Food Hydrocolloids 53 (2016) 75e83

Contents lists available at ScienceDirect

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

Physical properties of bacterial cellulose composites for wound

dressings
Wen-Shuo Chang, Hui-Huang Chen*
Department of Food Science, National Ilan University, 1 Sec. 1, Shen Nung Rd., I-Lan City, Taiwan, R.O.C

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

Article history: To test various formulas and techniques for manufacturing dry-fabricated bio-film (DFBF) that exhibits
Received 6 August 2014 physical properties advantageous to the use of the DFBF in wound dressings, a DFBF was fabricated by
Received in revised form adding chitosan (Chi) and alginate (Alg) to homogenized bacterial cellulose (BC) obtained from vinegar
2 December 2014
pellicles in vinegar brewing byproducts in this study. The results revealed that the degree of oxidation in
Accepted 8 December 2014
DFBF manufactured using hydrogen peroxide oxidized BC (HOBC), with 0.092% carboxyl group content,
Available online 16 December 2014
was lower than that in DFBF manufactured using periodic acid oxidized BC (POBC), but DFBF made using
HOBC exhibited higher elongation, rehydration, swelling ratios, and water vapor transmission than that
Keywords:
Bacterial cellulose
fabricated using POBC. A DFBF composite gel with 98.5% water content possessed appropriate fluidity for
Chitosan molding. After 10 min of rinsing cross-linked HOBC, 72 ppm of calcium remained in the final DFBF, which
Composite not only prevented cell toxicity but also demonstrated desirable mechanical and rehydration properties.
Physical property Overall, the modified DFBF possessed a high rehydration ratio of 51.69% and could absorb and gradually
Wound dressing release naringin by up to 80% within 24 h. This modified DFBF has the potential for exudate absorption
and the controlled release of medicinal substances at the initial stage of healing when used in wound
dressings.
© 2014 Elsevier Ltd. All rights reserved.

1. Introduction
Artificial materials with biomimetic behavior are essential in the
biomedical field. Scaffolds based on nanofibers (NFs) mimic the
natural extracellular matrix and its nanoscale fibrous structure.
Several approaches have been described for developing materials
based on NFs from synthetic or natural polymers (Ashammakhi,
Ndreu, Yang, Ylikauppila, & Nikkola, 2007). Bacterial cellulose
(BC), secreted by Gluconacetobacter xylinus, has been presented as a
biocompatible scaffold with a nanoscale fibrous structure for the
engineering of cartilage and blood vessels, wound dressings, and
guided tissue regeneration among other applications
(Chiaoprakobkij, Sanchavanakit, Subbalekha, & Pavasant, 2011;
Fernandes et al., 2010; Wiegand, Elsner, Hipler, & Klemm, 2006;
Wu et al., 2014).
BC has attracted increasing attention as a wound-dressing ma-
terial, because of its high purity, hydrophilicity, structure-forming
potential, chirality, and biocompatibility (Kawecki, Krystynowicz,
* Corresponding author. Tel.: þ886 3 9317764; fax: þ886 3 9310722.
E-mail address: hhchen@niu.edu.tw (H.-H. Chen).
Wysota, Czaja, Sakiel, 2006; Keshk, 2014). Because of its fine
network structure, it also exhibits controlled-release functionality,
a critical skin-barrier function in wound healing (Czaja,
Krystynowicz, Bielecki, & Brown, 2006).
Depending on synthetic conditions, BC has a water-holding ca-
pacity ranging from 60 to 700 times its dry weight. The cellulose in
typical statically cultured pellicles is approximately 1% of the total
weight, with the rest being water (Okiyama, Shirae, Kano, &
Yamanaka, 1992; Yamanaka et al., 1989). One reason for this hy-
drophilicity is that the cellulose ribbons are assembled extracellu-
larly in the liquid medium; abundant micelles are then formed that
trap large quantities of liquid. In addition, the hydrophilicity of the
cellulose pellicle is partially a result of the extensive interior surface
area of the interstitial space of the undried pellicle. However, BC
exhibits poor rehydration after drying, because of high crystallinity
(Huang, Chen, Lin, Hsu, & Chen, 2010). This limits its use as a bio-
film. In this study, two hydrophilic materials, chitosan and algi-
nate, were added to a dehydrated BC composite to increase its
rehydration ability.
Investigation of the use of BC in a liquid-loaded pad for wound
care began in the early 1980s and led to some BC products in the
wound-care market. For example, a purified gelatinous BC

http://dx.doi.org/10.1016/j.foodhyd.2014.12.009
0268-005X/© 2014 Elsevier Ltd. All rights reserved.
76 W.-S. Chang, H.-H. Chen / Food Hydrocolloids 53 (2016) 75e83
membrane was commercialized as an artificial skin, or wet wound
dressing (Fontana, Souza, Fontana, Toriani, & Moreschi, 1990). A
high mechanical strength in its wet state, substantial permeability
for liquids and gases, and low skin irritation indicated that the BC
gelatinous membrane was superior to conventional gauze as an
artificial skin for temporary covering of wounds (Fontana, Franco,
Souza, Lyra, & Souza, 1991). BC composites fabricated by blending
chitosan, polyethylene glycol (PEG), and gelatin were developed for
a possible tissue engineering scaffold and wound-dressing material
(Ougiya, Watanabe, Morinaga, & Yoshinaga, 1997). The BC com-
posites were prepared by immersing a wet BC pellicle in chitosan,
PEG, or a gelatin solution followed by freeze-drying.
An ideal wound-dressing material should maintain a moist
environment at the wound interface, allow gaseous exchange, act
as a barrier to microorganisms, and remove excess exudates. It
should also be nontoxic, non-allergenic, and non-adherent in the
sense that it can be easily removed without trauma; and it should
be made from a readily available biomaterial that requires minimal
processing, possesses antimicrobial properties, and promotes
wound healing (Shuangyun, Wenjuan, & Gu, 2008). Chitosan and
alginate were the best materials to control the microenvironment
on the wound surface including certain bioactive products that
stimulate some part of the healing cascade (Paul & Sharma, 2004;
Sudheesh Kumar et al., 2010; Yang, Yang, Lin, Wu, & Chen, 2008).
Furthermore, oxidized celluloses containing carboxylic groups
represent a new class of biodegradable materials, and have been
accepted for use in humans to stop bleeding during surgery and to
prevent the formation and reformation of postsurgical adhesions
(Bowman & Cooke, 1994).
However, the BC used in the composites or wound dressings in
the aforementioned studies were prepared using cultured pellicles.
The high production costs and low efficiency in yield limit the use
of this BC to high-value-added applications (Chang, Chen, Lin, &
Chen, 2012; Chawla, Bajaj, Survase, & Singhal, 2009; Hungund
et al., 2013). In the present study, homogenized BC was economi-
cally gathered from the film wastes of a vinegar factory for use as
the raw material. It could also be obtained from the offal from
“nata” or “biocellulose facial mask” industries. The end-product, a
DFBF, was composed of the homogenized BC, chitosan (Chi) and
alginate (Alg). To investigate the usefulness of this DFBF for wound
dressings, various fabrication techniques (oxidation treatment,
water content adjustment, and rinsing time) were adapted with
regard to specific physical properties (tensile properties, hydro-
philicity, and water vapor transmission) and biocompatibility.
2. Materials and methods
2.1. DFBF preparation
The BC used to prepare the DFBF was obtained from solid
byproducts of vinegar brewing (Sunny Way Biotech Co., Ltd., I-Lan,
Taiwan). Based on Huang et al. (2010), the vinegar film was
immersed in 0.5 N NaOH for 10 min and then in 0.5 N NaOH for 24 h
at room temperature to purify it, then rinsed with deionized water
to achieve a neutral pH. The BC was homogenized using a seven-
speed blender (Waring Commercial, Torrington, CT, USA) at
3500 rpm for 3 min and oxidized by hydrogen peroxide (H2O2,
Nippon Chemical Industrial Co., Ltd., Tokyo, Japan) or periodic acid
(H5IO6, Mallinckrodt Chemicals, St. Louis, MO, USA) for 6 h to obtain
hydrogen-peroxide-oxidized BC (HOBC) or periodic-acid-oxidized
BC (POBC), respectively. According to the optimized conditions of
fabricated composites film with BC from Yang (2011), the BC or
oxidized BC (OBC) was blended with sodium alginate (Alg, Food
Chemifa, Tokyo, Japan) and chitosan (Chi, DD 83%, 297 kDa, Ohka
Enterprises Co., Ltd., Kaohsiung, Taiwan) at a ratio of
BC:Alg:Chi ¼ 2:1:1 in 98.0%, 98.5% or 99.0% water content to form a
basic DFBF gel solution. We molded the gel solution, cross-linked it
with 3% (w/w) calcium chloride for 2 h in a 5-mm-thick wet-
fabricated pellicle (WFP), and rinsed it with phosphate-buffered
saline (PBS, pH ¼ 7.2). Finally, the rinsed fabricated film was
freeze-dried, to obtain the desirable rehydration ability, and the
resultant DFBF was stored in a dry cabinet.
2.2. COOH content analysis
According to Kumar and Yang (1999), about 0.5 g of the dried
sample was accurately weighed and dispersed in 50 ml of a 2%
weight by weight solution of calcium acetate for 30 min. The sus-
pension was then titrated with standardized 0.1 N NaOH solution
using phenolphthalein as an indicator. The volume of NaOH solu-
tion consumed was corrected for the blank. The carboxylic content
in the sample was calculated from the following relationship:
N$V$MWCOOH
COOH content ¼  100%
Weight of the sample ðmgÞ
where MWCOOH is the molecular weight of COOH, N is the normality
of NaOH, and V is the volume of NaOH in mls consumed in titration,
after correcting for the blank.
2.3. Fourier Transform Infrared (FT-IR) Spectroscopy
The FT-IR spectra of freeze-dried samples were recorded on a
Paragon 500 (Perkin Elmer, Waltham, USA) in absorption mode in
the range of 4000e450 cm1. Thirty-two scans were performed to
establish accuracy.
2.4. Rheological properties
The rheological properties of HOBC/Chi/Alg gel solution were
determined in a dynamic rheometer (Rheometer AR-550, TA In-
struments, New Castle, Delaware, U.S.A.) using a small amplitude
oscillatory test. The dynamic rheometer was equipped with paral-
lel-plate geometry of 40 mm diameter. Gap and strain were set at
2.0 mm and 1.0%, respectively. Steady shear tests were performed
over a shear rate range of 0.065e65 1/s and conducted between
0.01 and 10 Hz at 25  C in frequency sweeps. Three replicate scans
were conducted, with storage modulus (G0 ), loss modulus (G00 ), tand
and magnitude of the complex viscosity (h*) recorded.
2.5. Mechanical properties of DFBF
The mechanical properties of rehydrated DFBF were determined
since the rehydration would weaken the toughness of wound
dressing. The tensile strength and elongation at break of rehydrated
DFBF were measured by TPA (TA.XT2, Stable Micro Systems, God-
alming, Surrey, England), equipped with a 5-kg load cell and a cy-
lindrical probe (P/36R, 35 mm diameter). The dumbbell-shaped
sample was placed on an ASPR probe, the spacing and extension
speed were 50 mm and 1.0 mm/s, respectively, to determine the
tensile properties. Five replicates were performed and calculated as:
Tensile strength ðkPaÞ ¼ F=A
Elongation at break ð%Þ ¼ 100  DL=L
where F and A represent the maximum force (N) to break
composite and area (m2) of sample, respectively. DL and L represent
the elongation (mm) and original length (50 mm) of sample,
respectively.
 W.-S. Chang, H.-H. Chen / Food Hydrocolloids 53 (2016) 75e83
2.6. Hydrophilic properties
Following Bodhibukkana et al. (2006), the wet samples were
drained at 25  C for 2 min and weighed (Wwet). The freeze-dried
samples were weighed (Wdry) and immersed in deionized water
(W/V ¼ 1:2) until the weight of the rehydrated sample (Wrwet) no
longer increased (around 7 h). The Swelling ratio and rehydration
ratio, representing the degree to which removed water was
replaced by water, were calculated as:
Swelling ratio ð%Þ ¼ ðWrwet  WdryÞ=Wdry  100%
Rehydration ratio ð%Þ ¼ ðWrwet  WdryÞ=ðWwet  WdryÞ
 100%
2.7. Water vapor transmission rate (WVTR)
Following a modified ASTM E96-90 (ASTM, 2002) method, a
2.5 cm diameter DFBF specimen was mounted on brown bottle
(outer diameter of 2.6 cm, inner diameter of 1.4 cm, 5.9 cm deep)
containing 10 mL distilled water. The cap was sealed tightly on each
bottle with an O-ring. The bottles were weighed and placed in a
controlled chamber at 40% RH (achieved by adding a large amount
of MgCl2∙6H2O) for 24 h and weighed again. The controlled
chamber was placed in an incubator at a constant temperature of
37  C and humidity of 40% RH. Five specimens were tested for each
DFBF and the WVTR was calculated using.
  ðDw=DtÞ
WVTR g=m2 $day ¼
A
where Dw/Dt is the weight of moisture loss for 24 h (g/day) and A is
the film area exposed to the moisture transfer.
2.8. Cell viability assay
Detroit 551 cells (human embryonic skin fibroblast, BCRC 60118)
were purchased from the Bioresource Collection and Research
Center (FIRDI). The cell viability assay was performed according to
Gerlier and Thomasset (1986) with minor modifications. Detroit
551 cells were plated on 96-well plates at a density of 103 cells/
well. After 24 h incubation, 20 mL of serial dilutions of the sample
was added to the culture medium and the cells were incubated for
24 h. To each well, 100 mL of sterile filtered MTT reagent (5 mg/mL)
was added and reacted with the cells for 4 h at 37  C. Culture so-
lutions were removed and replaced with 100 mL/well DMSO to
dissolve formazan crystals. The absorbance was measured at
450 nm using an ELISA reader (E11010, BioLog, California, USA), and
the relative cell viability (%) was calculated as:
Cell viability ð%Þ ¼ ½A570 ðsampleÞ  A570 ðblankÞ=A570 ðcontrolÞ
 A570 ðblankÞ  100%
2.9. Calcium content analysis
According to CNS9638 N5200 (1982), the rinsed WFP was cut
into 2  2 cm samples and immersed in 10 mL of deionized water or
10 mL of PBS (pH ¼ 7.2) for 30 min. Five milliliters of deionized
water, 1 mL of a NH4Cl/NH4OH buffer solution (pH ¼ 10.0), and two
drops of eriochrome black T indicator (Koch-Light Laboratories,
Buckingham, England) were added in sequence to a 5-mL immer-
sion solution. Titration was conducted using an EDTA standard
77
solution (Nippon Chemical Industrial Co., Ltd.) from purple to blue.
The calcium concentration was calculated using.
MEDTA $VEDTA $40:0
CaðppmÞ ¼

Sample mL $103
where MEDTA and VEDTA are the molarity and titration volume (mL)
of EDTA, respectively, demarcated using a CaCO3 standard solution.
2.10. SEM observation
DFBF was lacerated into small pieces after freeze-drying. Dried
specimens were mounted on aluminum studs and coated with a
gold/palladium alloy under high vacuum conditions. Specimens
were examined using a Tescan Vega TS5136MM microscope (Tes-
can s.r.o., Brno, Czech Republic) to observe the microstructure of
ruptured surfaces.
2.11. Release of naringin from DFBF
For determining the release of naringin (SigmaeAldrich) from
BC composites, DFBF was cut into 2 cm square. Equal amount (2 mL)
of naringin solution (250 mg/L) was absorbed in the DFBF and then
being freeze dried. The naringin contained DFBF was immersed in
5 mL PBS (pH 7.2) (SigmaeAldrich) for release test conducted under
ambient temperature.
The simple exponential relation Mt/M∞ ¼ ktn was introduced to
describe the general solute release behavior of the controlling
release polymeric devices, where
Mt: the amount of compound released at time t,
M∞: the amount of compound released as time approaches
infinity,
t: the release time,
k: the kinetic constant and
n: the diffusional exponent.
For a thin film, Fickian diffusion (Case I) is characterized when
n ¼ 0.5, and Non-Fickian diffusion or anomalous diffusion is char-
acterized when 0.5 < n < 1.0. The diffusion is too fast compared
with matrix relaxation (Case II) when n ¼ 1. Super Case II transport
is characterized when n > 1.0 (Ritger & Peppas, 1987; Costa & Sousa
Lobo, 2001; Da Silva, Bierhalz, & Kieckbusc, 2012).
2.12. Statistical analysis
Statistical analysis of data was performed using an SPSS system.
When analysis of variance (ANOVA) revealed a significant effect
(P < 0.05), data means were compared using a least significant dif-
ference (LSD) test.
3. Results and discussion
3.1. Physical properties of DFBF prepared using OBC
The main peaks, characteristic frequency (4000 cm1e1300 cm1),
and fingerprint regions (below 1300 cm1) of HOBC, POBC, and BC
were nearly the same in FTIR spectra (Fig. 1). Only the widths of peaks
1740 cm1 and 880 cm1 were in the order of POBC > HOBC > BC.
Because the C]O peak appeared at 1740 cm1 (Kim, Kuga, Wada,
Okano, & Kondo, 2000), the eCOOH contents were further quanti-
fied. The eCOOH contents of POBC and HOBC were 0.092% and 0.39%,
respectively (Table 1). Because both hydrogen peroxide and periodic
acid demonstrate a high oxidative ability, the eOH of C-6 in glucose
were oxidized to eCOOH during the oxidation treatment of
78 W.-S. Chang, H.-H. Chen / Food Hydrocolloids 53 (2016) 75e83

Fig. 1. FTIR spectra of BC (A), HOBC (B), POBC (C), Chitosan (D), Alginate (E), BC/Chi/Alg (F), HOBC/Chi/Alg (G) and POBC/Chi/Alg (H).

homogenized BC (Fig. 1BeC). Furthermore, the periodic acid cleaved
the glucoside bond in C-2 and C-3, producing two eHO groups. This
cleavage process led the absorption peaks of the hemiacetal bond
formed by eHO and neighboring eOH; these peaks appeared at
approximately 880 cm1 (Li, Wan, Li, Liang, & Wang, 2009). The
cleavage of the glucoside bond to form eHO was slowed by the
oxidation of hydrogen peroxide (Menachen & Aharon, 1969). How-
ever, the peak at 1740 cm1 disappeared when OBC was combined
with chitosan and alginate (Fig. 1FeH). Concurrently, the peaks of
chitosan at 1645 cm1 (corresponding to C]O stretching in amide I)

Table 1
Hydrophobic properties and COOH contents of BC, HOBC, POBC, HOBC/Alg, DFBFand the mechanical properties of rehydrated DFBF.

Films Mechanical propertiesa Hydrophilic propertiesa COOH content (%)

b c
TS (kPa) E (%) RR (%)d SR(%)e

BC 15.99 ± 3.70D 202.99 ± 21.78C UMf

HOBC 0.09 ± 0.01
POBC 0.39 ± 0.05
HOBC/Alg 6.99 ± 0.42E 155.30 ± 11.61D UM
BC/Chi/Alg 208.23 ± 4.08BC 24.62 ± 1.32B 47.14 ± 1.26A 830.22 ± 44.22A UM
HOBC/Chi/Alg 223.48 ± 14.35B 27.48 ± 2.87A 45.04 ± 0.39B 748.59 ± 25.86AB UM
POBC/Chi/Alg 265.08 ± 4.94A 12.70 ± 0.13C 42.18 ± 1.51C 659.79 ± 18.75B UM
a
The different letters, AeЕ, for sample in each column represents significant difference (P < 0.05, n ¼ 5).
b
TS: Tensile strength.
c
E: Elongation at break.
d
RR: Rehydration ratio.
e
SR: Swelling ratio.
f
UM: unmeasurable.
 W.-S. Chang, H.-H. Chen / Food Hydrocolloids 53 (2016) 75e83
and 1584 cm1 (corresponding to NeH bending in HNeCOCH3) as
well as the peak of alginate at 1614 cm1 (corresponding to asym-
metric eCOOH) merged and migrated to 1635 cm1, corresponding to
the superposition of the bands assigned to the carboxylate group of
alginate and the amine group of chitosan (Venkatesan, Bhatnagar, &
Kim, 2014). In general, the electrostatic interaction between the
carboxylate group of alginate and the amine group of chitosan forms a
polyelectrolyte complex. Furthermore, the interaction between the
carboxylate group of OBC and the amine group of chitosan might also
form a polyelectrolyte complex to enhance the mechanical strength of
the film (Li et al., 2009).
The maximum tensile strength and elongation at break were
obtained for the rehydrated DFBF prepared using POBC and HOBC,
respectively (Table 1). The tensile strength of rehydrated DFBF
positively correlated to its eCOOH content, which also resulted in
the greater amount of formed polyelectrolyte complex and hence
the stronger DFBF scaffold. The alginate and chitosan molecules
were filled in the voids and formed a large number of junction
zones to construct a continuous phase.
A porous and layered structure was observed in SEM images of
the DFBF cross-section (Fig. 2AeC). During the cross-linking treat-
ment, the surface of the DFBF gel solution made contact with CaCl2,
and the hardened surface became a barrier against the CaCl2. The
difference in the speed of cross-linking at the surface and in the
internal part of the gel solution caused this asymmetric structure of
the DFBF. The outer layer was a dense skin and the inner layer was
sponge-like and porous (Fig. 2). This asymmetric structure is
similar to a wound dressing prepared using BC/Alg composites, the
outer dense layer of which prevents the entrance of microorgan-
isms and water evaporation, and the inner layer of which, with
100e500 mm pores, offers the absorption of wound fluids and the
transmission of oxygen and water vapor (Chiaoprakobkij et al.,
2011). The SEM image also showed that fibrous BC remained in
the BC/Chi/Alg DFBF (Fig. 2a) and a network structure existed in the
Fig. 2. SEM images of various DFBF cross-section: BC/Chi/Alg (A) (a), HOBC/Chi/Alg (B) (b) and POBC/Chi/Alg film (C) (c) (A, B, C are 100 magnification; a, b, c are 3000
magnification).
79
HOBC/Chi/Alg DFBF (Fig. 2b). In the POBC/Chi/Alg DFBF, the fibrous
BC and the network structure were difficult to discern although the
layer matrices were formed (Fig. 2c). The various structures rec-
ommended the high tensile strength of POBC/Chi/Alg and high
elongation at break of HOBC/Chi/Alg.
The hydrophilicity of the DFBF depended on the addition of
chitosan, because the films without chitosan, namely the BC, HOBC,
POBC, and HOBC/Alg films, exhibited low rehydration ratios
(Table 1). In an as-prepared BC containing more than 90% water,
however, water is easily expelled, with no further recovery in
swelling properties because of the formation of crystalline regions
by H-bonds between cellulose fibers (Nakayama et al., 2004). Chi-
tosan exhibited high hydrophilicity and decreased the crystallinity
of BC in BC/Chi film (Phisalaphong & Jatupaiboon, 2008). Therefore,
the high rehydration ratio of chitosan-added DFBF recommends its
application in wound dressings to absorb wound fluids or blood.
The DFBF prepared using HOBC possessed the highest elonga-
tion at break and swelling ratio, and the best rehydration ratio. The
following experiments were carried out by using HOBC/Chi/Alg to
evaluate DFBF for wound dressings.
3.2. Physical properties of DFBF prepared using various water-
content WFPs
As the water content of the DFBF gel solution was increased, the
moduli G0 , G00 , and h* decreased and the tand increased (Fig. 3). This
indicated the flow was improved at high water content and showed
the frequency dependence. The pseudogel possesses a properties of
the moduli and h* are frequency-dependent (Morris & Ross-
Murphy, 1981). The DFBF gel solution was then speculated to
exhibit a shear thinning property, which might facilitate molding a
gel solution because the solution exhibited desirable mobility
during agitation and became stable when agitation was stopped.
80 W.-S. Chang, H.-H. Chen / Food Hydrocolloids 53 (2016) 75e83

Fig. 3. Rheological properties of HOBC/Chi/Alg gel solution under various moisture contents.

Table 2
Mechanical properties of rehydrated DFBF and hydrophilic properties of DFBF prepared with HOBC/Chi/Algunder various water contents in DFBF gel solution.

Water content (%) Mechanical propertiesa Hydrophilic propertiesa WVTRc (g/m2$day)

b b b b
TS (kPa) E (%) RR (%) SR(%)

98.0 378.51 ± 8.32A 21.34 ± 1.14A 38.54 ± 0.31C 613.30 ± 12.23C 2865.03 ± 93.30C
98.5 276.96 ± 5.68B 18.12 ± 0.21B 45.89 ± 1.32B 830.00 ± 14.22B 3034.38 ± 77.86B
99.0 72.82 ± 1.67C 12.25 ± 1.00C 50.99 ± 1.20A 1240.00 ± 3.74A 3465.00 ± 60.00A
a
The different letters, AeC, for sample in each column represents significant difference (P < 0.05, n ¼ 5).
b
The abbreviation of physical properties are same as in Table 1.
c
WVTR: Water vapor transmission rate.

Overall, the DFBF gel solution's ease and uniformity of molding prepared using gel solutions of 98.0% and 98.5% water content were
were in the order of 99.0% > 98.5% > 98.0%. 2865 and 3034 g/m2∙d, respectively, and were close to the ideal
The tensile strength and elongation at break of rehydrated DFBF dressing. However, the lower fluidity of a DFBF gel solution with
decreased as the water content of the DFBF gel solution increased. 98.0% water content is not conducive to molding. The following
Concurrently, the rehydration and swelling ratio increased as the experiments were carried out by using a gel solution of 98.5% water
water content of the WFP increased (Table 2). The alginate cross- content to evaluate DFBF for wound dressings. This solution
linked with Caþ2 could form a strong three-dimensional gel and exhibited the most desirable mechanical properties, hydrophilicity,
appeared to be the main contributor to the mechanical strength of and WVTR.
the DFBF. The increased water content caused the concentration and
molecular density to decrease. The distant molecules weaken the gel
strength of the WFP after cross-linking. However, for a DFBF gel so-
lution with higher water content, the freeze-drying process left more
pores in DFBF to adsorb water, resulting in higher hydrophilicity.
The WVTR of the DFBF increased as the water content of the DFBF
gel solution increased (Table 2). The vapor transmission was regar-
ded as a small-molecule prevaporation. The vapor molecules were
absorbed when they made contact with the DFBF surface and the
porous DFBF allowed vapor molecules to diffuse through the film.
Finally, desorption and evaporation of vapor molecules occurred on
the other side of the DFBF surface. The DFBF prepared using 99%
water content gel solution possessed a loose structure, and its
porosity facilitated the diffusion and desorption of water molecules.
The WVTR is a quality index for wound dressings. An excessively
high WVTR accelerates the dehydration and scabbing of a wound,
whereas an excessively low WVTR causes wound fluids to accu-
mulate, and impedes healing, and raises the risk of bacterial
contamination. A desirable WVTR is 2500e3000 g/m2∙d (Li, Li, Ma, Fig. 4. The effect of rising time of rehydrated HOBC/Chi/Alg film on residual calcium in
Wang, & Zhang, 2014; Paul & Sharma, 2004). The WVTR of DFBF WFP.
 W.-S. Chang, H.-H. Chen / Food Hydrocolloids 53 (2016) 75e83 81

3.3. Physical properties and biocompatibility of DFBF after various

rinsing durations

When the chitosan and alginate were added to the OBC during
the preparation of the DFBF, the eCOO of the alginate and the OBC
interacted with the eNHþ 3 of the chitosan, and the cross-linking was
conducted because of the ion attraction between residual eCOO in
the alginate and Caþ2 (Meng et al., 2010). Although BC has been
shown not to be cytotoxic or genotoxic (Moreira et al., 2009), the
high concentration of Caþ2 could connect with the negatively
charged cell membrane, thereby inhibiting the transport of nutri-
ents and impeding cell growth (Maeno et al., 2005). Therefore, the
WFP was rinsed with PBS before dehydration. A 2 min rinse halved
the Caþ2 concentration of the WFP, and a 10 min rinse reduced the
residual Caþ2e72 ppm, after which it stabilized (Fig. 4). In a SEM
image, a quantity of 2e5 mm of crystalline spherical particles were
Fig. 5. The cell viability of Detroit 551 cell cultured with immersing solution of DFBF observed in a non-rinsed WFP (Fig. 4A), and a pronounced reduction
(HOBC/Chi/Alg), prepared by different rinsing time, on polystylene culture plate. The of particles was seen after 10 min of rinsing (Fig. 4B).
different letters, aec, for each cell viability represents a significant difference
Rinsing durations were evaluated for the biocompatibility of the
(P < 0.05).
DFBF. The cell viability was 54.49% and became noncytotoxic to
Detroit 551 cells after 10 min of rinsing (Fig. 5). Ten minutes was
considered sufficiently long to achieve biocompatibility in applying
Table 3 DFBF in wound, because of the sufficient elimination of Caþ2.
Mechanical properties of rehydrated DFBF and hydrophilic properties of DFBF pre- Only a small decrease in the tensile strength of rehydrated DFBF
pared with WFP (HOBC/Chi/Alg) under various rising time. was observed when the rinsing time of the WFP exceeded 30 min,
Rinsing Mechanical propertiesa Hydrophilic propertiesa whereas the elongation at break decreased in conjunction with the
time (min)
TS (kPa)b E (%)b RR (%)b SR(%)b
rinsing duration. The highest rehydration and swelling ratios were
A A BC
observed after 10 min of rinsing (Table 3). Rinsing the WFP caused
0 217.12 ± 9.58 15.44 ± 1.23 46.21 ± 1.21 1350.35 ± 39.58C
parts of hydrophilic molecules such as chitosan and uncross-linked
2 222.01 ± 2.12A 12.67 ± 0.50B 52.66 ± 4.43A 1283.89 ± 5.75C
5 223.49 ± 5.98A 13.07 ± 1.35B 51.85 ± 3.96AB 1672.77 ± 80.22B alginate to be washed off, thus weakening the mechanical prop-
10 212.67 ± 6.81AB 12.45 ± 0.44B 53.43 ± 1.74A 2046.11 ± 153.10A erties of the DFBF. Furthermore, its hydrophilicity is attributed to
30 200.72 ± 10.60B 9.72 ± 0.05C 48.15 ± 2.35B 1798.74 ± 66.97B the hydrophilic components and porous structure. The porosity
60 208.94 ± 13.64AB 8.92 ± 0.55C 45.43 ± 1.99C 1821.52 ± 49.27AB
increased as the rinsing time of the WFP increased, which contra-
a
The different letters, AeC, for sample in each column represents significant dicted the hydrophilic molecule content in DFBF. Ten min was
difference (P < 0.05, n ¼ 5). regarded as the optimal rinsing time, yielding the highest hydro-
b
The abbreviation of physical properties are same as in Table 1.
philicity. The following experiments were conducted by rinsing the
WFP for 10 min to evaluate the potential for the release of antiin-
flammatory substances in DFBF.

Fig. 6. Release ratio curve of naringin from HOBC/Chi/Alg films.

82 W.-S. Chang, H.-H. Chen / Food Hydrocolloids 53 (2016) 75e83
3.4. Release of naringin from DFBF
When the DFBF was immersed in PBS for 2 h, approximately half
of its naringin was released. The release ratio reached 80% at 24 h
immersion and then stabilized (Fig. 6). According to the simple
exponential relation Mt/M∞ ¼ ktn, the release exponent (n) and
kinetic constant (k) were 0.1609 and 0.5713, respectively. This
indicated that the release of naringin in DFBF contributed to a Fick's
diffusion model because n < 0.45 (Da Silva et al., 2012). The release
mechanism of Fick's diffusion posits that the chemicals in the
matrix of a porous macromolecule matrix are released in the im-
mersion solution by diffusion through these pores (Gurny, Doelker,
& Peppas, 1982). Although naringin is water-insoluble, swelled
chitosan expanded the pore size of chitosanealginate composites
(Ho, Mi, Sung, & Kuo, 2009) and facilitated the diffusion of naringin.
The successful release of naringin suggests that using DFBF in
wound dressings is effective for the controlled release of substances
that reduce infection and inflammation.
4. Conclusion
The homogenized BC oxidized by H2O2 increased the mechan-
ical strength and hydrophilicity of DFBF. Rinsing eliminated H2O2,
ensuring that the DFBF is eligible for Detroit 551 cell proliferation.
The modified DFBF exhibited desirable mechanical properties and
hydrophilicity, WVTR, biocompatibility, and the ability to control
the release of anti-inflammatory substances. These properties
make DFBF a suitable candidate for use in wound dressings
designed for the initial stages of healing, during which hemostasis,
absorption of wound fluids, and the application of anti-
inflammatory medications are particularly beneficial.
Acknowledgments
The authors thank the National Science Council (NSC 100-2313-
B-197-003) for their financial support of this study. The authors
also express appreciation to partners in the Nano-Biomaterial
Application Lab (NBA) for their assistance with the experiments.
References
Ashammakhi, N., Ndreu, A., Yang, Y., Ylikauppila, H., & Nikkola, L. (2007). Nanofiber
based scaffolds for tissue engineering. European Journal of Plastic Surgery, 35,
135e149.
ASTM. (2002). Standard test methods for tensile properties thin plastic sheeting (pp.
882e902). American Society for Testing and Materials D.
Bodhibukkana, C., Srichana, T., Kaewnopparat, S., Tangthong, N., Bouking, P.,
Martin, G. P., et al. (2006). Composite membrane of bacterially-derived cellulose
and molecularly imprinted polymer for use as a transdermal enantioselective
controlled-release system of racemic propranolol. Journal of Controlled Release,
113, 43e56.
Bowman, M. C., & Cooke, I. D. (1994). The efficacy of synthetic adhesion barriers in
infertility surgery. British Journal of Obstetrics and Gynaecology, 101, 3e6.
Chang, S. T., Chen, L. C., Lin, S. B., & Chen, H. H. (2012). Nano-biomaterials appli-
cation: morphology and physical properties of bacterial cellulose/gelatin com-
posites via crosslinking. Food Hydrocolloids, 27, 137e144.
Chawla, P. R., Bajaj, I. B., Survase, S. A., & Singhal, R. S. (2009). Microbial cellulose:
fermentative production and applications. Food Technology and Biotechnology,
47, 107e124.
Chiaoprakobkij, N., Sanchavanakit, N., Subbalekha, K., & Pavasant, P. (2011). Char-
acterization and biocompatibility of bacterial cellulose/alginate composite
sponges with human keratinocytes and gingival fibroblasts. Carbohydrate
Polymers, 85, 548e553.
CNS9638-N5200. (1982). Method of test for calcium and phosphorus in food stuff.
In Chinese National Standard. Taipei: Bureau of Standards, Metrology & In-
spection, M.O.E.A., Taiwan.
Costa, P., & Sousa Lobo, J. M. (2001). Modeling and comparison of dissolution
profiles. European Journal of Pharmaceutical Sciences, 13, 123e133.
Czaja, W., Krystynowicz, A., Bielecki, S., & Brown, R. M. (2006). Microbial cellulose-
the natural power to heal wounds. Biomaterials, 27, 145e151.
Da Silva, M. A., Bierhalz, A. C. K., & Kieckbusc, T. G. (2012). Modelling natamycin
release from alginate/chitosan active films. International Journal of Food Science
and Technology, 47, 740e746.
Fernandes, S. C. M., Freire, C. S. R., Silvestre, A. J. D., Neto, C. P., Gandini, A.,
Berglund, L. A., et al. (2010). Transparent chitosan films reinforced with a high
content of 317 nanofibrillated cellulose. Carbohydrate Polymers, 81, 394e401.
Fontana, J. D., Franco, V. C., Souza, S. J., Lyra, I. N., & Souza, A. M. (1991). Nature of
plant stimulators in the production of Acetobacter xylinum (“Tea Fungus”) bio-
film used in skin therapy. Applied Biochemistry and Biotechnology, 28-29,
341e351.
Fontana, J. D., Souza, A. M., Fontana, C. K., Toriani, I. L., & Moreschi, J. C. (1990).
Acetobacter cellulose pellicle as a temporary skin substitute. Applied Biochem-
istry and Biotechnology, 24e25, 253e264.
Gerlier, D., & Thomasset, N. (1986). Use of MTT colorimetric assay to measure cell
activation. Journal of Immunological Methods, 94, 57e63.
Gurny, R., Doelker, E., & Peppas, N. A. (1982). Modeling of drug sustained release of
water soluble drugs from porous, hydrophobic polymers. Biomaterials, 3,
27e32.
Ho, Y. C., Mi, F. L., Sung, H. W., & Kuo, P. L. (2009). Heparin-functionalized chitosan-
alginate scaffolds for controlled release of growth factor. International Journal of
Pharmaceutics, 376, 69e75.
Huang, H. C., Chen, L. C., Lin, S. B., Hsu, C. P., & Chen, H. H. (2010). In situ modification
of bacterial cellulose network structure by adding interfering substances during
fermentation. Bioresource Technology, 101, 6084e6091.
Hungund, B. S., Prabhu, S., Shetty, C., Acharya, S., Prabhu, V., & Gupta, S. G. (2013).
Production of bacterial cellulose from Gluconacetobacter persimmonis GH-2
using dual and cheaper carbon sources. Microbial & Biochemical Technology, 5,
31e33.
Kawecki, M., Krystynowicz, A., Wysota, K., Czaja, W., & Sakiel, S. (2006). Microbial
cellulosedthe natural power to heal wounds. Biomaterials, 27, 145e151.
Keshk, S. M. A. S. (2014). Bacterial cellulose production and its industrial applica-
tions. Journal of Bioprocessing & Biotechniques, 4, 150.
Kim, U. J., Kuga, S., Wada, M., Okano, T., & Kondo, T. (2000). Periodate oxidation of
crystalline cellulose. Biomacromolecules, 1, 488e492.
Kumar, V., & Yang, T. (1999). Analysis of carboxyl content in oxidized celluloses by
solid-state 13C CP: MAS NMR spectroscopy. International Journal of Pharma-
ceutics, 184, 219e226.
Li, X., Li, B., Ma, J., Wang, X., & Zhang, S. (2014). Development of a silk fibroin/HTCC/
PVA sponge for chronic wound dressing. Journal of Bioactive and Compatible
Polymers, 29, 398e411.
Li, J., Wan, Y., Li, Y., Liang, H., & Wang, J. (2009). Preparation and characterization of
2,3-dialdehyde bacterial cellulose for potential biodegradable tissue engineer-
ing scaffolds. Materials Science and Engineering C, 29, 1635e1642.
Maeno, S., Niki, Y., Matsumoto, H., Morika, H., Yatabe, T., & Funayama, A. (2005). The
effect of calcium ion concentration on osteoblast viability, proliferation and
differentiation in monolayer and 3D culture. Biomaterials, 26, 4847e4855.
Menachen, C., & Aharon, E. (1969). Oxidaton of cellulose by hydrogen peroxide.
Cellulose Chemistry and Technology, 3, 9e20.
Meng, X., Tian, F., Yang, J., He, C. N., Xing, N., & Li, F. (2010). Chitosan and
alginate polyelectrolyte complex membranes and their properties for wound
dressing application. Journal of Materials Science: Materials in Medicine, 21,
1751e1759.
Moreira, S., Silva, N. B., Almeida-Lima, J., Rocha, H. A., Medeiros, S. R., Alves, C., Jr.,
et al. (2009). BC nanofibers: In vitro study of genotoxicity and cell proliferation.
Toxicology Letters, 189, 235e241.
Morris, E. R., & Ross-Murphy, S. B. (1981). Chain flexibility of polysaccharides and
glycoproteins from viscosity measurement. In D. H. Northcote (Ed.), Techniques
in carbohydrate metabolism (pp. 1e46). Amsterdam: Elsevier/North-Holland.
Nakayama, A., Kakugo, A., Gong, J. P., Osada, Y., Takai, M., Erata, T., et al. (2004). High
mechanical strength double-network hydrogel with bacterial cellulose.
Advanced Functional Materials, 14, 1124e1128.
Okiyama, A., Shirae, H., Kano, H., & Yamanaka, S. (1992). Bacterial cellulose. I. Two-
stage fermentation process for cellulose production by Acetobacter aceti. Food
Hydrocolloids, 6, 471e477.
Ougiya, H., Watanabe, K., Morinaga, Y., & Yoshinaga, F. (1997). Emulsion effect of
bacterial cellulose. Bioscience, Biotechnology, and Biochemistry, 61, 1541e1545.
Paul, W., & Sharma, C. P. (2004). Chitosan and alginate wound dressings: a short
review. Trends in Biomaterials & Artificial Organs, 18, 18e23.
Phisalaphong, M., & Jatupaiboon, N. (2008). Biosynthesis and characterization of
bacterial cellulose-chitosan film. Carbohydrate Polymers, 74, 482e488.
Ritger, P. L., & Peppas, N. A. (1987). A simple equation for description of solute
release. I. Fickian and non-Fickian release from non-swellable devices in the
form of slabs, spheres, cylinders or discs. Journal of Controlled Release, 5, 23e36.
Shuangyun, L., Wenjuan, G., & Gu, H. Y. (2008). Construction, application and
biosafety of silver nanocrystalline chitosan wound dressing. Burns, 34, 623e628.
Sudheesh Kumar, P. T., Abhilash, S., Manzoor, K., Nair, S. V., Tamura, H., &
Jayakumar, R. (2010). Preparation and characterization of novel b-chitin/nano-
silver composite scaffolds for wound dressing applications. Carbohydrate Poly-
mers, 80, 761e767.
Venkatesan, J., Bhatnagar, I., & Kim, S. K. (2014). Chitosan-alginate bio-
composite containing fucoidan for bone tissue engineering. Marine Drugs,
12, 300e316.
Wiegand, C., Elsner, P., Hipler, U., & Klemm, D. (2006). Protease and ROS activities
influenced by a composite of bacterial cellulose and collagen typeⅠin vitro.
Cellulose, 13, 689e696.
 W.-S. Chang, H.-H. Chen / Food Hydrocolloids 53 (2016) 75e83 83

Wu, J., Zheng, Y., Wen, X., Lin, Q., Chen, X., & Wu, Z. (2014). Silver nanoparticle/
bacterial cellulose gel membranes for antibacterial wound dressing: investi-
gation in vitro and in vivo. Biomedical Materials, 9 (on line).
Yamanaka, S., Watanabe, K., Kitamura, N., Iguchi, M., Mitsuhashi, S., Nishi, Y., et al.
(1989). The structure and mechanical properties of sheets prepared from bac-
terial cellulose. Journal of Materials Science, 24, 3141e3145.
Yang, S. J. (2011). Development of fabricated composites film with bacterial cellulose
and the investigation of its physical properties. Master thesis of Department of
Food Science. I-Lan, Taiwan: National Ilan University.
Yang, J. M., Yang, S. J., Lin, H. T., Wu, T. H., & Chen, H. J. (2008). Chitosan containing
PU/Poly (NIPAAm) thermosensitive membrane for wound dressing. Materials
Science and Engineering C, 28, 150e156.