Carbohydrate Polymers 127 (2015) 64–71

Contents lists available at ScienceDirect

Carbohydrate Polymers
journal homepage: www.elsevier.com/locate/carbpol

Kinetics and functional effectiveness of nisin loaded antimicrobial

packaging film based on chitosan/poly(vinyl alcohol)
Hualin Wang a,c,∗ , Ru Zhang a , Heng Zhang a , Suwei Jiang a , Huan Liu a , Min Sun a ,
Shaotong Jiang b,c
a
School of Chemistry and Chemical Technology, Hefei University of Technology, Hefei 230009, Anhui, People’s Republic of China
b
School of Biotechnology and Food Engineering, Hefei University of Technology, Hefei 230009, Anhui, People’s Republic of China
c
Anhui Institute of Agro-Products Intensive Processing Technology, Hefei 230009, Anhui, People’s Republic of China

a r t i c l e i n f o a b s t r a c t

Article history: The aim of this study was to evaluate the kinetics and functional effectiveness of Nisin loaded chi-
Received 12 February 2015 tosan/poly(vinyl alcohol) (Nisin–CS/PVA) as an antibacterial packaging film. The films were prepared
Received in revised form 10 March 2015 by coating method and Staphylococcus aureus (S. aureus, ATCC6538) was used as test bacterium. The
Accepted 12 March 2015
intermolecular hydrogen bonds between CS and PVA molecules were confirmed. The elasticity of films
Available online 30 March 2015
was significantly improved by the incorporation of PVA, and the film could also bear a relative high
tensile strength at 26.7 MPa for CS/PVA = 1/1. As CS/PVA ratio decreased, the water vapor permeability
Keywords:
(WVP) decreased and reached its minimum value 0.983 × 10−10 g m−1 s−1 at CS/PVA = 1/1, meanwhile,
Nisin–chitosan/poly(vinyl alcohol)
Antimicrobial packaging film
oxygen permeability (OP) increased but still lower than 0.91 cm3 mm m−2 d−1 kPa−1 for CS/PVA = 1/1 as
Kinetics the CS/PVA ratio was above 1:1. The initial diffusion of nisin (Mt /M∞ < 2/3) from CS/PVA film could be well
Thermodynamics described by the Fickian diffusion equation. Owing to the positively charged nisin at pH below isoelec-
Functional effectiveness tric point (pI, 8.8) and its increasing dissolubility in water as the pH reduced, the diffusion of nisin from
the films strongly depended on pH and ionic strength besides CS/PVA ratio and temperature. Moreover,
the thermodynamic parameters suggested the spontaneous and endothermic diffusion of nisin from the
films. The resulting data can provide some valuable information for the design of film in structure and
ingredient.
© 2015 Elsevier Ltd. All rights reserved.

1. Introduction
Recently, a growing attention has been paid to antibacterial
packaging, which was widely investigated as a response to con-
sumer demands or industrial production trends toward mildly
preserved, fresh, tasty and convenient food products with extended
shelf-life and controlled quality (Aider, 2010). As antimicrobial
packaging films have been showed more efficient in controlling
migration of antimicrobials into the food and keeping antimicrobial
activity over time during the transport and storage (Dutta, Tripathi,
Mehrotra, & Dutta, 2009; Park, Daeschel, & Zhao, 2004), it could play
a profound role in food security assurance. The principle action of
antimicrobial packaging is based on the release of antimicrobial
∗ Corresponding author at: Hefei University of Technology, School of Chemical
Technology, Tunxi 193, Hefei 230009, Anhui, People’s Republic of China.
Tel.: +86 551 62901450; fax: +86 551 62901450.
E-mail address: hlwang@hfut.edu.cn (H. Wang).
entities in film, hence, the entity carrier film and entity diffusion in
film are the two key factors.
As a deacetylated derivative of chitin, chitosan (CS) has shown
a great promise for its application in food preservation by the
virture of non-toxic, biodegradable, bio-functional and biocom-
patible properties. On account of its excellent oxygen barrier and
film-forming ability, CS has been used in bio-packaging for food
preservation (Agulló, Rodríguez, Ramos, & Albertengo, 2003; Dutta
et al., 2009). Nevertheless, the brittleness and moisture barrier
limited the application of pure CS film in packaging. Poly(vinyl
alcohol) (PVA) is a non-toxic water-soluble polymer and has good
flexibility and film-forming ability. Furthermore, it is convenient to
prepare CS/PVA composite film by solution blend method (Chen,
Wang, Mao, & Yang, 2007). At present, serials of investigations
were concerned on CS/PVA composite film in preparation (Chen,
Wang, Mao, Liao, & Hsieh, 2008), mechanical, thermal and surface
properties (El-Hefian, Nasef, & Yahaya, 2011), compatibility (Kumar
et al., 2010), physicochemical and bioactivity (Tripathi, Mehrotra,
& Dutta, 2009) and hydrogel property (Yang, Su, & Yang, 2004).
However, no literature was reported on the diffusion behavior of

http://dx.doi.org/10.1016/j.carbpol.2015.03.058
0144-8617/© 2015 Elsevier Ltd. All rights reserved.
 H. Wang et al. / Carbohydrate Polymers 127 (2015) 64–71
antimicrobial entities in CS/PVA composite film in antibacterial
packaging to the best of our knowledge at present.
As an antibacterial entity, the release and antibacterial
behaviors of nisin have been investigated in films such as
methylcellulose-hydroxypropyl methylcellulose (Grower,
Cooksey, & Getty, 2004), starch-based (Sanjurjo, Flores,
Gerschenson, & Jagus, 2006), cellulose (Nguyen, Gidley, & Dykes,
2008) and heat-pressed and cast films (Cha, Cooksey, Chinnan,
& Park, 2003). The diffusivity of antimicrobial entity in film can
provide important information for the design of film in structure
and ingredient. Much valuable work has been done on deter-
mining the diffusivity of antimicrobial entities in antimicrobial
films in food systems such as acetic and propionic acids from
chitosan-based (Ouattara, Simard, Piette, Begin, & Holley, 2000),
linalool and methylchavicol from polyethylene-based (Suppakul,
Sonneveld, Bigger, & Miltz, 2011) and potassium sorbate from
k-carrageenan-based (Choi et al., 2005) antimicrobial packaging
films. Thermodynamic parameters, including enthalpy (1H0 ),
entropy (1S0 ) and Gibbs free energy (1G0 ), can provide additional
information regarding the inherent energetic changes associated
with the diffusion. However, few attentions were paid on the
diffusion of antimicrobial entities from films.
In the present work, we aimed to evaluate the kinetics and
functional effectiveness of nisin loaded chitosan/poly(vinyl alco-
hol) (Nisin–CS/PVA) as an antibacterial packaging film and focused
on the diffusion of nisin from the film by kinetics and thermody-
namics, and the effects of CS/PVA ratio, temperature, pH and ionic
strength on the diffusion efficiency. To this aim, an operationally
simple coating method was used to prepare the films by an auto-
matic film applicator. The mechanical property, structure, water
vapor permeability (WVP) and oxygen permeability (OP) of the
CS/PVA films as well as the antimicrobial activity of Nisin–CS/PVA
films were assessed. Nisin was used as a model antimicrobial entity
and Staphylococcus aureus (S. aureus, ATCC6538) as test bacterium.
2. Experimental
2.1. Materials
Chitosan (CS, Mw 300 kDa, viscosity 100 MPa s, DD 95%) was
purchased from Zhejiang Aoxing Biochemical CO., Ltd., (Zhejiang,
Chain). Polyvinyl alcohol (PVA, DP 1750 ± 50) was supplied from
Sinopharm Chemical Reagent CO., Ltd., (Shanghai China). The 2.5%
nisin preparation was a powdered product from MP Biomedicals,
LLC (Solon, Ohio, USA). The product was labeled with a nisin con-
centration of 1000 IU mg−1 solids. Staphylococcus aureus (S. aureus,
ATCC6538) was provided by China Center of Industrial Culture Col-
lection (Beijing, China). All the other chemical reagents used were
of analytical grade and available from Sinopharm Chemical Reagent
Co. Ltd., (Shanghai, China).
2.2. Preparation of samples
Chitosan was dissolved into 2% (v/v) acetic acid to obtain 1 wt.%
CS solution. A PVA aqueous solution (5 wt.%) was prepared by
dissolving PVA powder into distilled water at 85 ◦ C. Afterwards,
serials of CS/PVA blend solutions with different ratio of CS/PVA
(3:1, 2:1, 1:1, 1:2 and 1:3, w/w) were prepared by mixing. Cor-
respondingly, the films prepared were defined as CS/PVA = 3/1,
CS/PVA = 2/1, CS/PVA = 1/1, CS/PVA = 1/2 and CS/PVA = 1/3, respec-
tively. All the films were preformed on an AFA-III automatic film
applicator (Hefei Kejing Material Technology Co., Ltd., China). The
CS/PVA blend solution was coated onto a substrate polyethylene
(PE) film. After drying at 35 ◦ C for 72 h, the films were peeled from
65
PE film and vacuum dried at 60 ◦ C for 24 h to remove the residue
water and acetic acid.
Nisin–CS/PVA films were prepared on the AFA-III automatic film
applicator and the parameters were the same as that of CS/PVA
films. Required nisin (1, 3, 5, 8, 10 and 20%, w/w, based on the
weight of CS/PVA) was gradually added into the CS/PA solutions and
stirred continuously at room temperature for 4 h before coating on
the substrate film.
2.3. Measurement and characterization
Film thickness was measured with a hand-held micrometer
(BC Ames Co., Waltham, MA, USA). Nine film thickness measure-
ments were taken from along the length of each specimen and
the mean value was used in calculating tensile strength and dif-
fusion coefficient. The tensile strength (TS) and elongation at break
(EB) of specimens were conducted on a TA-XTPlus Texture Anal-
yser (Stable Micro Systems, Co., UK). All the dry films were cut into
rectangular blocks (1 cm × 10 cm in width and length).
The morphologies of films were estimated by scanning electron
microscopy (SEM; SU8020, Hitachi, Japan) under an accelerating
voltage of 15 kV, and all of the specimens were sputter-coated with
a layer of gold. Fourier transform infrared (FTIR) spectroscopy was
conducted with a Nicolet 6700 spectrometer (Thermo Nicolet, USA)
with KBr pellets.
2.4. Permeability of films
Water vapor permeability (WVP) were measured using a mod-
ified method as described by Limpan, Prodpran, Benjakul, and
Prasarpran (2010). The specimens, sealed on beakers, containing
silica gel (0% RH) were placed in incubator containing distilled
water. The chamber of incubator was provided with a psychrometer
for relative humidity, and the temperature of incubator was main-
tained at 30 ◦ C. The moisture absorbed was estimated by weighing
of beakers at 3 h intervals during 3 days. WVP (g m−1 s−1 Pa−1 ) was
determined for three replicate specimens each type, as follows:
w×x
WVP = (1)
A × t × 1P
where w is the weight gain of the beaker (g), x is the film thickness
(m), A is the area of exposed film (m2 ), t is the time of weight gain
(s), and 1P is the water vapor partial pressure difference (Pa) across
the two sides of film calculated on the basis of relative humidity.
Oxygen transmission rate (OTR, according to ASTMD1434)
of films was determined at 23 ◦ C and 0% RH on a N500 gas
permeameter (Guangzhou Biaoji packaging equipment Co., Ltd.,
Guangzhou, China). Oxygen permeability (OP) was calculated from
OTR (cm3 m−2 d−1 kPa−1 ) by the following equation:
OP = OTR × thickness (2)
The thickness and open testing area of each sample in three
parallel measurements were approximately 100 mm and 50 cm2 .
2.5. Antibacterial activity assay
The antibacterial activity evaluation of film against S. aureus
was performed in liquid culture medium (Wang et al., 2013). The
concentration of S. aureus for bacterial suspension after rejuvenat-
ing was set as 1.0, and the absorbance at 600 nm for each sample
with concentration of S. aureus at 0.05, 0.1, 0.15, 0.2 and 0.25 was
measured. On the basis of experimental data, the linear regression
equation from the calibration curve was determined as following:
Y = 0.8196X − 0.00406(R2 = 0.99925) (3)
66 H. Wang et al. / Carbohydrate Polymers 127 (2015) 64–71
where the X is the relative concentration of S. aureus and Y is the
absorbance.
1 ml rejuvenated bacterial suspension was transferred into 9 ml
KMB broth, then 30 mg Nisin–CS/PVA film was soaked into liquid
culture medium containing S. aureus after sterilization by ultravi-
olet light. The culture media were propagated for 24 h on a shaker
platform (SHZ-82, Changzhou Guohua Electric Appliance Co., Ltd.,
Jiangsu, China) at 160 rpm and 37 ◦ C. The absorbance at 600 nm was
measured using a spectrophotometer (UV-754PC, Shanghai Jinghua
Technology Instruments Co., Ltd., Shanghai, China) and relative
concentration of S. aureas was calculated based on Eq. (3).
2.6. Diffusion test
Series of nisin solutions (50, 100, 150, 200, 250 and 300 mg ml−1 )
were prepared by dissolving nisin into distilled water in Erlenmeyer
flasks. 1 ml above solution was stained with 2 ml Coomassie bril-
liant blue G250 solution. The absorbance at 595 nm was measured
for each solution (1 ml distilled water and 2 ml Coomassie brilliant
blue G250 blend solution as blank). On the basis of experimental
data, the regression line from the calibration curve was expressed
as following:
Y = 0.00142X + 0.09179(R2 = 0.99548)
where X (mg ml−1 ) is the concentration of nisin, and Y is the
absorbance.
Films were cut into 4 cm squares, then the square film was
covered with aluminum foil tape on one side and immersed in a
Erlenmeyer flak containing 100 ml distilled water. The flaks were
shaken on the shaker platform (100 rpm, 25 ◦ C) and achieved diffu-
sion equilibration. The absorbances at 595 nm were measured with
the spectrophotometer. The concentrations of nisin at different
time and diffusion equilibration were determined based on Eq. (4).
The accumulative release percentage of nisin from the films was
calculated using the following equation:
Mt
Accumulative release(%) = × 100%
M0
where Mt (mg) is the released nisin at time t, M0 (mg) is the total
trapped or entrapped nisin.
The pH values of diffusion solutions were adjusted by adding
a thimbleful of 0.01 M HCl or 0.01 M NaOH solution. Most
experiments were carried out at pH 6.5 ± 0.1. According to our
experimental data, the diffusion percentage is higher if the solu-
tion pH is lower, while the diffusion percentage is lower if the pH
is higher. Thus, a pH approximately 6.5 is representative.
(A) PVA
1335
CS 1430 1139
3352
2944
Tranmittance
1654 1596
1390
2875
3430
CS/PVA
3330
4000 3500 3000 2500 2000 1500
-1
Wave number(cm )
Fig. 1. (A) FT-IR spectra of samples; (B) proposed structure of CS/PVA (Wan Ngah, Kamari, & Koay, 2004). Intermolecular hydrogen bonds existed between CS and PVA
molecules.
2.7. Statistical analysis
Each experiment was repeated three times. Statistical analysis
was performed using the unpaired Student’s t-test, and the results
were expressed as the means ± standard deviation (SD). A value of
p < 0.05 was considered to be statistically significant.
3. Results and discussion
3.1. Structure and mechanical property of CS/PVA film
FTIR spectroscopy was used to investigate the interaction
between CS and PVA and the spectra were shown in Fig. 1. In
the spectrum of PVA, characteristic peaks for PVA were present
at: 3352 ( OH stretching); 2944 ( CH anti-symmetric stretch-
ing); 1430, 1335 ( CH angular deformation) and 1090, 1142 cm−1
(C O stretching, band at 1142 cm−1 was mostly attributed to the
crystallinity of PVA) (Wang et al., 2014). In the spectrum of CS,
characteristic peaks of its saccharide structure were present at
902, 1020 and 1083 cm−1 (skeletal vibrations referring to the C O
stretching), and 1151 cm−1 (anti-symmetric stretching of C O C
bridge, glycosidic band) (Ávila, Bierbrauer, Pucci, López-González,
& Strumia, 2012; Li, Hu, Ren, Worley, & Huang, 2013). The absorp-
(4) tion bands at 1654, 1596 and 1390 cm−1 were assigned to the
stretching vibration of C O (amide I), bending vibration of NH2
(amide II) and bending vibration of CH2 , respectively, while the
absorption band at 2875 cm−1 corresponded to the bending vibra-
tion of CH in CH3 from NHCOCH3 . In addition, the strong
and broad band at around 3430 cm−1 was due to the stretching
vibration of OH, extension vibration of NH and intermolecu-
lar hydrogen bonds of the polysaccharide moieties (Sajomsang,
Ruktanonchai, Gonil, & Warin, 2010). As compared to the spectra
of CS and PVA, the band at around 3430 cm−1 concerned with OH
and NH was broadened and obviously shifted to a lower wave
number around 3330 cm−1 in the spectrum of CS/PVA film, indicat-
ing the formation of intermolecular hydrogen bonds between CS
and PVA molecules (Fig. 1B).
(5) Tensile strength (TS) and elongation at break (EB) can reflect
strength and elasticity of film. Fig. 2 demonstrated the results of TS
and EB of the films. For pure CS film, it had a high TS (46.8 MPa),
but very low EB (21.6%), while pure PVA film exhibited better elas-
ticity and the EB reached 222.8%. As expected, its EB was sharply
improved by incorporating PVA into CS matrix, and the film could
also bear a relative high TS at 26.7 MPa for CS/PVA = 1/1. The reason
was likely that CS had a high molecular weight and hard back-
bones as compared to PVA. Similar trend was reported by Srinivasa,
Ramesh, Kumar, and Tharanathan (2003). To better understand the
(B) CH2 CH
n
O
H
1090
Intermolecular
HOH2C
HOH2C O
665
O
902 O
O
1151 1083 1022 O
O NH 2
NH2
O r H
ula Intermolecular
H
olec H
ram
Int O
CH2 CH
n
1000 500
 H. Wang et al. / Carbohydrate Polymers 127 (2015) 64–71 67

Fig. 2. Tensile properties of films in: (A) tensile strength and (B) elongation at break. Tensile strength of films decreased seriously (p < 0.05) by incorporating PVA into CS
matrix, meanwhile, the corresponding elongation at break was sharply improved (p < 0.05). The data were representative of the results from repeated experiments (n = 3)
and expressed as the mean ± SD.

Fig. 3. Cross-section SEM photographs of specimens at break. Cross-section of CS/PVA film showed lots of tensile cracks compared to CS and PLA films.

tensile properties of films, SEM study was performed at the cross-
section of the corresponding pure CS, pure PVA and CS/PVA = 1/1
films at break (Fig. 3). The cross-section SEM photograph of pure
CS (Fig. 3a) and PVA film (Fig. 3c) at break showed a smooth
surface and no obvious tensile cracks. However, lots of cracks
appeared on surface of CS/PVA = 1/1 film at break (Fig. 3b), suggest-
ing the incorporation of PVA into CS matrix improved elasticity of
the film.
3.2. Water vapor and oxygen permeability
As could be seen from WVP results of the films (Fig. 4A),
the WVP decreased with the decrease of CS/PVA ratio and
then increased again after reached the minimum value
0.983 × 10−10 g m−1 s−1 Pa−1 for CS/PVA = 1/1. The reason may be
attributed to the high crosslink effects formed by intermolecular
hydrogen bonds between CS and PVA molecules for CS/PVA = 1/1,

Fig. 4. Effect of CS/PVA ratio on (A) water vapor permeability and (B) oxygen permeability. With decreasing CS/PVA ratio, water vapor permeability of films decreased and
then increased after reached the minimum value (p < 0.05), meanwhile, the corresponding oxygen permeability improved significantly (p < 0.05). The data were representative
of the results from repeated experiments (n = 3) and expressed as the mean ± SD.
68 H. Wang et al. / Carbohydrate Polymers 127 (2015) 64–71

which makes the film structure become more compact (Zhang,

Xu, & Jiang, 2012). Furthermore, the WVP of the pure PVA film

Relative concentration of S. aureus

was slightly higher than that of pure CS film owing to the higher
hydrophilicity toward CS.
Fig. 4B illustrated OP of pure and blended films. Pure CS and PVA
had OP values of 0.05 and 3.13 cm3 mm m−2 d−1 kPa−1 . As non-polar
molecules, oxygen molecules would dissolve to a greater extent in
less polar polymers, thus, a less polar film would have higher OP
due to higher sorption toward oxygen. When PVA was incorporated
into CS matrix, the polarity of matrix was weakened owing to the
decrease in amount of NH2 and OH. As a result, the blended
films showed a decrease in oxygen barrier with decreasing CS/PVA
ratio, but OP value was still lower than 0.91 cm3 mm m−2 d−1 kPa−1
for CS/PVA = 1/1 as the CS/PVA ratio was above 1:1. Based on the
mechanical property, WVP and OP results, the CS/PVA ratios were
chosen at 1:1 and above to prepare Nisin–CS/PVA antimicrobial
films.
3.3. Antimicrobial activity evaluation in liquid culture medium
As it is known, antimicrobial activity is inversely proportional
to the relative concentration of S. aureus. Fig. 5 depicted the rel-
ative concentration of S. aureus for Nisin–CS/PVA films in liquid
culture medium after incubation for 24 h at 37 ◦ C. It was found
that the relative concentration of S. aureus decreased sharply
from 100% to 11.65% when nisin content increased from 0 to
10%, and then decreased gradually. The releasing dosage of nisin
into bacterial suspension increased in quantity with the increase
of the nisin content in films, therefore, the growth of S. aureus
was inhibited effectively. While nisin content was above 10%, the
increase of nisin content had no obvious effect on the antimicrobial
activity.
3.4. Diffusion kinetics of nisin from Nisin–CS/PVA film
3.4.1. Release of nisin
For assessing the release behavior and acquiring basic informa-
tion for diffusion kinetics, the accumulative release percentages of
nisin from films with CS/PVA ratios at 3:1, 2:1 and 1:1 were cal-
culated by Eq. (5) and plotted versus time in Fig. 6A. Each curve
showed initial burst release phenomenon, and reached a plateau
after a significantly increase in accumulative release percentage,
1.0
0.8
0.6
0.4
0.2
0.0
0 5 10 15 20
Nisin content (%)
Fig. 5. Relative concentrations of S. aureus of Nisin–CS/PVA film (CS/PVA = 1:1, prop-
agated 24 h, 37 ◦ C). The relative concentration of S. aureus decreased significantly
compared to CS/PVA film without nisin (p < 0.05). The data were representative of
the results from repeated experiments (n = 3) and expressed as the mean ± SD.
suggesting a good controlled release behavior for Nisin–CS/PVA
film (Selvam, Manikandan, Kennedy, & Vijaya, 2013). It was
noteworthy that the accumulative release percentage at diffu-
sion equilibrium decreased with the decrease of CS/PVA ratio.
The diffusion of nisin on or near surface of film resulted in the
initial burst release under the diffusion driving force by nisin con-
tent (Moditsi, Lazaridou, Moschakis, & Biliaderis, 2014; Ozdemir &
Floros, 2001). While the nisin being trapped into the inner core
of the matrix would take longer time to be released owing to
longer diffusion pathway and showed an increase in accumula-
tive release. Nisin was an amphiphilic molecule simultaneously
displaying hydrophobic and hydrophilic behaviors, however, it
had the better affinity toward hydrophilic materials (Karam et al.,
2013a, 2013b). Therefore, the incorporation of CS into PVA matrix
increased the hydrophobicity of film, resulting in that the accumu-
lative release percentage at equilibrium increased with the increase
of CS/PVA ratio.

70
(A) (B)
1.0
60
Accumulative release(Mt/M0,%

50 0.8

40
0.6
Mt /M

30
0.4
20

10 0.2
CS/PVA=1:1
CS/PVA=2:1
0
CS/PVA=3:1
0.0
-10
0 3000 6000 9000 12000 15000 0 3000 6000 9000 12000 15000
t (min) t (min)

Fig. 6. (A) Accumulative release of nisin from Nisin–CS/PVA films (T = 25 ◦ C, pH = 6.5 ± 0.1, nisin 10 wt.%). Each curve showed initial burst release phenomenon, and reached
a plateau after a significantly increase (p < 0.05); (B) fractional diffusion of nisin in films Mt /M∞ versus time t, inset: linear regression of Mt /M∞ versus square root of
time t1/2 (T = 25 ◦ C, pH = 6.5 ± 0.1, nisin 10 wt.%). Mt /M∞ increased with the increase of CS/PVA ratio, and each curve showed a sharp increase before a plateau (p < 0.05).
The initial diffusion of nisin (Mt /M∞ < 2/3) from films could be well described by the Fickian diffusion equation. The data (mean ± SD) are results from three independent
experiments.
 H. Wang et al. / Carbohydrate Polymers 127 (2015) 64–71 69

3.4.2. Determination of diffusivity of nisin -26.5

An initial mathematical model for analysis of nisin diffusion
from Nisin–CS/PVA film could be derived from Fick’s second law
(Crank, 1975): -27.0

2
∂C ∂ C
=D (6) -27.5
∂t ∂x2
where C (kg m−3 ) is the nisin concentration in the film, D (m2 s−1 )

ln D
is the diffusivity or diffusion coefficient, x (m) is the coordinate -28.0
measured in the direction of transport, and t (s) is the time. The fol-
lowing assumptions were made: (1) The film prepared was plane
sheet. (2) The dispersion of nisin in film was uniform and the nisin -28.5
CS/PVA=1:1
concentration in aqueous medium was zero. (3) The diffusion of CS/PVA=2:1
nisin was regarded as one-dimensional diffusion (As mentioned CS/PVA=3:1
-29.0
in experimental, the specimen was covered with aluminum foil
tape on one side) and a non-steady state phenomenon of non-
concentration-dependent diffusion. Based on these assumptions, 0.0031 0.0032 0.0033 0.0034 0.0035 0.0036

a classical solution was obtained (Crank, 1975):

-1
1/T(K )
∞
D(2n + 1)2 2 t
 
4C0 X 1 (2n + 1) x Fig. 7. Temperature effects on diffusion coefficient (CS/PVA = 1:1, pH = 6.5 ± 0.1,
C= exp − sin (7) nisin 10 wt.%). The effect of temperature on diffusion coefficient could be described
 2n + 1 h2 h
n=0 by Arrhenius activation model. The data (mean ± SD) are results from three inde-
pendent experiments.
where C0 (kg m−3 ) is the initial content of nisin in film, h (m) is the
thickness of Nisin–CS/PVA film. Eq. (7) integrated over the thickness
cases, linearity with respect to t1/2 predicted by Eq. (9) for the ini-
of the film could be expressed in linear form (Choi et al., 2005;
tial portion of the curve (Mt /M∞ < 2/3) was strong, indicating that
Moditsi et al., 2014; Yoshida, Bastos, & Franco, 2010).
the diffusion mechanism could be well described by Fickian diffu-
∞
D(2n + 1)2 2 t sion. The linear regressions of Mt /M∞ versus t1/2 for CS/PVA films
 
Mt X 8
=1− exp − (8) at different ratios were shown in the inset of Fig. 6B, correspond-
M∞ (2n 1)2 2
+ h2
ingly, the diffusion coefficient D calculated by Eq. (10) was listed
n=0
in Table 1. Similarly, the D values for CS/PVA films at 5 ◦ C and 45 ◦ C
where Mt (mg) is the amount of nisin diffused at time t, M∞ (mg)
were obtained and summarized in Table 1. The high correlation
is the amount of nisin diffused at equilibrium. For Mt /M∞ < 2/3,
coefficient R2 > 0.998 indicated that the predominant diffusion was
Mt /M∞ could be calculated by the following formula (Crank, 1975;
Fickian diffusion. As expected, Table 1 showed a decrease in D value
Malley, Bardon, Rollet, Taverdet, & Vergnaud, 1987; Moditsi et al.,
with the decrease CS/PVA ratio at each given temperature due to
2014; Peppas, 1985; Yoshida et al., 2010):
the higher affinity of nisin toward PVA, and higher temperature
Mt Dt
 1/2 caused an increase in D value for the same CS/PVA films.
=4 = kt 1/2 (9)
M∞ h2
3.4.3. Temperature dependence of diffusivity
where k (1/s1/2 ) is the slope of the linear regression of Mt /M∞ versus The temperature effect on diffusivity is generally described by
t1/2 . Consequently, the diffusivity could be counted by the following the Arrhenius activation energy mode.
equation:  E 
a
 kh 2 D = D0 exp − (11)
D=  (10) RT
4
where D0 (m2 s−1 ) is a constant, Ea (J mol−1 ) is activation energy of
Based on Fig. 6A, the whole curves of Mt /M∞ versus time were the diffusivity, R (J mol−1 K−1 ) is universal gas constant and T (K) is
demonstrated in Fig. 6B. As could be seen from Fig. 6B, Mt /M∞ absolute temperature.
depended on film composition at given time and increased with The Arrhenius plots were derived from logarithmic transform
the increase of CS/PVA ratio owing to the lower affinity of nisin of Eq. (11) (Fig. 7). From the slope of line, the corresponding
toward CS as compared to PVA. Moreover, each curve showed a Ea were summarized in Table 2. The high correlation coefficient
similar shape by increasing sharply before reaching a plateau. In all values (R2 > 0.999) implied that the effect of temperature on diffu-
sion coefficient could be described by Arrhenius activation model.
Table 1 The greater the Ea value, the more sensitive the diffusivity to
Diffusivity of nisin from Nisin–CS/PVA films (pH = 6.5 ± 0.1, nisin 10 wt.%). temperature changes. In addition, Ea might predict the molecu-
CS/PVA ratio Temperature (◦ C) D (×10−13 m2 /s)a R2 b lar interactions between the diffusate and diffusion medium. The

3:1 5 8.071 ± 0.26 0.99925

25 14.644 ± 0.65 0.99971 Table 2
45 27.366 ± 0.80 0.99910 Activation energy of diffusion for nisin from Nisin–CS/PVA films (pH = 6.5 ± 0.1, nisin
2:1 5 4.252 ± 0.14 0.99827 10 wt.%).
25 9.197 ± 0.54 0.99942
CS/PVA ratio Ea (kJ/mol)a R2 b
45 18.814 ± 0.73 0.99916
1:1 5 2.468 ± 0.15 0.99903 3:1 21.997 ± 0.813 0.99178
25 6.489 ± 0.23 0.99992 2:1 26.843 ± 1.342 0.99838
45 16.074 ± 0.68 0.99831 1:1 33.813 ± 1.276 0.99798
a a
Nisin diffusivity was calculated using Eq. (10). The data (mean ± SD) are results Ea was calculated using Eq. (11). The data (mean ± SD) are results from three
from three independent experiments. independent experiments.
b
Given correlation coefficient (R2 ) was the largest one among replications (n = 3). b
Given correlation coefficient (R2 ) was the largest one among replications (n = 3).
70 H. Wang et al. / Carbohydrate Polymers 127 (2015) 64–71

higher Ea value indicated a stronger interaction between nisin and
CS/PVA matrix, because more work in the form of energy was
needed to overcome the energetic barrier. As mentioned above,
nisin had higher affinity toward PVA as compared to CS. Hence, the
decrease of CS/PVA ratio caused an increase in Ea value at same
temperature.
repulsion suppressed by the increasing ionic strength, meanwhile,
a higher ionic strength could also enhance the hydrophobicity of
nisin (Zhong, Jin, Davidson, & Zivanovic, 2009). However, no dis-
tinguish in M∞ /M0 was observed in NaCl solutions owing to the
poor solubility of nisin at pH > 8.8.
3.5. Thermodynamic parameter

3.4.4. Effects of pH and ionic strength on diffusion A distribution coefficient (Kd ) associated with the total
There was a significant difference in the diffusion percentage entrapped nisin M0 in film and the amount of nisin diffused at
at equilibrium (M∞ /M0 ) as a function of pH for the diffusion of equilibrium M∞ was adapted:
nisin from film in NaCl solutions (Fig. 8A), and each curve could
M∞ m
be divided into three regions. As pH reduced, M∞ /M0 showed a Kd = · (12)
M0 − M∞ V
slightly increase in region I (pH > 8.8), a significantly increase in
region II (pH 4.5–8.8) and a high level at pH < 4.5 (region III). Nisin where V (ml) was the volume of the solution and m (g) was the
was positively charged at pH below the isoelectric point (pI, 8.8) weight of each film.
(Xiao, Gömmel, Davidson, & Zhong, 2011), and the dissolubility The enthalpy change (1H0 ) and entropy change (1S0 ) for the
in water was improved as the pH reduced (Bani-Jaber, McGuire, diffusion of nisin from the film were calculated by the intercept and
Ayres, & Daeschel, 2000). On the other hand, CS was also positively slope of the plot of lnKd versus 1/T (Fig. 8B) based on temperature-
charged in acidic situation and the quantity of amino groups pro- dependent distribution coefficient relationship:
tonated increased at a reduced pH (Jiang, Wang, Li, Jiang, & Lu, 2005;
1S 0 1H 0
Katas & Alpar, 2006). In region I, M∞ /M0 depended on the solubility ln Kd = − (13)
R RT
of nisin and showed a slightly increase at a reduced pH. Neverthe-
less, the electrostatic repulsion from the positively charged nisin where R (8.314 J mol−1 K−1 ) is the ideal gas constant, and T (K) is
and CS in acidic situation favored nisin to diffuse from film to aque- the temperature in Kelvin. Meanwhile, the corresponding 1G0 was
ous solution. As a result, a significantly increase in M∞ /M0 appeared obtained by the general expression:
in region II (especially at pH < 7.0) owing to the enhancing electro-
1G0 = 1H 0 − T1S 0 (14)
static repulsion at a reduced pH.
As shown in Fig. 8A, M∞ /M0 also depended on NaCl concentra- Thermodynamic parameters based on Eqs. (13) and (14) were
tions at each given pH. An increase in NaCl concentration resulted summarized in Table 3. The positive 1H0 suggested the diffu-
in a lower M∞ /M0 at pH < 8.8. This behavior was most likely due sion was endothermic, because kinetic energy was needed for the
to the Debye length and the effective distance of electrostatic entrapped nisin to travel through the CS/PVA matrix. The positive

(A) 100 region III (B) 2.4

90 region II
/M0,%)

CS/PVA=1:1
2.1 CS/PVA=2:1
80 CS/PVA=3:1
region I
70 1.8
Diffusion percentage(M

60
ln Kd(g/L)

1.5
50
1.2
40
1.00M NaCl
0.9
30 0.10M NaCl
0.01M NaCl
20 0.6
0 M NaCl
10 0.3
0 0.0031 0.0032 0.0033 0.0034 0.0035 0.0036
2 4 6 8 10 -1
1/T (K )
pH

Fig. 8. (A) Diffusion percentage at equilibrium (M∞ /M0 ) as a function of pH in NaCl solutions (T = 25 ◦ C, CS/PVA = 1:1, nisin 10 wt.%). At given NaCl solution, M∞ /M0 showed
a slightly increase at pH > 8.8, a significantly increase at pH 4.5–8.8 (p < 0.05) and a high level at pH < 4.5. Moreover, an increase in NaCl concentration resulted in a lower
M∞ /M0 at pH < 8.8. (B) Plots of lnKd versus 1/T for the diffusion of nisin from the film (pH = 6.5 ± 0.1, nisin 10 wt.%). Plots showed a good linear correlation between d lnKd and
1/T. The data (mean ± SD) are results from three independent experiments.

Table 3
Thermodynamic parameters of diffusion for nisin from Nisin–CS/PVA films (pH = 6.5 ± 0.1, nisin 10 wt.%).

CS/PVA ratio 1S0 (J (mol K)−1 )a 1H0 (kJ mol−1 )a 1G0 (kJ mol−1 )b

278.15 K 298.15 K 318.15 K

3:1 84.929 ± 2.387 9.006 ± 0.279 −2.602 ± 0.082 −4.321 ± 0.116 −6.020 ± 0.125
2:1 63.729 ± 2.874 16.260 ± 0.748 −1.466 ± 0.063 −2.741 ± 0.092 −4.015 ± 0.103
1:1 34.632 ± 1.153 21.000 ± 0.861 −0.627 ± 0.018 −1.318 ± 0.064 −2.012 ± 0.092
a
1S0 and 1H0 were calculated using Eq. (13). The data (mean ± SD) are results from three independent experiments.
b
1G0 were calculated using Eq. (14). The data (mean ± SD) are results from three independent experiments.
 H. Wang et al. / Carbohydrate Polymers 127 (2015) 64–71
1S0 might be related to the affinity of nisin toward CS/PVA matrix
and the dispersion change of nisin in CS/PVA matrix. It was note-
worthy that the value of 1G0 was negative and decreased with an
increase in temperature, indicating that the diffusion of nisin in
CS/PVA matrix was spontaneous and the spontaneity improved by
increasing temperature.
4. Conclusion
The present study reported Nisin–CS/PVA antibacterial films
prepared by coating method. The interaction between CS and PVA
molecules was confirmed to be intermolecular hydrogen bonds.
The incorporation of PVA into CS matrix could sharply improve the
elasticity of the film, and the film could also bear a relative high
tensile strength at 26.7 MPa for CS/PVA = 1/1. With the decrease
of CS/PVA ratio, the WVP decreased and reached the minimum
value 0.983 × 10−10 g m−1 s−1 at CS/PVA = 1/1, at the same time,
OP increased but was still lower than 0.91 cm3 mm m−2 d−1 kPa−1
for CS/PVA = 1/1 as the CS/PVA ratio was above 1:1. The initial
diffusion of nisin (Mt /M∞ < 2/3) from CS/PVA film could be well
described by the Fickian diffusion equation. M∞ /M0 showed a
slightly increase in region I (pH > 8.8), a significantly increase in
region II (pH 4.5–8.8) and a high level at pH < 4.5 (region III). The for-
eign ionic strength had significant influence on the diffusion of nisin
in CS/PVA films at pH < 8.8, and no distinguish pH > 8.8. The ther-
modynamic parameters indicated that the diffusion of nisin was
endothermic and spontaneous. The experimental results showed
that the films revealed well controlled release and better antimi-
crobial activity against S. aureus, which may have potential as an
active film in food packaging.
Acknowledgments
Financial support from National Natural Science Foundation of
China (31371859) is gratefully acknowledged.
References
Agulló, E., Rodríguez, M. S., Ramos, V., & Albertengo, L. (2003). Present and future
role of chitin and chitosan in food. Macromolecular Bioscience, 3(10), 521–530.
Aider, M. (2010). Chitosan application for active bio-based films production and
potential in the food industry: Review. LWT—Food Science and Technology, 43(6),
837–842.
Ávila, A., Bierbrauer, K., Pucci, G., López-González, M., & Strumia, M. (2012). Study
of optimization of the synthesis and properties of biocomposite films based on
grafted chitosan. Journal of Food Engineering, 109(4), 752–761.
Bani-Jaber, A., McGuire, J., Ayres, J., & Daeschel, M. (2000). Efficacy of the antimi-
crobial peptide nisin in emulsifying oil in water. Journal of Food Science, 65(3),
502–506.
Cha, D., Cooksey, K., Chinnan, M., & Park, H. (2003). Release of nisin from var-
ious heat-pressed and cast films. LWT—Food Science and Technology, 36(2),
209–213.
Chen, C.-H., Wang, F.-Y., Mao, C.-F., Liao, W.-T., & Hsieh, C.-D. (2008). Stud-
ies of chitosan: II. Preparation and characterization of chitosan/poly(vinyl
alcohol)/gelatin ternary blend films. International Journal of Biological Macro-
molecules, 43(1), 37–42.
Chen, C. H., Wang, F. Y., Mao, C. F., & Yang, C. H. (2007). Studies of chitosan. I. Prepa-
ration and characterization of chitosan/poly(vinyl alcohol) blend films. Journal
of Applied Polymer Science, 105(3), 1086–1092.
Choi, J., Choi, W., Cha, D., Chinnan, M., Park, H., Lee, D., et al. (2005). Diffusivity of
potassium sorbate in k-carrageenan based antimicrobial film. LWT—Food Science
and Technology, 38(4), 417–423.
Crank, J. (1975). The mathematics of diffusion (2nd ed.). New York: Oxford University
Press.
Dutta, P., Tripathi, S., Mehrotra, G., & Dutta, J. (2009). Perspectives for chitosan based
antimicrobial films in food applications. Food Chemistry, 114(4), 1173–1182.
El-Hefian, E. A., Nasef, M. M., & Yahaya, A. H. (2011). Preparation and characterization
of chitosan/poly(vinyl alcohol) blended films: Mechanical, thermal and surface
investigations. Journal of Chemistry, 8(1), 91–96.
Grower, J., Cooksey, K., & Getty, K. (2004). Release of Nisin from methylcellulose-
hydroxypropyl methylcellulose film formed on low-density polyethylene film.
Journal of Food Science, 69(4), FMS107–FMS111.
71
Jiang, L., Wang, R., Li, X., Jiang, L., & Lu, G. (2005). Electrochemical oxidation behav-
ior of nitrite on a chitosan-carboxylated multiwall carbon nanotube modified
electrode. Electrochemistry Communications, 7(6), 597–601.
Karam, L., Jama, C., Mamede, A.-S., Fahs, A., Louarn, G., Dhulster, P., et al. (2013).
Study of nisin adsorption on plasma-treated polymer surfaces for setting up
materials with antibacterial properties. Reactive and Functional Polymers, 73(11),
1473–1479.
Karam, L., Jama, C., Nuns, N., Mamede, A. S., Dhulster, P., & Chihib, N. E. (2013). Nisin
adsorption on hydrophilic and hydrophobic surfaces: Evidence of its interactions
and antibacterial activity. Journal of Peptide Science, 19(6), 377–385.
Katas, H., & Alpar, H. O. (2006). Development and characterisation of chi-
tosan nanoparticles for siRNA delivery. Journal of Controlled Release, 115(2),
216–225.
Kumar, H. N., Prabhakar, M., Prasad, C. V., Rao, K. M., Reddy, T. A. K., Rao, K. C., et al.
(2010). Compatibility studies of chitosan/PVA blend in 2% aqueous acetic acid
solution at 30 ◦ C. Carbohydrate Polymers, 82(2), 251–255.
Li, R., Hu, P., Ren, X., Worley, S. D., & Huang, T. S. (2013). Antimicrobial N-halamine
modified chitosan films. Carbohydrate Polymers, 92(1), 534–539.
Limpan, N., Prodpran, T., Benjakul, S., & Prasarpran, S. (2010). Properties of
biodegradable blend films based on fish myofibrillar protein and polyvinyl alco-
hol as influenced by blend composition and pH level. Journal of Food Engineering,
100(1), 85–92.
Malley, I., Bardon, J., Rollet, M., Taverdet, J., & Vergnaud, J. (1987). Modelling of
controlled drug-release in case of carbopol-sodium salicylate matrix in gastric
liquid. Drug Development and Industrial Pharmacy, 13(1), 67–79.
Moditsi, M., Lazaridou, A., Moschakis, T., & Biliaderis, C. G. (2014). Modifying the
physical properties of dairy protein films for controlled release of antifungal
agents. Food Hydrocolloids, 39, 195–203.
Nguyen, V. T., Gidley, M. J., & Dykes, G. A. (2008). Potential of a nisin-containing
bacterial cellulose film to inhibit Listeria monocytogenes on processed meats.
Food Microbiology, 25(3), 471–478.
Ouattara, B., Simard, R., Piette, G., Begin, A., & Holley, R. (2000). Diffusion of acetic
and propionic acids from chitosan-based antimicrobial packaging films. Journal
of Food Science, 65(5), 768–773.
Ozdemir, M., & Floros, J. (2001). Analysis and modeling of potassium sorbate dif-
fusion through edible whey protein films. Journal of Food Engineering, 47(2),
149–155.
Park, S. I., Daeschel, M., & Zhao, Y. (2004). Functional properties of antimicrobial
lysozyme–chitosan composite films. Journal of Food Science, 69(8), M215–M221.
Peppas, N. (1985). Analysis of Fickian and non-Fickian drug release from polymers.
Pharmaceutica Acta Helvetiae, 60(4), 110.
Sajomsang, W., Ruktanonchai, U. R., Gonil, P., & Warin, C. (2010). Quaternization of N-
(3-pyridylmethyl) chitosan derivatives: Effects of the degree of quaternization,
molecular weight and ratio of N-methylpyridinium and N,N,N-trimethyl ammo-
nium moieties on bactericidal activity. Carbohydrate Polymers, 82(4), 1143–1152.
Sanjurjo, K., Flores, S., Gerschenson, L., & Jagus, R. (2006). Study of the perfor-
mance of nisin supported in edible films. Food Research International, 39(6),
749–754.
Selvam, N. C. S., Manikandan, A., Kennedy, L. J., & Vijaya, J. J. (2013). Comparative
investigation of zirconium oxide (ZrO2 ) nano and microstructures for structural,
optical and photocatalytic properties. Journal of Colloid and Interface Science,
389(1), 91–98.
Srinivasa, P. C., Ramesh, M. N., Kumar, K. R., & Tharanathan, R. N. (2003). Properties
and sorption studies of chitosan–polyvinyl alcohol blend films. Carbohydrate
Polymers, 53(4), 431–438.
Suppakul, P., Sonneveld, K., Bigger, S. W., & Miltz, J. (2011). Diffusion of linalool
and methylchavicol from polyethylene-based antimicrobial packaging films.
LWT—Food Science and Technology, 44(9), 1888–1893.
Tripathi, S., Mehrotra, G., & Dutta, P. (2009). Physicochemical and bioactivity of cross-
linked chitosan–PVA film for food packaging applications. International Journal
of Biological Macromolecules, 45(4), 372–376.
Wan Ngah, W. S., Kamari, A., & Koay, Y. J. (2004). Equilibrium and kinetics studies
of adsorption of copper(II) on chitosan and chitosan/PVA beads. International
Journal of Biological Macromolecules, 34(3), 155–161.
Wang, H., Ma, X., Li, Y., Jiang, S., Zhai, L., Jiang, S., et al. (2013). Synthesis, antimicrobial
and release of chloroamphenicol loaded poly(l-lactic acid)/ZrO2 nanofibrous
membranes. International Journal of Biological Macromolecules, 62, 494–499.
Wang, H., Zhang, P., Ma, X., Jiang, S., Huang, Y., Zhai, L., et al. (2014). Prepara-
tion, characterization of electrospun meso-hydroxylapatite nanofibers and their
sorptions on Co(II). J Hazardous Materials, 265, 158–165.
Xiao, D., Gömmel, C., Davidson, P. M., & Zhong, Q. (2011). Intrinsic Tween 20 improves
release and antilisterial properties of co-encapsulated nisin and thymol. Journal
of Agricultural and Food Chemistry, 59(17), 9572–9580.
Yang, J. M., Su, W. Y., & Yang, M. C. (2004). Evaluation of chitosan/PVA blended
hydrogel membranes. Journal of Membrane Science, 236(1), 39–51.
Yoshida, C. M., Bastos, C. E. N., & Franco, T. T. (2010). Modeling of potassium sor-
bate diffusion through chitosan films. LWT—Food Science and Technology, 43(4),
584–589.
Zhang, R., Xu, W., & Jiang, F. (2012). Fabrication and characterization of dense
Chitosan/polyvinyl-alcohol/poly-lactic-acid blend membranes. Fibers and Poly-
mers, 13(5), 571–575.
Zhong, Q., Jin, M., Davidson, P. M., & Zivanovic, S. (2009). Sustained release of
lysozyme from zein microcapsules produced by a supercritical anti-solvent pro-
cess. Food Chemistry, 115(2), 697–700.