B. (2010). Bio-degradable films based on rice starch and rice flour. Journal of Cereal Science,

51(2), 213e

219.

Table 5

Properties offilms prepared with different types offlour.

Property Film matrixa

Achira flour Amaranth flour Quinoa flour Rice flour Banana flour

Plasticizer (g/100 g offlour) 17 22.5 21 20 19Amylose (g/100 g of starch) 30.7 0.1 7.6 0.4 e e 27.8 0.1Thickness (mm) 84 2 83 5 80 2 100 40 87 3Mechanical properties

Tensile strength (MPa) 7.0 0.3 1.5 0.1 4.1 0.5 10.3 1.0 9.2 0.2Elongation at break (%) 14.6 1.1 83.7 5.1 88.4 8.9 2.7 0.5 24.2 1.9Youngs modulus (MPa) 231.7 19.9 21.5 1.4 138.0 40.0 560.7 64.3 583.4 46.4Barrier properties

Solubility in water (%) 38.3 1.9 42.3 1.8 18.7 0.1 e 27.9 1.3WVP (1010 g/m s Pa) 5.3 0.2 0.7 0.2 0.6 0.1 1.1 0.1 2.1 0.2Optical properties

DE* 14.4 0.7 8.9 0.6 18.1 0.4 e 2.7 0.1Opacity (%) 18.0 0.3 6.5 0.9 5.3 0.8 e 51.3 0.3

a References: Andrade-Mahecha et al., 2012 (achira flour), Tapia-Blcido et al., 2005a (amaranth flour), Araujo-Farro, 2008 (quinoa flour), Dias et al., 2010 (rice flour), thiswork (banana flour).



11/11

Diemair, W. (1963). Laboratoriumsbuch fr den Lebensmittelchemiker (8th ed.).Dresden: Verlag Von Theodor Steinkopff.

Garca, M. A., Martino, M. N., & Zaritzky, N. E. (1999). Edible starch films andcoatings characterization: scanning electron microscopy, water vapor, and gaspermeabilities. Scanning, 21(5), 348e353.

Garca, M. A., Martino, M. N., & Zaritzky, N. E. (2000). Lipid addition to improvebarrier properties of edible starch-based films and coatings. Journal of FoodScience, 65(6), 941e944.

Gontard, N., Guilbert, S., & Cuq, J.-L. (1992). Edible wheat gluten films: Influence ofthe main process variables on film properties using response surface method-

ology. Journal of Food Science, 57(1), 190e195.HunterLab. (1996). Application note: CIE L* a* b* color scale, 8, (7), Virginia.International organization for standardization. (1987). Norme internationale: Riz

dtermination de la teneur en amilose . Suisse, 4pp., ISO 6647.Jagannath, J. H., Nanjappa, C., Das Gupta, D. K., & Bawa, A. S. (2003). Mechanical and

barrier properties of edible starch-protein-based films. Journal of Applied Poly-mer Science, 88(1), 64e71.

Kayisu, K., Hood, L. F., & Vansoest, P. J. (1981). Characterization of starch and fiber ofbanana fruit. Journal of Food Science, 46(6), 1885e1890.

Khuri, A. J., & Cornell, F. A. (1996). Response surfaces: design and analyses (2nd ed.).New York: Marcel Dekker.

Krochta, J. M., & De Mulder-Johnston, C. (1997). Edible and biodegradable polymerfilms: challenges and opportunities. Food Technology, 51(2), 61e74.

Mali, S., Grossmann, M. V. E., Garca, M. A., Martino, M. N., & Zaritzky, N. E. (2004).Barrier, mechanical and optical properties of plasticized yam starch films.Carbohydrate Polymers, 56(2), 129e135.

Mali, S., Grossmann, M. V. E., Garca, M. A., Martino, M. N., & Zaritzky, N. E. (2006).Effects of controlled storage on thermal, mechanical and barrier properties ofplasticized films from different starch sources. Journal of Food Engineering,75(4), 453e460.

Mali, S., Sakanaka, L. S., Yamashita, F., & Grossmann, M. V. E. (2005). Water sorptionand mechanical properties of cassava starch films and their relation to plasti-cizing effect. Carbohydrate Polymers, 60(3), 283e289.

Mariniello, L., Di Pierro, P., Esposito, C., Sorrentino, A., Masi, P., & Porta, R. (2003).Preparation and mechanical properties of edible pectin-soy flour films obtainedin the absence or presence of transglutaminase. Journal of Biotechnology, 102(2),191e198.

Marsh, K., & Bugusu, B. (20 07). Food packaging: roles, materials, and environmentalissues. Journal of Food Science, 72(3), R39eR55.

Myllrinen, P., Partanen, R., Seppl, J., & Forssell, P. (2002). Effect of glycerol onbehaviour of amylose and amylopectin films. Carbohydrate Polymers, 50(4),355e361.

Pelissari, F. M., Andrade-Mahecha, M. M., Sobral, P. J. A., & Menegalli, F. C. (2012).Isolation and characterization of the flour and starch of plantain bananas (Musaparadisiaca). Starch - Strke, 64(5), 382e391.

Pitak, N., & Rakshit, S. K. (2011). Physical and antimicrobial properties of ba-nana flour/chitosan biodegradable and self sealing films used for preserving

fresh-cut vegetables. LWT e Food Science and Technology, 44(10), 2310e2315.Rayas, L. M., Hernandez, R. J., & Ng, P. K. W. (1997). Development and character-

ization of biodegradable/edible wheat protein films. Journal of Food Science,62(1), 160e162.

Rindlav-Westling, ., Stading, M., Hermansson, A.-M., & Gatenholm, P. (1998).Structure, mechanical and barrier properties of amylose and amylopectinfilms.Carbohydrate Polymers, 36(2e3), 217e224.

Rodrigues, M. I., & Iemma, A. F. (2005). Planejamento de experimentos e otimizao deprocessos: uma estratgia sequencial de planejamentos. Campinas: Editora Casado Po.

Romero-Bastida, C. A., Bello-Prez, L. A., Garca, M. A., Martino, M. N., Solorza-Feria, J., & Zaritzky, N. E. (2005). Physicochemical and microstructural charac-terization of films prepared by thermal and cold gelatinization from non-conventional sources of starches. Carbohydrate Polymers, 60(2), 235e244.

Shaw, N. B., Monahan, F. J., O Riordan, E. D., & OSullivan, M. (2002). Effect of soya oiland glycerol on physical properties of composite WPI films. Journal of FoodEngineering, 51(4), 299e304.

Sobral, P. J. A., Menegalli, F. C., Hubinger, M. D., & Roques, M. A. (2001). Mechanical,water vapor barrier and thermal properties of gelatin based edible films. FoodHydrocolloids, 15(4e6), 423e432.

Sothornvit, R., & Krochta, J. M. (2001). Plasticizer effect on mechanical properties ofb-lactoglobulin films. Journal of Food Engineering, 50(3), 149e155.

Sothornvit, R., & Pitak, N. (2007). Oxygen permeability and mechanical properties ofbanana films. Food Research International, 40(3), 365e370.

Stading, M., Rindlav-Westling, ., & Gatenholm, P. (2001). Humidity-inducedstructural transitions in amylose and amylopectin films. Carbohydrate Polymers,45(3), 209e217.

Suntharalingam, S., & Ravindran, G. (1993). Physical and biochemical properties ofgreen banana flour. Plant Foods for Human Nutrition, 43(1), 19e27.

Suppakul, P. , Chalernsook, B., Ratisuthawat, B., Prapasitthi, S., &Munchukangwan, N. (2013). Empirical modeling of moisture sorption charac-teristics and mechanical and barrier properties of cassava flour film and theirrelation to plasticizing-antiplasticizing effects. LWT - Food Science and Tech-nology, 50(1), 290e297.

Talja, R. A., Heln, H., Roos, Y. H., & Jouppila, K. (2008). Effect of type and content ofbinary polyol mixtures on physical and mechanical properties of starch-basededible films. Carbohydrate Polymers, 71(2), 269e276.

Tapia-Blcido, D. R., Sobral, P. J. A., & Menegalli, F. C. (2005a). Development andcharacterization of biofilms based on amaranth flour (Amaranthus caudatus).Journal of Food Engineering, 67(1e2), 215e223.

Tapia-Blcido, D. R., Sobral, P. J. A., & Menegalli, F. C. (2005b). Effects of drying temper-ature and relative humidity on the mechanical properties of amaranth flour filmsplasticized with glycerol. Brazilian Journal of Chemical Engineering, 22, 249e256.

Tapia-Blcido, D. R., Sobral, P. J. A., & Menegalli, F. C. (2011). Optimization ofamaranth flour films plasticized with glycerol and sorbitol by multi-responseanalysis. LWTe Food Science and Technology, 44(8), 1731e1738.

Zamudio-Flores, P. B., Bautista-Baos, S., Salgado-Delgado, R., & Bello-Prez, L. A.(2009). Effect of oxidation level on the dual modification of banana starch: themechanical and barrier properties of its films. Journal of Applied Polymer Science,112(2), 822e829.

Zamudio-Flores, P. B., Vargas-Torres, A., Prez-Gonzlez, J., Bosquez-Molina, E., &

Bello-Prez, L. A. (2006). Films prepared with oxidized banana starch: me-chanical and barrier properties. Starch - Strke, 58(6), 274e282.Zhang, P., Whistler, R. L., BeMiller, J. N., & Hamaker, B. R. (2005). Banana starch:

production, physicochemical properties, and digestibility e a review. Carbohy-drate Polymers, 59(4), 443e458.


optimization of process conditions for the production of films based

Documents