Carbohydrate Polymers 82 (2010) 173–180

Contents lists available at ScienceDirect

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

Physicochemical and mechanical characterization of bacterial cellulose produced

with an excellent productivity in static conditions using a simple fed-batch
cultivation strategy
Omer Shezad a , Salman Khan a , Taous Khan b , Joong Kon Park a,∗
a
Department of Chemical Engineering, Kyungpook National University, Daegu, 702-701, Republic of Korea
b
Department of Pharmacy, COMSATS Institute of Information Technology, Abbottabad, Pakistan

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

Article history: The aim of the present study was to investigate the structural modifications and physico-mechanical
Received 18 February 2010 properties of the BC sheets produced in static cultures using a newly developed fed-batch strategy and
Received in revised form 19 April 2010 a new medium (waste from beer fermentation broth; WBFB). Characteristics of these BC samples were
Accepted 20 April 2010
compared with BC sheets produced in batch and fed-batch cultivations using chemically defined medium
Available online 29 April 2010
(CDM). FT-IR and 13 C NMR spectra showed almost same results for all the BC samples. XRD analysis
revealed that BC produced in fed-batch cultivation using CDM had larger crystallite size than other BC
Keywords:
samples. Crystallinity index of BC from fed-batch cultivations was higher than batch cultivation. BC from
Bacterial cellulose
Gluconacetobacter hansenii PJK
WBFB had slightly lower degree of polymerization than BC from CDM. Mechanical strength of BC from
Fed-batch cultivation fed-batch cultivation using CDM was highest. WHC and WRR for BC from WBFB were greater than BC
Static culture from CDM.
Water holding capacity © 2010 Elsevier Ltd. All rights reserved.
Mechanical strength

1. Introduction Chang, 2007; Park, Hyun, & Jung, 2004; Park, Jung, & Park, 2003;
Park, Park, & Jung, 2003; Shah, Ha, & Park, 2010) as well waste
Cellulose, composed of (1 → 4)-␤-d-glucopyranose repeating from beer fermentation broth (WBFB) (Ha et al., 2008; Park, Hyun,
units, is the most abundant biopolymer on earth. It is the main & Ahn, 2006). However, the previous studies were mostly related
constituent of plant cell wall. Some kinds of bacteria of the genera to the exploration of the fundamental biotechnology and enhance-
Acetobacter, Rhizobium, Agrobacterium, and Sarcina can also pro- ment of productivity in agitated conditions. The agitated culture
duce cellulose called as bacterial cellulose (BC) (Khan, Park, & Kwon, method is mainly used for industrial production of BC (El-Saied,
2007). Recently, BC has received much attention as a new func- Basta, & Gobran, 2004) but most of the biomedical (e.g., artificial
tional material for biomedical and industrial applications because skin) and cosmoceutical (e.g., face mask) applications require BC to
of its superiority to plant cellulose in purity and supermolecu- be in a proper shape (e.g., film, membrane or sheet), which can be
lar structure. It has unique physical and chemical properties that produced only in static cultivations (Park et al., 2009). The process
lack in plant cellulose including high tensile strength, high water by which BC is conventionally produced using a static culture is
holding capacity, high crystallinity, ultra-fine and finely pure fiber not applicable to large-scale industrial production due to the low
network structure, transparency, fiber-binding ability, biocom- productivity. In order to overcome this low productivity problem
patibility, biodegradability and moldability (Ishihara, Matsunaga, in static cultures and to enhance it to a level suitable for com-
Hayashi, & Tišler, 2002; Klemm, Schumann, Udhardt, & Marsch, mercial applications, a simple fed-batch strategy was successfully
2001; Shezad, Khan, Khan, & Park, 2009; Vandamme, De Baets, applied for the production of BC sheets in a fermenter using CDM
Vanbaelen, Joris, & De Wulf, 1998). and WBFB (Shezad et al., 2009). The produced BC sheets appeared
Gluconacetobacter hansenii PJK is known to produce BC under to be applicable for biomedical and other applications.
various experimental conditions using a chemically defined It is known that the composition of the culture medium and
medium (CDM) (Jung, Park, & Chang, 2005; Jung, Khan, Park, & fermentation conditions greatly influence the chemical struc-
ture, composition and viscosity of the microbial polysaccharides
including BC (Duta, França, & Lopes, 2006; Khan & Park, 2008;
∗ Corresponding author. Tel.: +82 53 950 5621; fax: +82 53 950 6615. Nieduszynski & Preston, 1970; Watanabe, Tabuchi, Morinaga, &
E-mail address: [email protected] (J.K. Park). Yoshinaga, 1998). These modifications (if any) in turn may affect

0144-8617/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carbpol.2010.04.052
174 O. Shezad et al. / Carbohydrate Polymers 82 (2010) 173–180

the properties of BC. Therefore, it seemed important to investi- Lorentzian function; here  is the wavelength of X-ray radiation
gate the possible modifications in the physicochemical properties (1.54 Å) (Song et al., 2009).
of the BC sheets that may have been produced using a new strat- The crystallite index (CrIXRD ) was calculated with the help of the
egy (fed-batch) and using the new medium (WBFB) employed in following formula:
our previous studies (Shezad et al., 2009). For this purpose, the
I2 0 0 − Iam
BC sheets produced in fed-batch cultivation using CDM and the CrIXRD = × 100
I2 0 0
WBFB were subjected to detailed structural analyses using FT-IR,
XRD, CP/MAS 13 C NMR and gel permeation chromatography while where I2 0 0 is the maximum intensity of the (2 0 0) lattice diffraction
the various physical properties investigated included morphology, and Iam is the intensity diffraction at 2 (Focher et al., 2001).
mechanical strength, water holding capacity and water release rate.
These properties were compared with the physicochemical proper- 2.4. FT-IR analysis
ties of BC produced in batch cultivation using CDM. This study will
be helpful in determining if the BC obtained in the new conditions The surface properties of the BC samples were examined by a
is appropriate for commercial applications. Perkin-Elmer S2000 Fourier transform infrared spectrometer (FT-
IR, Nicolet Magna IR 560) in an attenuated total reflectance (ATR)
mode. BC samples were milled with KBr to form a very fine powder
2. Experimental
which were then compressed into a thin pellet and analyzed. All FT-
IR spectra were recorded in the transmittance mode in the range
2.1. Production and processing of BC sheets
of 4000–400 cm−1 with a resolution of 0.25 cm−1 .
The BC sheets used in this study were obtained in previous inves-
2.5. Cross polarization/magic angle spinning (CP/MAS) 13 C NMR
tigations via batch and fed-batch cultivation strategy using CDM
spectroscopy
and WBFB (Shezad et al., 2009). Briefly, the colonies of G. hansenii
PJK (KCTC 10505BP) were inoculated into a 50 mL CDM (contain-
The CP/MAS 13 C NMR spectra were recorded (at 294 ± 1 K) on
ing 10 g glucose, 10 g yeast extract, 7 g peptone, 1.5 mL acetic acid,
the Bruker Avance II+ -400 instrument operating at 9.4 T. A dou-
and 0.2 g succinate dissolved in 1 L of distilled water and its pH
ble air-bearing probe and a zirconium oxide rotor were used. The
adjusted to 5) in a 250 mL flask which was shaken at 200 rpm and
MAS rate was in the range of 6–10 kHz. Acquisition was performed
cultured at 30 ◦ C for 24 h. For batch cultivation, 100 mL pre-culture
with a standard CP pulse sequence using a 4.2 ␮s proton 90 pulse, a
was inoculated into 2 L CDM containing 1% ethanol in a 3 L jar fer-
2000 ␮s contact pulse and 3 s delays between repetitions. Adaman-
menter. In fed-batch cultivation, 25 mL pre-culture was inoculated
tine was used as an external standard for the chemical shift scale
into 500 mL of culture medium (CDM or WBFB) in a 3 L jar fer-
relative to tetramethylsilane.
menter. A fresh medium (250 mL) was fed periodically. In either
case, the fermentations were carried out at 30 ◦ C with an aeration
2.6. Gel permeation chromatography (GPC)
rate of 1 vvm.
The BC sheets were harvested after 30 days cultivation in each
Weight average molecular weight (Mw), polydispersity index
case and washed with distilled water repeatedly in order to remove
(Mw/Mn) and degree of polymerization (DP) for the nitrated
the residual medium and other impurities. These BC sheets were
BC samples were determined by a high-performance gel per-
then repeatedly treated by boiling in 0.3 N NaOH solution for
meation chromatography system (GPCV 2000, Water Alliance,
15–20 min followed by washing with distilled water. Finally, the
USA) according to the technique described elsewhere (Kuga et al.,
sheets were stored in containers containing distilled water while
1989). Each cellulose sample was nitrated in a solution of fuming
for characterization; portions of the BC sheets were freeze-dried.
nitric acid/diphosphorous pentaoxide according to the method of
Alexander and Michell (1949). Briefly, the nitrating mixture was
2.2. FE-SEM analysis prepared by dissolving 120 g of phosphorous pentaoxide in 300 g
of nitric acid (90%, Sigma–Aldrich). Then 1 g BC was introduced into
In order to determine the morphology and surface topography nitrating mixture (40 g), and reacted at 20 ◦ C for 20 min with the
of the BC sheets, micrographs of the platinum-coated samples of sample being swirled at about 5 min intervals. The resulting BC
freeze dried BC were taken with a field-emission scanning electron product was thoroughly washed with cold (10 ◦ C) distilled water
microscope (S-4300; Hitachi Co., Japan). and neutralized with 5% sodium carbonate solution followed by
three washes with distilled water. The BC product was then boiled
2.3. XRD analysis in distilled water for 20 min and soaked in methyl alcohol (50 mL)
for 10 min. The final product was obtained by drying the BC sam-
The XRD technique was employed in order to determine the ples in a mechanical convection oven at 50o C for about 1 h. The
crystallite size and crystallinity of the BC samples. The freeze- molecular weight of the samples was checked with GPC using
dried BC samples were studied using a Rigaku D/max 2500 X-ray tetrahydrofuran as eluent.
diffractometer with a thin film attachment and operated at room
temperature. The Cu K␣ X-ray source was set to 40 kV and 100 mA. 2.7. Mechanical strength
The samples were scanned from 10◦ to 80◦ with a scan speed of
4◦ /min. The data were obtained with the help of a MDI/JADE6 The tensile properties of the BC membranes were measured
software package attached to the Rigaku XRD instrument. The using an Instron Universal Testing Machine (Model 4465, USA)
Scherrer’s formula: according to the procedure of American Society for Testing and
Materials (ASTM D 882). Two metal clamps were placed at either
k
Crystallite Size = cos  end of a 100 mm × 10 mm rectangular strip of dried BC sample in
W each case. The clamps were then mounted on an Instron 4465 that
(with a shape factor k = 0.89) was employed to determine the crys- measured the maximum tensile strength before fracture. Thickness
tallite sizes of samples with full width at half maximum (fwhms, of each sample was measured by using a Peacock Model G (made
W) and peak centers obtained by fitting the (1 0 1) peak to the in Japan).
 O. Shezad et al. / Carbohydrate Polymers 82 (2010) 173–180 175

2.8. Measurement of water holding capacity (WHC) and water

release rate (WRR)

The water holding capacity of the samples was measured using

the shake method (Schrecker & Gostomski, 2005). Three BC samples
for each batch (with the dimension 7.5 cm × 5 cm × 0.8 cm were
removed from the storage container using tweezers. The samples
were shaken twice quickly and then weighed. These samples were
dried at room temperature for 48 h and their weights were mea-
sured at different time intervals. Then, BC samples were dried for
12 h at 60 ◦ C in order to completely remove water from them.
Finally, they were transferred quickly to the balance for weigh-
ing. The water holding capacity was calculated using the following
formula:

Water Holding Capacity

Mass of water removed during drying (g)
=
Dry weight of cellulose (g)

For the determination of the water release rate, the wet weight
of BC samples was measured followed by continuously weighing
the samples stored at ambient conditions at different time intervals.
Finally the dry weight of the respective BC samples was taken which
was subtracted from all the readings. Similarly the loss of water at
different time intervals was plotted against time.

3. Results and discussion

3.1. Morphology of BC sheets prepared under various conditions

BC is accumulated at the surface of the culture medium as a

gelatinous membrane in a static culture (Hestrin & Schramm, 1954).
Nascent chains of BC aggregate to form subfibrils which are then
crystallized into microfibrils, these into bundles, and the latter into
ribbons (Park et al., 2009).
Fig. 1 shows the scanning electron micrographs of the BC
sheets produced in different cultivation modes and media. These
micrographs clearly show the various morphological features of
these sheets. All samples showed a reticulated structure con-
sisting of ultra fine cellulose fibrils. The gross morphological
structure seemed to be similar for all specimens. However, detailed
examination of the micrographs revealed profound morphological
differences among the nanofibrils and the reticulated structures
of BC produced under different culture conditions. Depending on
their origin, these BC nanofibrils have transverse dimensions that
range from 10 to 180 nm. The fibrils of BC produced in fed-batch
Fig. 1. FE-SEM micrographs of bacterial cellulose samples produced in batch culti-
cultivation using CDM are highly extended and almost uniform in vation using chemically defined medium (A), fed-batch cultivation using chemically
size. In contrast, the fibrils of BC sheets produced in batch cultiva- defined medium (B), and fed-batch cultivation using waste from beer fermentation
tion appeared to be smaller in size. The fibrils of BC produced in broth (C) in static conditions in a Jar fermenter.
fed-batch cultivation using WBFB are more crowded and thinner
than BC fibrils produced using CDM. The exact width of individ- diagram indicated that BC synthesized in different culture con-
ual microfibrils in each case, however, is difficult to estimate. The ditions has both I␣ and I␤ crystal cellulose. Fig. 2 revealed three
morphological changes in BC sheets affect the microstructures and characteristic diffraction peaks in the region of 10–35◦ under dif-
various properties including the degree of polymerization, crys- ferent conditions while the crystalline parameters of (0 0 2) crystal
tallinity, content of cellulose I␣ and water holding capacity (Park et plate of BC synthesized in different culture modes are listed in
al., 2009; Watanabe et al., 1998). Table 1.
Fig. 2 represents X-ray diffraction patterns of BC samples pro-
3.2. XRD analysis duced using CDM in batch and fed-batch cultivation modes and
WBFB in fed-batch cultivation. These patterns are apparently simi-
The morphological changes in the ribbons of BC are related to lar to each other which are based on the BC polymorphs examined
the changes in microstructures such as crystallinity, I␣ fraction in this study. The fiber diagram of all the cellulose samples is well
and hydrogen holding between the molecules (Sun et al., 2007; resolved pattern of the I␣ rich type of cellulose (55.58, 58.94 and
Watanabe et al., 1998). In order to compare the microstructural 58.73% of I␣ phase respectively), which identifies the cellulose pro-
changes in the BC sheets produced under different cultivation duction by bacteria. However, there are some slight differences in
modes and conditions, X-ray diffraction was used. The diffraction their structural features. Fig. 2B and C approximately shows simi-
176 O. Shezad et al. / Carbohydrate Polymers 82 (2010) 173–180

Fig. 2. X-ray diffraction patterns of bacterial cellulose samples produced in batch

cultivation using chemically defined medium (A), fed-batch cultivation using
chemically defined medium (B), and fed-batch cultivation using waste from beer
fermentation broth (C) in static conditions in a Jar fermenter.

Fig. 3. FT-IR spectrum of bacterial cellulose samples produced using chemically

defined medium in batch cultivation (A), chemically defined medium in fed-batch
lar cellulose I␣ and cellulose I␤ mass fraction while Fig. 2A shows cultivation (B), and waste from beer fermentation broth in fed-batch cultivation (C)
slightly lower mass fraction of cellulose I␣ . The former two have in static conditions in a Jar fermenter.
been produced in fed-batch cultivation while the later in batch cul-
tivation mode. This shows that cultivation mode has an effect on
cellulose I␣ and cellulose I␤ mass fraction, while change in culture 3.3. FT-IR analysis
medium has no significant effect. Moreover, the d-spacing is almost
the same in all the cases. Fourier transform infrared (FT-IR) spectra of BC samples were
BC produced in fed-batch cultivation using CDM has larger crys- taken in order to detect any peak shift that could be attributed to
tallite size of a crystallographic plane (1 1 0) than BC produced in BC produced in different media and cultivation modes. Fig. 3 shows
batch cultivation using CDM or fed-batch cultivation using WBFB the FT-IR spectra of BC produced in different media and cultivation
(Table 1). The latter sample has the smallest crystal size. Similarly, modes. Table 2 indicates the characteristic bands of BC synthe-
the crystallinity index of BC samples produced in fed-batch culti- sized in different modes and media appeared at ∼3430–3435 cm−1
vation using WBFB is highest. WBFB is a complex natural medium for hydroxyl groups stretching vibration, at ∼2927–2949 cm−1 for
consisting of several ingredients. It may also contain some con- C–H stretching vibration, at ∼1433–1456 cm−1 for C–H bending
stituent(s) which can serve as a dispersant to lessen the aggregation vibration and at ∼1045–1067 cm−1 for C–O–C and C–O–H stretch-
of particles. The cultivation conditions may also affect the aggrega- ing vibration of sugar ring (Socrates, 2001; Sun et al., 2007). This
tion of particles. carbonyl amide group may come from proteins and bacterial cells
From these results it can be concluded that the cultivation mode which are not completely wiped off after the NaOH treatment of
has a greater effect on cellulose I␣ and cellulose I␤ mass frac- the BC membrane (Sun et al., 2007). The carbonyl amide peak was
tion, crystallite size and crystallinity index. However, the culture shorter in the spectra of BC produced in fed-batch cultivation using
medium has a significant effect on crystallite size and crystallinity WBFB or in the batch cultivation using CDM than the in fed-batch
index. cultivation using CDM. This means that the former two samples are

Table 1
d-Spacing, crystallite sizes and cellulose I␣ and I␤ content (%) of different bacterial cellulose samples determined from X-ray diffractograms.

BC sample d-Spacing (Å) Crystallite sizes (nm) Cellulose type (%) Crystallinity index (%)

Dl D2 Crystal 1 Crystal 2 I␣ I␤

Batch cultivation (CDM)b 6.12a 3.93a 10.37 6.86 55.58 44.41 78.0
Fed-batch cultivation (C DM)b 6.18a 3.91a 10.95 7.22 58.94 41.05 75.5
Fed-batch cultivation (WBFB)b 6.13a 3.94a 9.07 5.74 58.73 41.26 80.8
a
Where d is the perpendicular distance sepamting each lattice plane in a stack.
b
Abbreviations: CDM, chemically defined medium; WBFB, waste from beer fermentation broth.

Table 2
Assignment of FT-IR spectra of bacterial cellulose samples produced at various culture conditions.

Band assignments Wave number (cm−1 )

Batch cultivation (CDM)a Fed-batch cultivation (CDM)a Fed-batch cultivation (WBFB)a

O–H stretching 3430 3435 3435

C–H stretching 2922 2964 2923
C–H bending 1456 1443 1456
C–O–H bending, C–OH stretching 1060 1070 1056
–CO–NH 1641 1600 1632
O–D stretching Negligible 2489 Negligible
–CO–Br Negligible 1772 Negligible
N–H out of plan bending Negligible 885 Negligible
a
Abbreviations: CDM, chemically defined medium; WBFB, waste from beer fermentation broth.
 O. Shezad et al. / Carbohydrate Polymers 82 (2010) 173–180 177

Fig. 4. The CP/MAS 13 C NMR spectra of bacterial cellulose samples produced in batch cultivation using chemically defined medium (A), fed-batch cultivation using chemically
defined medium (B), and fed-batch cultivation using waste from beer fermentation broth (C) in static conditions in a Jar fermenter.

more pure than the later one. The band position in the FT-IR spec- previously (Horii, Hirai, & Kitamaru, 1984; Newman, 1998;
tra for all the three BC samples were in the same range and did not Newman & Hemmingson, 1995; Wickholm, Larsson, & Iversen,
show any significant difference. 1998). A slight shoulder on the C6 peak, between 63 and 65 ppm,
Three additional dominant peaks are found in the spectra of BC was attributed to the amorphous and disordered component in cel-
produced in fed-batch cultivation using CDM as shown in Table 2, lulose (Horii et al., 1984; Newman, 1998; Newman & Hemmingson,
at ∼2489 cm−1 for O–D stretching (Socrates, 2001), at 885 cm−1 for 1995; Wickholm et al., 1998). Such a shoulder is absent in the spec-
N–H out of plan bending (Pavia et al., 2001) and at ∼1772 cm−1 for tra obtained in the present studies.
acid halide (–CO–Br) (Socrates, 2001). However, two uncommon peaks are found in 13 C NMR spec-
tra of BC produced in batch and fed-batch cultivation using CDM as
shown in Table 3. The uncommon peaks at 173.81, 164.48 ppm rep-
3.4. CP/MAS 13 C NMR spectroscopy resent the presence of carbonyl amide group (Pavia et al., 2001; Sun
et al., 2007). These peaks are negligible in the spectra of BC produced
CP/MAS is a useful technique for monitoring the morphological in fed-batch cultivation using WBFB. The higher chemical shift
changes taking place in cellulose during processing (Bhattacharya, value for BC obtained in batch compared to fed-batch cultivation
Louis, & William, 2008). Therefore, this technique was applied in the using CDM represents the presence of secondary or tertiary amine
present study in order to monitor the changes in cellulose caused attached to the carbonylic carbon. The lower chemical shift value
by the changes in the fermentation conditions. 13 C NMR spectra of for BC obtained in fed-batch cultivation using CDM represents the
BC produced in batch and fed-batch cultivation using CDM and fed- possible presence of primary amine attached to the carbonylic car-
batch cultivation using WBFB are presented in Fig. 4. The peaks for bon. The electron withdrawing effect of tertiary/secondary amine
the six carbon atoms of the cellulose molecule were dominant in group is higher compared to primary amine (Pavia et al., 2001)
the spectra in all cases. The chemical shift values ranged from 105 hence caused the high chemical shift in case of BC produced in batch
to 65 ppm. The anomeric carbon (C1 ) appeared at around 105 ppm cultivation. This can be coincident with a relatively evident band at
followed by the signal from the C4 atom at around 89 ppm. Peaks 885 cm−1 in FT-IR spectra for N–H out of plan bending in case of BC
arising due to C2 , C3 and C5 atoms make their appearance between produced in fed-batch cultivation using CDM. Similarly, the peaks
70 and 75 ppm and finally the C6 peak has a chemical shift value at at 32.87 ppm are also dominant for BC produced using CDM which
around 65 ppm. These values are in agreement with those reported indicate the presence of saturated carbon attached to the carbonyl
in the literature for cellulose (Bhattacharya et al., 2008). amide group (Pavia et al., 2001). The overall studies of NMR and
In previous studies on solid-state NMR spectra of cellulose two FT-IR spectra indicate that the representative peaks for carbonyl
peaks appeared in the chemical shift range 80–92 ppm which were groups attached with amines are dominant for BC produced using
assigned to the C4 carbon atom (Atalla, Gast, Sindorf, Bartuska, & CDM compared to WBFB. The slight reduction in wave number for
Maciel, 1980; Earl & VanderHart, 1981). The relatively sharp peak O–H stretching and an increase in chemical shift for BC produced
was due to crystalline regions while broader peak was assigned using CDM compared to WBFB indicate the presence of traces of
to the crystallite surfaces and the amorphous/disordered domains bacterial bodies in the BC even after treatment with NaOH solution.
(Atalla et al., 1980; Earl & VanderHart, 1981). In contrast, a sin- Hence, from this expanded NMR and FT-IR studies given in
gle and sharp peak at 89 ppm, assigned to C4 , was obtained in the Tables 2 and 3, both the spectra argue for the presence of some pro-
current study which showed the crystalline nature of the obtained teinacious bodies which gave rise to additional peaks. The spectra
BC. The correlation of the cellulose NMR spectra to its structure clarify that the peaks for carbonyl amide are relatively domi-
and morphology was also revealed by the assignments reported nant for BC produced using CDM compared to WBFB. Hence, this
178 O. Shezad et al. / Carbohydrate Polymers 82 (2010) 173–180

Table 3
13
C Chemical shifts of bacterial cellulose samples produced at various culture conditions.
13
Culture type C chemical shifts (ppm)

C1 C2 , C3 , C5 C4 C6 R–CO–NH2 Saturated card on

Batch cultivation (CDM)a 105.11 70–75 89.00 65.29 173.81 32.81

Fed-batch cultivation (CDM)a 105.11 70–75 89.03 65.29 164.48 32.81
Fed-batch cultivation (WBFB)a 105.13 70–75 89.03 65.27 Negligible Negligible
a
Abbreviations: CDM, chemically defined medium; WBFB, waste from beer fermentation broth.

support that the thick and long fibrils of BC produced using CDM Generally, the higher the molecular weight of a polymer, the higher
hold the bacterial cells more firmly and acted as a barrier during is its mechanical strength. The mechanical strength of the BC with
washing which resulted in the remain of some bacterial cell debris. high DP will be greater than that of BC with low DP. The polydisper-
sity index (PDI), an index of polymeric uniformity, of BC produced
3.5. Degree of polymerization in all culture condition is similar (Table 4).

Molecular weight is an important property of polymeric mate- 3.6. Mechanical strength

rials (Choi, Song, Kim, Chang, & Kim, 2009). Many of the unique
physical properties of BC are attributed to the high molecular Tensile testing was used to evaluate the inherent mechanical
weight, molecular homogeneity and chemical purity. In the cur- properties of the produced BC sheets and to determine the effect
rent study the Mw characterization of BC samples were carried of various culture conditions on the mechanical properties. Table 4
out using the GPC method. This method furnishes comprehensive shows the maximum tensile stress of BC sample obtained by batch
information regarding all important molecular parameters of the cultivation using CDM and fed-batch cultivation using WBFB and
polymer, that is, number and weigh-average molecular weights CDM. The results displayed that the mechanical strength of BC
(Mn and Mw), polydispersity (Mw/Mn) and degree of polymeriza- sample obtained by fed-batch cultivation using CDM is highest
tion (Mn/monomer molecular weight). compared to fed-batch cultivation using WBFB and batch cultiva-
The retention times of BC samples produced in batch and fed- tion using CDM. The GPC results (Fig. 5 and Table 4) showed that
batch cultivation using CDM, and fed-batch cultivation using WBFB the degree of polymerization for batch cultivation using CDM is
were 22.450, 22.540, and 22.542 min, respectively. Fig. 5 shows greater than fed-batch cultivation using CDM and WBFB. As dis-
the degree of polymerization patterns of BC samples produced in cussed earlier, DP has a greater effect on the mechanical strength.
different culture conditions. It is obvious from these results that The mechanical strength of BC with higher DP will be greater than
BC samples obtained from fed-batch cultivation using WBFB has BC with lower DP (Cheng, Catchmark, & Demirci, 2009). However,
slight lower DP than BC produced in batch and fed-batch cultivation the results obtained for mechanical strength of BC samples pro-
using CDM. As shown in Table 4, the highest DP value (8011) was duced under various culture conditions do not obey this theory.
obtained for the BC produced in batch culture using CDM. The low- The discrepancy can be explained from the FE-SEM images. The
est DP value for BC produced in fed-batch cultivation using WBFB fibrils of BC sample produced using batch cultivation using CDM
may be due to the fact that WBFB is a complex natural medium are less uniform and smaller in size as compared to the BC sam-
composing of several ingredients. It may also contain some con- ples produce in fed-batch cultivation and thus possess the least
stituent(s) which can serve as dispersant to lessen the aggregation mechanical strength. Similarly, the fibrils of BC produced in fed-
of particles or constituent(s) which may interferes with the hydro- batch cultivation using WBFB are also less uniform in size but they
gen bonding between microfibrils resulting in a reduced length of are more crowded and thus possess the higher mechanical strength.
the microfibrils, and thus the Mw of BC (Choi et al., 2009). The The mechanical strength of BC sheets also depends on the orien-
cultivation conditions may also effect the aggregation of particles. tation of the micro-fibrils in ribbons (Keshk, 2006; Yano, Maeda, &
Nakajima, 2008). BC sheets produced in fed-batch cultivation using
CDM have considerable uniplanar orientation of micro-fibrils com-
pared to other BC samples. This may be another basic reason for
the high mechanical strength of BC sheets produced in fed-batch
cultivation using CDM.

3.7. Water holding capacity (WHC) and water release rate (WRR)

Maintenance of proper wound moisture is important in mod-

ern therapy because the penetration of active substances into the
wound is facilitated by a moist environment. This also enables an
easy and painless dressing change without damage to the newly
formed skin (Newman, 1998). In order to estimate the usefulness of
the BC sheets produced under various cultivation conditions in the
treatment of wounds requiring wet conditions and frequent dress-
ing change, their water holding capacity (WHC) and water release
rate (WRR) was determined.
The WHC for the BC sample produced in fed-batch cultivation
using CDM was found to be 273 times its dry weight while the WHC
for batch cultivation was 296 times its dry weight. This means that
Fig. 5. GPC analysis of bacterial cellulose samples produced in batch cultivation
WHC for BC produced in batch cultivation is higher than the BC pro-
using chemically defined medium (A), fed-batch using chemically defined medium
(B), and fed-batch using waste from beer fermentation broth (C) in static conditions duced in fed-batch using chemically defined medium. However, the
in a Jar fermenter. reverse was true for WRR i.e., release of water from fed-batch BC
 O. Shezad et al. / Carbohydrate Polymers 82 (2010) 173–180 179

Table 4
Molecular weight and its distribution for bacterial cellulose produced at various culture conditions.

BC sample Mna Mwa Mpa PDIa DPa Max tensile stress (MPa)
a
Batch cultivation (CDM) 1,297,924 1,968,687 3,020,792 1.516797 8011 76.7
Fed-batch cultivation (CDM)a 1,286,410 1,945,804 2,766,903 1.512584 7940 19.8
Fed-batch cultivation (WBFB)a 1,250,774 1,914,041 2,753,342 1.530285 7720 4.0
a
Abbreviations: Mn, number average molecular weight; Mw, weight average molecular weight; Mp, molecular weight peak at top; PDI, polydispersity index (Mw/Mn); DP,
degree of polymerization; CDM, chemically defined medium; WBFB, waste from beer fermentation broth.

References

Alexander, W. J., & Michell, R. L. (1949). Rapid measurement of cellulose viscosity

by the nitration method. Analytical Chemistry, 21, 1497–1500.
Atalla, R. H., Gast, J. C., Sindorf, D. W., Bartuska, V. J., & Maciel, G. E. (1980). Carbon-13
NMR spectra of cellulose polymorphs. Journal of American Chemical Society, 102,
3249–3251.
Bhattacharya, D., Louis, T. G., & William, T. W. (2008). Isolation, preparation and
characterization of cellulose microfibers obtained from bagasse. Carbohydrate
Polymers, 73, 371–377.
Cheng, K.-C., Catchmark, J. M., & Demirci, A. (2009). Effect of different additives on
bacterial cellulose production by Acetobacter xylinum and analysis of material
property. Cellulose, 16, 1033–1045.
Choi, C. N., Song, H. J., Kim, M. J., Chang, M. H., & Kim, S. J. (2009). Properties of bacte-
rial cellulose produced in a pilot-scale spherical type bubble column bioreactor.
Korean Journal of Chemical Engineering, 26, 136–140.
Ciechańska, D. (2004). Multifunctional bacterial cellulose/chitosan composite mate-
rials for medical applications. Fibres & Textiles in Eastern Europe, 12, 69–72.
Duta, F. P., França, F. P., & Lopes, L. M. A. (2006). Optimization of culture for
exopolysaccharide production in Rhizobium sp. using the response surface
method. Electronic Journal of Biotechnology, 9, 391–399.
Earl, W. L., & VanderHart, D. L. (1981). Observations by high resolution carbon-13
nuclear magnetic resonance of cellulose I related to morphology and crystal
structure. Macromolecules, 14, 570–574.
El-Saied, H., Basta, A. H., & Gobran, R. H. (2004). Research progress in friendly
environmental technology for the production of cellulose products (bacterial
Fig. 6. Water release rate for bacterial cellulose samples produced in batch using
cellulose and its application. Polymer-Plastics Technology and Engineering, 43,
chemically defined medium (A), fed-batch cultivation using chemically defined 797–820.
medium (B), and fed-batch cultivation using waste from beer culture broth (C) in Focher, B., Palma, M. T., Canetti, M., Torri, G., Cosentino, C., & Gastaldi, G. (2001).
static conditions in a Jar fermenter. Structural differences between non-wood plant celluloses: Evidence from solid
state NMR, vibrational spectroscopy and X-ray diffractometry. Industrial Crops
and Products, 13, 193–208.
took more time as compared to batch cultivated BC sample (Fig. 6). Ha, J. H., Shehzad, O., Khan, S., Lee, S. Y., Park, J. W., Khan, T., et al. (2008). Production
After 64 h almost all of the water was released from the batch sam- of bacterial cellulose by a static cultivation using the waste from beer culture
ple, while in case of the fed-batch sample water was completely broth. Korean Journal of Chemical Engineering, 25, 812–815.
Hestrin, S., & Schramm, M. (1954). Synthesis of cellulose by Acetobacter xylinum:
released after 75 h. As revealed from SEM morphology that fed- preparation of freeze dried cells capable of polymerizing glucose to cellulose.
batch BC fibrils are thicker in size as compared to batch BC fibrils. Biochemical Journal, 58, 345–352.
Therefore, the water molecules are sandwiched between the large Horii, F., Hirai, A., & Kitamaru, R. (1984). CP/MAS carbon-13 NMR study of spin
relaxation phenomena of cellulose containing crystalline and noncrystalline
and thick fibrils of fed-batch BC, and these fibrils act as a shield components. Journal of Carbohydrate Chemistry, 3, 641–662.
for water molecules. Hence, these fibrils prevent the water from Ishihara, M., Matsunaga, M., Hayashi, N., & Tišler, V. (2002). Utilization of D-xylose
moving out of the BC membrane at high speeds. The fibrils of batch as carbon source for production of bacterial cellulose. Enzyme and Microbial
Technology, 31, 986–991.
cultivation BC are smaller, and having a large surface area. Thus Jung, J. Y., Khan, T., Park, J. K., & Chang, H. N. (2007). Production of bacterial cellulose
holding more water as compared to fed-batch BC sample. The WHC by Gluconacetobacter hansenii using a novel bioreactor equipped with a spin
of fed-batch BC sample using WBFB was found to be 302 times its filter. Korean Journal of Chemical Engineering, 24, 265–271.
Jung, J. Y., Park, J. K., & Chang, H. N. (2005). Bacterial cellulose production by
dry weight and is larger when compared to the BC samples from
Gluconacetobacter hansenii in an agitated culture without living non-cellulose
the chemically defined medium (both batch and fed-batch BC). producing cells. Enzyme and Microbial Technology, 37, 347–354.
Ougiya, Watanabe, Matsumura, and Yoshinaga (1998) reported that Keshk, S. (2006). Physical properties of bacterial cellulose sheets produced in pres-
ence of lignosulfonate. Enzyme and Microbial Technology, 40, 9–12.
BC samples with thinner and longer fibrils have high WHC capacity.
Khan, T., & Park, J. K. (2008). The structure and physical properties of glucuronic acid
The high WHC of BC produced in fed-batch cultivation using WBFB oligomers produced by a Gluconacetobacter hansenii strain using the waste from
as culture medium may be due to thinner and more crowded fibrils. beer fermentation broth. Carbohydrate Polymers, 73, 438–445.
The amount of the bound water in bacterial cellulose was Khan, T., Park, J. K., & Kwon, J.-H. (2007). Functional biopolymers produced by
biochemical technology considering applications in food engineering. Korean
reported to be negligible (Wickholm et al., 1998) and this water is Journal of Chemical Engineering, 24, 816–826.
trapped physically at the surface and on the inside of the particles Klemm, D., Schumann, D., Udhardt, U., & Marsch, S. (2001). Bacterial synthesized
composed of the reticulated fibrils (Watanabe et al., 1998). The slow cellulose-artificial blood vessels for microsurgery. Progress in Polymer Science,
26, 1561–1603.
release of water from BC is important in biomedical applications i.e., Kuga, S., Muton, N., Usuda, M., & Brown, R. M., Jr. (1989). Structural and functional
wound dressing (Ciechańska, 2004; Okiyama, Motoki, & Yamanaka, aspects. In J. F. Kennedy, G. O. Phillips, & P. A. Williams (Eds.), Cellulose (pp.
1992). In this context, the BC produced using WBFB seems to be 81–86). Chichester, UK: Ellis Horwood Limited.
Newman, R. H. (1998). Evidence for assignment of 13 C NMR signals to cellulose
superior to that produced using chemically defined medium. crystallite surfaces in wood, pulp, and isolated celluloses. Holzforschung, 52,
157–159.
Acknowledgement Newman, R. H., & Hemmingson, J. A. (1995). Carbon-13 NMR distinction between
categories of molecular order and disorder in cellulose. Cellulose, 2, 95–110.
Nieduszynski, I., & Preston, R. D. (1970). Crystallite size in natural cellulose. Nature,
This work was supported by the Korea Research Foundation 225, 273–274.
Grant funded by the Korean Government (MOHERD) (KRF-2007- Okiyama, A., Motoki, M., & Yamanaka, S. (1992). Bacterial cellulose II. Processing of
the gelatinous cellulose for food materials. Food Hydrocolloids, 6, 479–487.
521-D00105).
180 O. Shezad et al. / Carbohydrate Polymers 82 (2010) 173–180

Ougiya, H., Watanabe, K., Matsumura, T., & Yoshinaga, F. (1998). Relationship Shezad, O., Khan, S., Khan, T., & Park, J. K. (2009). Production of bacterial cellulose
between suspension properties and fibril structure of disintegrated bacterial in static conditions by a simple fed-batch cultivation strategy. Korean Journal of
cellulose. Bioscience, Biotechnology, and Biochemistry, 62, 1714–1719. Chemical Engineering, 26, 1689–1692.
Park, J. K., Hyun, S. H., & Ahn, W. S. (2006). Production of bacterial cellulose using Socrates, G. (2001). Infrared and Raman characteristics group frequencies (3rd ed.).
waste of beer fermentation broth. Korean Chemical Engineering Research, 44, John Wiley and Sons Limited.
52–57. Song, H. J., Li, H., Seo, J. H., Kim, M. J., & Kim, S. J. (2009). Pilot scale production of bac-
Park, J. K., Hyun, S. H., & Jung, J. Y. (2004). Conversion of G. hansenii PJK into non- terial cellulose by a spherical type buble column bioreactor using saccharified
cellulose-producing mutants according to the culture condition. Biotechnology food wastes. Korean Journal of Chemical Engineering, 26, 141–146.
and Bioprocess Engineering, 9, 383–388. Sun, D., Zhou, L., Wu, Q., & Yang, S. (2007). Preliminary research on structure and
Park, J. K., Jung, J. Y., & Park, Y. H. (2003). Cellulose production by Gluconacetobacter properties of nano-cellulose. Journal of Wuhan University of Technology: Materials
hansenii in a medium containing ethanol. Biotechnology Letters, 25, 2055–2059. Science Edition, 22, 677–680.
Park, J. K., Khan, T., & Jung, J. Y. (2009). Bacterial cellulose. In G. O. Phillips, & P. A. Vandamme, E. J., De Baets, S., Vanbaelen, A., Joris, K., & De Wulf, P. (1998). Improved
Williams (Eds.), Handbook of hydrocolloids (2nd ed., pp. 724–739). Cambridge, production of bacterial cellulose and its application potential. Polymer Degrada-
UK: Woodhead Publishing Limited. tion and Stability, 59, 93–99.
Park, J. K., Park, Y. H., & Jung, J. Y. (2003). Production of bacterial cellulose by Watanabe, K., Tabuchi, M., Morinaga, Y., & Yoshinaga, F. (1998). Structural features
Gluconacetobacter hansenii PJK isolated from rotten apple. Biotechnology and and properties of bacterial cellulose produced in agitated culture. Cellulose, 5,
Bioprocess Engineering, 8, 83–88. 187–200.
Pavia, D. L., Lampman, G. M., & Kriz, G. S. (2001). Introduction to spectroscopy Wickholm, K., Larsson, P. T., & Iversen, T. (1998). Assignment of non-crystalline forms
(3rd ed.). Hartcourt College Publishers. in cellulose I by CP/MAS carbon-13 NMR spectroscopy. Carbohydrate Research,
Schrecker, S. T., & Gostomski, P. A. (2005). Determining the water holding capacity 312, 123–129.
of microbial cellulose. Biotechnology Letters, 27, 1435–1438. Yano, S., Maeda, H., & Nakajima, M. (2008). Preparation and mechanical properties
Shah, N., Ha, J. H., & Park, J. K. (2010). Effect of reactor surface on production of bacte- of bacterial cellulose nanocomposites loaded with silica nanoparticles. Cellulose,
rial cellulose and water soluble oligosaccharides by Gluconacetobacter hansenii 15, 111–120.
PJK. Biotechnology and Bioprocess Engineering, 15, 110–118.