Cellulose (2009) 16:1047–1055

DOI 10.1007/s10570-009-9340-y

Mechanical and structural properties of native and alkali-

treated bacterial cellulose produced by Gluconacetobacter
xylinus strain ATCC 53524
Brigid A. McKenna Æ Deirdre Mikkelsen Æ
J. Bernhard Wehr Æ Michael J. Gidley Æ
Neal W. Menzies

Received: 16 April 2009 / Accepted: 30 June 2009 / Published online: 17 July 2009
Ó Springer Science+Business Media B.V. 2009

Abstract Mechanical properties of hydrated bacte- Keywords Cellulose
Gluconacetobacter xylinus

rial cellulose have been tested as a function of Mechanical properties
Scanning electron
fermentation time and following the alkali treatment microscopy
Tensile testing
required for sterilisation prior to biomedical applica-
tions. Bacterial cellulose behaves as a viscoelastic
material, with brittle failure reached at approximately Introduction
20% strain and 1.5 MPa stress under uniaxial tension.
Treatment with 0.1 M NaOH resulted in minimal Gluconacetobacter xylinus (formerly Acetobacter xyl-
effects on the mechanical properties of bacterial inum and Acetobacter xylinus) is a non-pathogenic,
cellulose. Fermentation time had a large effect on Gram-negative, rod-shaped, obligate aerobe, belong-
both bacterial numbers and cellulose yield but only ing to the family Acetobacteraceae. (Kersters et al.
minor effects on mechanical properties, showing that 2006; Yamada et al. 1997) A notable feature of this
the fermentation system is a robust method for bacterium is its ability to produce extracellular cellu-
producing cellulose with predictable materials prop- lose as a pure, ultra-fine random fibre network,
erties. The failure zone in uniaxial tension was shown possessing high crystallinity, high water absorption
to be associated with large-scale fibre alignment, capacity and mechanical strength (Cannon and Ander-
consistent with this being the major determinant of son 1991; Jonas and Farah 1998; Ross et al. 1991). The
mechanical properties. Under uniaxial tension, elastic cellulose network produced is referred to as a pellicle.
moduli and failure stresses are an order of magnitude In recent years, bacterial cellulose has been a focus for
lower than those obtained under biaxial tension, the development of health foods (nata-de-coco), high-
consistent with the fibre alignment mechanism which end acoustic diaphragms, specialty papers, biomedical
is not available under biaxial tension. wound care products and other industrial applications
such as biodegradable food packaging materials (Czaja
et al. 2007; Iguchi et al. 2000; Klemm et al. 2005).
Furthermore, the growth of Ga. xylinus strain ATCC
D. Mikkelsen
M. J. Gidley (&)
Centre for Nutrition and Food Sciences, The University of 53524 in the presence of various plant cell wall
Queensland, Brisbane, QLD 4072, Australia polysaccharides has provided a convenient in vitro
e-mail: m.gidley@uq.edu.au model system to investigate the properties of plant cell
walls (Astley et al. 2001; Chanliaud et al. 2002;
B. A. McKenna
J. B. Wehr
N. W. Menzies
School of Land, Crop and Food Sciences, The University Chanliaud and Gidley 1999; Tokoh et al. 2002; Touzel
of Queensland, Brisbane, QLD 4072, Australia et al. 2003; Whitney et al. 1995).

123
1048 Cellulose (2009) 16:1047–1055

Due to its applications in the biomedical and other deionised water, with frequent changes until the
markets, there is a need for sterilised bacterial water reached pH 7.0.
cellulose; this is typically achieved with boiling
NaOH treatment. At high concentrations, NaOH is a Dry weight and density measurements
harsh treatment and is used to initiate the degradation
of cellulose through the process of mercerization. The Whole pellicles were freeze dried (10 °C,
resulting chemical degradation would be expected to 0.006 mbar) and weighed on an analytical balance.
affect the mechanical properties of bacterial cellu- Dry weights were determined as the average of at
lose. As the unique mechanical properties of bacterial least three replicates. The relative density of a whole
cellulose are highly sought after, any adverse effects pellicle was calculated by dividing the average dry
on mechanical properties should be avoided. weight by the average volume of the hydrated pellicle
A first aim of this study was to characterise the (calculated using the dimensions of the growth vessel
physico-mechanical properties of native bacterial and the thickness of the hydrated pellicle, measured
cellulose produced by Ga. xylinus strain ATCC using digital callipers).
53524 as a function of fermentation time (48, 72,
96 h) and the effect of boiling 0.1 M NaOH on those pH and total viable counts
properties. A second aim of this study was to use
scanning electron microscopy (SEM) to observe the The pH of the medium was recorded for the 48, 72
micro-structure of native and NaOH-treated bacterial and 96 h fermentations using a TPS basic benchtop
cellulose and to identify the dominant mechanism pH meter (TPS Pty Ltd, Springwood, Australia). The
involved in uniaxial tensile deformation and subse- number of viable bacteria in the inoculum and at the
quent failure. A third aim was to reconcile the end of fermentation was determined by the spread
different moduli obtained under uni-axial and bi-axial plate technique. The diluent used was 0.1% peptone
tension, and to identify implications for end-use (pH 5.0) and the plating medium was HS agar.
application. Colonies were counted after 4 days of incubation at
30 °C using an Omron H7EC digital colony counter
(Applethorn, Australia).

Experimental Scanning electron microscopy

Bacterial strain and culture conditions Cellulose samples (from 48, 72 and 96 h fermenta-
tions) were freeze-substituted according to the method
Gluconacetobacter xylinus strain ATCC 53524 from of Wharton (1991). Approximately 1 cm2 sample
the American Type Culture Collection (Manassas, pieces were frozen in liquid nitrogen for 10 s and
VA, USA) was used in this study. The bacterial strain immediately transferred to a solution of 3% glutaral-
was cultivated in Hestrin and Schramm (HS) medium dehyde in methanol at -20 °C for 24 h. Cellulose
containing 5 g L-1 peptone, 5 g L-1 yeast extract, samples were then transferred to methanol (100%,
2.7 g L-1 Na2HPO4, 1.15 g L-1 citric acid and 2% without glutaraldehyde) at -20 °C for a further 24 h,
(w/v) glucose (Hestrin and Schramm 1954). Incuba- removed from the freezer, allowed to warm to room
tions were performed at 30 °C for 48, 72 and 96 h temperature, and dried using a Balzers critical point
under static conditions. The starting pH of the drier. Thereafter, the freeze-substituted samples were
medium was 5.0. At the end of each incubation coated with approximately 10 nm of Pt using an Eiko
period, pellicles were harvested, shaken to dislodge IB-5 sputter coater and examined using a field
some of the embedded cells and subjected to one of emission Scanning Electron Microscope (JEOL JSM
two treatments. In a control treatment pellicles were 6300F) at 6 kV and 3–5 mm working distance.
washed in ice cold triple deionised water under gentle Samples presented in Fig. 4 were taken through the
agitation, with frequent rinses until the pellicles were above process after stretching by tensile deformation
white. In the second treatment pellicles were boiled in had occurred. Samples were processed carefully to
0.1 M (0.4%) NaOH for 30 min then soaked in triple protect the fracture edge (point of failure).

123
Cellulose (2009) 16:1047–1055 1049

Mechanical characterisation each fermentation time and treatment. The effect of

time and NaOH treatment were analysed as a
Mechanical properties of hydrated cellulose pellicles completely randomised design using two-way anal-
were assessed by uniaxial tensile testing using an ysis of variance, calculated using GenStat 7 (GenStat
Instron 5543 (Instron, Melbourne, Australia). Dumb- 2003). Comparisons between means were made using
bell shaped strips of pellicle with dimensions of Fisher’s protected LSD test.
2.639149 individual sample thickness (mm) were cut
using a dumbbell press (ISO 37-4). Thickness was
measured on individual samples using digital calli- Results and discussion
pers. Three dumbbells were cut per pellicle, five
pellicles were tested for each treatment at each time Bacterial growth, cellulose yield and structural
point, totalling 15 replicates per treatment. The two characterisation
ends were placed directly between vice grips, and
moved apart at a constant speed of 10 mm/min. A 5-N At the end of all fermentation periods the media was
load cell was used to record the force required for between pH 5.5 and 5.6, differing from the initial pH
extension as a function of time. From the geometrical of 5.0. The time course of cellulose yield as a function
measurements, force-deformation data were con- of Ga. xylinus ATCC 53524 cell numbers is presented
verted to stress-strain profiles. Engineering stress (r) in Fig. 1 showing that bacterial cellulose yield
was calculated by F/A where A is the area measured increased with fermentation time up to 72 h, then
and F is the force in N. Strain (e) was calculated by remained similar at 96 h despite a major increase in
DL/L0 where DL is exerted extension from the starting bacterial cell count. The difference in dry weights
point L0, and converted to a percentage. The apparent obtained for native bacterial cellulose and NaOH
Young’s modulus was estimated from the average treated bacterial cellulose can be attributed to the
slope over the strain range of 5–15%. removal of alkali-soluble non-cellulosic (bacterial)
material by the NaOH treatment (Nishi et al. 1990)
Statistical analysis and possibly to the removal of some cellulose.
The micro-architecture of cellulose pellicles was
Graphical results in Fig. 1 are presented as means of investigated by SEM. Multiple micrographs were
the triplicates with error bars representing standard obtained for the pellicles from each time point (48, 72
errors of the mean. Figure 2 graphs are representative and 96 h) and from each treatment (washed and
stress/strain curves selected from 15 replicates for NaOH treated), with a representative micrograph

Fig. 1 Cell growth (bars) 1e+10 0.05

and bacterial cellulose yield
Ga. xylinus ATCC 53524 counts

of native (d) and NaOH Bacterial cellulose yield (g)

treated (s) cellulose over 1e+8 0.04
the 96 h experimental
(log10 CFU mL )
-1

period
1e+6 0.03

1e+4 0.02

1e+2 0.01

1e+0 0.00
0 48 72 96
Time (h)

123
1050 Cellulose (2009) 16:1047–1055

Fig. 2 Scanning electron

micrographs illustrating the
micro-architecture of native
cellulose microfibres
produced by Ga. xylinus
strain ATCC 53524 when
cultivated for 48 h (a), 72 h
(b) and 96 h (c); and NaOH
treated cellulose microfibres
harvested after 48 h (d),
72 h (e) and 96 h (f)

presented here (Fig. 2). Micrographs obtained for properties of the pellicle and to determine the effect
these samples revealed a densely packed network of of fermentation time and NaOH boiling treatment on
cellulose fibres with no major differences apparent the mechanical properties. Figure 3 shows represen-
between timescales (Fig. 2). The cellulose fibres tative stress/strain curves of washed and NaOH
within pellicles appeared to be random in orientation boiled bacterial cellulose at three harvest times.
at the micron length scale, with no indication of Overall the shape of the curve is characteristic of a
microfiber directionality (Fig. 2). In the NaOH viscoelastic material that does not exhibit linear
treated samples, occasional damaged (appearing stress/strain behaviour at low deformations. Bacterial
snapped or broken) microfibers were observed, cellulose pellicles, exhibit strains between 10 and
possibly as a result of some cellulose being damaged 20%, before brittle failure. These findings are
by washing during the NaOH treatment. consistent with past investigations into the mechan-
ical properties of bacterial cellulose (Backdahl et al.
Mechanical properties 2006; Chanliaud et al. 2002; Hsieh et al. 2008). The
absence of a linear response at low deformations
Deformation and failure mechanism suggests there is some rearrangement of the network
whilst undergoing deformation.
Uniaxial tensile testing (‘stretching’) of bacterial Individual curves for each of the 15 replicates
cellulose was used to assess the inherent mechanical were variable with LSD 5% values being of the order

123
Cellulose (2009) 16:1047–1055 1051

2.0 showed less, but still high, variability (results not

48 h Washed
shown) indicating some heterogeneity both within
72 h
1.5 96 h and between pellicles. This should be expected for a
Stress (MPa)

biological sample such as a randomly deposited

network of cellulose microfibrils. Large numbers of
1.0 samples (as used in this study) are therefore required
to determine statistically significant effects.
0.5 The bacterial cellulose pellicle consists of highly
entangled, randomly distributed cellulose microfi-
brils, evident by the micrographs in Fig. 2. This
0.0
0 10 20 30 random nature of deposition and extrusion results in
an assortment of levels of aggregation and distances
Strain (%)
between physical cross-links (entanglements) for
2.0
48 h NaOH boiled cellulose microfibrils. Upon deformation, the stress
72 h experienced in any one area is distributed between all
1.5 96 h the microfibrils present. However, regions of the
Stress (MPa)

pellicle that contain chains present in shorter physical

cross-links undergo deformation and failure before
1.0
zones containing longer chains. As deformation
continues, regions containing the longer chains
0.5 eventually reach their limit of extension and failure
occurs. This chain of events offers an explanation for
0.0 the occurrence of double peaks that were often
0 10 20 30 observed in the stress/strain curves (Fig. 3). This
Strain (%) hypothesis was investigated by deforming cellulose at
a much slower rate, 0.5 mm/min (results not shown).
Fig. 3 Representative stress/strain curves obtained by uniaxial This resulted in similar apparent modulus values but
tensile testing of native and NaOH treated bacterial cellulose
harvested at 48, 72 and 96 h. Deformation rate = 10 mm/min
a higher stress before failing, and no double peak.
The lower deformation rate presumably overcomes
of 10% of values for tensile strain and apparent some of the consequences of chain entanglement, by
modulus, and of the order of 20% for tensile stress allowing more time for microfibrils to disentangle
(see Table 1). The three dumbbell cuts per pellicle and slide past each other.

Table 1 Physical and mechanical properties of native and NaOH treated bacterial cellulose
Samples Tensile stress (MPa) Tensile strain (%) Apparent Young’s Relative density (g/cm3)
modulus (MPa)

Native cellulose (48 h) 2.22 ± 0.44c 27.74 ± 4.84b 9.94 ± 1.85a 0.048 ± 0.005
NaOH-treated (48 h) 1.68 ± 0.26b 26.93 ± 4.60b 9.09 ± 1.77a 0.029 ± 0.002
b a b
Native cellulose (72 h) 1.61 ± 0.18 17.49 ± 1.85 14.22 ± 2.25 0.060 ± 0.001
NaOH-treated (72 h) 1.49 ± 0.13b 17.92 ± 2.34a 13.16 ± 1.66b 0.042 ± 0.003
Native cellulose (96 h) 1.23 ± 0.14a 18.72 ± 2.66a 10.12 ± 1.65a 0.055 ± 0.002
NaOH-treated (96 h) 1.11 ± 0.14a 16.65 ± 1.50b 10.27 ± 1.69a 0.037 ± 0.001
LSD (5%) 0.23* 1.74** 1.38***
Within a column, means (±SD) with the same letter are not significantly different at the 5% level
* Effect of both fermentation time and NaOH treated versus native bacterial cellulose
** Effect of fermentation time
*** Effect of fermentation time

123
1052 Cellulose (2009) 16:1047–1055

Astley et al. (2003) using small angle X-ray Modulus values

scattering whilst simultaneously deforming bacterial
cellulose uniaxially, proposed the fibril realignment The reported modulus of individual cellulose micro-
model of bacterial cellulose fibrils undergoing tensile fibrils, ranging from 2.7 to 138 GPa, provides an
deformation. As native bacterial cellulose consists of upper limit of modulus for cellulose fibres within
a random network of microfibrils, upon deformation composites (Hamad 1998; Ishikawa et al. 1997;
the microfibrils reorient quickly, followed by a period Nishino et al. 1995). Hydration and random deposi-
of slow reorientation, until failure occurs and the tion of cellulose within composites markedly reduces
microfibrils relax (Astley et al. 2003). Figure 4 composite modulus compared to fibril modulus
provides a series of micrographs from 48 h native (Vincent 1990). The apparent Young’s moduli of
cellulose that provide the first direct experimental bacterial cellulose pellicles, as determined in this
support for the model proposed by Astley et al. work from the linear stress/strain region under
(2003). Initially the network is randomly deposited uniaxial tensile testing, are similar to those reported
(see Fig. 2). Upon deformation, at the site of stress previously (1–10 MPa) for uniaxial extension (Back-
the microfibrils align in the direction of deformation dahl et al. 2006; Nakayama et al. 2004; Putra et al.
(Fig. 4c) resulting in an increased gradient of the 2008; Svensson et al. 2005).
stress strain curve (Fig. 3). The bulk fractured edge Stress and strain at break was highest for 48 h
shows breaking in tiered formation (Fig. 4a) suggest- fermented bacterial cellulose and decreased with an
ing that (to a limited extent) sheet-like microstruc- increase in fermentation time (Table 1). The mechan-
tures slide past each other at the point of fracture. At ical properties were significantly (p \ 0.05) affected
higher magnification, frayed edge fibres are observed by fermentation time (Table 1). Tensile stress signif-
to be orientated in the direction of deformation icantly decreased with an increase in fermentation
(Fig. 4b). In areas near to the fractured edge, some time. Tensile strain was only significantly higher for
curled fibres (also within the direction of deforma- 48 h bacterial cellulose. It is proposed that mechan-
tion) were observed (Fig. 4d). It is proposed that ical properties are affected by fermentation time due
these fibres have relaxed post fracture. to an increase in heterogeneity within the pellicle

Fig. 4 Scanning electron

micrographs of tensile
fractured surfaces for native
bacterial cellulose harvested
at 48 h illustrating the
overall fractured edge (a), a
close up of aligned fibres at
the fracture edge (b), the
alignment of fibres (c) and
relaxation of fibres (d).
Arrow indicates the
direction of tensile
deformation

123
Cellulose (2009) 16:1047–1055 1053

along with increases in fibre entanglement, leading to the relieving effect of slippage/rearrangement of the
less resistance to deformation. Calculated apparent microstructure that facilitates fibre alignment under
Young’s modulus was significantly (p \ 0.05) higher uniaxial tension.
for 72 h fermented cellulose, there was no significant
difference between 48 and 96 h. It is interesting that NaOH treatment
the highest Young’s modulus was found for the
intermediate fermentation time of 72 h. Whilst it Stress at break, after treatment with NaOH was
might be expected that an increase in cellulose reduced slightly for all pellicles but only significantly
deposition between 48 and 72 h (Fig. 1) would lead for 48 h pellicles (Table 1). A possible explanation
to an increase in modulus, it is less obvious why for this may be a slight increase in swelling, leading
further fermentation to 96 h would lead to a reduced to a decrease in entanglement density and order
modulus when the cellulose content and pellicle within the pellicle (George et al. 2005a). There was
density are similar. We speculate that pellicles no effect on Young’s modulus or strain at break.
fermented for longer periods may form a more There was, however, an interesting difference in bulk
laminated structure and upon deformation these density with NaOH-treated pellicles less dense than
laminated structures may be a possible cause for the their non-treated counterparts. This suggests that the
reduction in modulus of the 96 h samples (Damm- negative charges imparted to cellulose hydroxyl
strom and Gatenholm 2006). The large increase in groups, although not enough to change microfibril
bacterial counts between 72 and 96 h (Fig. 1) may architecture (Fig. 2), were sufficient to cause a
have disrupted the cellulose network, either by direct marked swelling of the pellicle. The reason why
physical presence of bacterial cells or through such a change in bulk density is not reflected in
metabolites secreted during this period where cellu- mechanical measurements, is probably because fluid
lose synthesis is apparently minimal. is readily expressed during the tensile tests (effec-
In this study, values for apparent Young’s modulus tively squeezed out). This means that for tests carried
and tensile stress at failure of ca. 10 and 1 MPa out on samples of such low cellulose concentration,
respectively were obtained. This contrasts with the original bulk density is not a relevant factor after
values for apparent Young’s modulus of between application of strain. The effect of NaOH treatment
300 and 500 MPa and a failure stress of 11 MPa for on the mechanical properties of bacterial cellulose
cellulose from the same bacterial strain measured has been investigated on dried cellulose membranes
under biaxial tension (Chanliaud et al. 2002). On a by George et al. (2005b) and Nishi et al. (1990) To
practical level, this shows that applications of bacte- our knowledge this is the first study to specifically
rial cellulose where stresses are mostly biaxial (e.g. investigate the effects of NaOH on hydrated bacterial
tubular structures) will enjoy a greater deformation cellulose.
modulus and reduced chances of failure compared to Boiling in NaOH solution is a common method for
those applications where uniaxial stresses are dom- sterilisation and has been used in a number of studies
inant. It is also noted that the natural function of plant to treat bacterial cellulose prior to mechanical testing
cell walls in resisting the (biaxial) stresses of cellular (Backdahl et al. 2006; Hsieh et al. 2008; Nakayama
turgor pressure is well served by a cellulose fibre et al. 2004; Putra et al. 2008; Svensson et al. 2005).
assembly that is particularly strong/robust under High levels of NaOH treatment affect inter and intra
biaxial deformation. The rationale for the large molecular H-bonds between adjacent cellulose fibres
difference between uniaxial and biaxial deformation causing fibre swelling, as exploited in commercial
follows from the proposed deformation mechanism of processes such as mercerization. The consequential
fibre alignment (evidenced in Fig. 4). Whilst uniaxial effect on mechanical properties is ambiguous. George
deformation allows fibre rearrangement and ‘pull- et al. (2005a) concluded that 0.1 M NaOH treatment
out’ by linear stretching of entangled fibres, no such reduced mechanical properties compared to native
mechanism is available under biaxial (‘bulging’) bacterial cellulose, Nishi et al. (1990) found that
tension. Modulus and failure stress values under treatment with up to 1.25 M NaOH improved
biaxial tension are proposed to result from direct mechanical properties (cf. Young’s modulus). In this
stretching of entangled cellulose microfibrils without study boiling samples in 0.1 M NaOH for 30 min, did

123
1054 Cellulose (2009) 16:1047–1055

not have a significant (p \ 0.05) effect on either Martin for use of the Instron 5543 instrument. James Riesz, Dr.
tensile strain or Young’s modulus for any of the Grant Edwards and Dr. Polly Burey are thanked for their
helpful discussions and Dr. Peter Kopittke for statistical
fermentation periods (Table 1). Tensile stress was analysis.
significantly (p \ 0.05) affected by NaOH treatment
in 48 h bacterial cellulose only, there was no effect
on 72 and 96 h bacterial cellulose. As such, this is a References
significant finding as it allows for sterilisation and
removal of all non-cellulosic components without Astley OM, Chanliaud E, Donald AM, Gidley MJ (2001)
adversely affecting the inherent and unique mechan- Structure of Acetobacter cellulose composites in the
hydrated state. Int J Biol Macromol 29:193–202. doi:
ical properties of bacterial cellulose. It should be 10.1016/S0141-8130(01)00167-2
noted however, that higher concentrations will most Astley OM, Chanliaud E, Donald AM, Gidley MJ (2003)
likely affect mechanical properties (Nishi et al. Tensile deformation of bacterial cellulose composites. Int
1990). J Biol Macromol 32:28–35. doi:10.1016/S0141-8130(03)
00022-9
Backdahl H, Helenius G, Bodin A, Nannmark U, Johansson
BR, Risberg B, Gatenholm P (2006) Mechanical proper-
ties of bacterial cellulose and interactions with smooth
Conclusions muscle cells. Biomaterials 27:2141–2149. doi:10.1016/
j.biomaterials.2005.10.026
Due to its high mechanical strength and physico- Cannon RE, Anderson SM (1991) Biogenesis of bacterial
chemical properties bacterial cellulose has attracted cellulose. Crit Rev Microbiol 17:435–447. doi:10.3109/
10408419109115207
much interest in various biomedical and industrial Chanliaud E, Gidley MJ (1999) In vitro synthesis and prop-
applications. The effects of fermentation time and erties of pectin/Acetobacter xylinus cellulose composites.
0.1 M NaOH treatment on the inherent mechanical Plant J 20:25–35. doi:10.1046/j.1365-313X.1999.00571.x
properties of hydrated bacterial cellulose were inves- Chanliaud E, Burrows KM, Jeronimidis G, Gidley MJ (2002)
Mechanical properties of primary plant cell wall ana-
tigated. Cellulose yield increased with time. Mechan- logues. Planta 215:989–996. doi:10.1007/S00425-002-
ical stress and strain tended to decrease with 0783-8
fermentation time. Young’s modulus increased at Czaja WK, Young DJ, Kawecki M, Brown RM (2007) The
72 h but then decreased at 96 h fermentation time. future prospects of microbial cellulose in biomedical
applications. Biomacromolecules 8:1–12
Treatment with 0.1 M NaOH treatment did not Dammstrom S, Gatenholm P (2006) Preparation and properties
consistently affect bacterial cellulose mechanical of cellulose/xylan nanocomposites. In: Stokke DD,
properties, with significant effects only seen for Groom LH (eds) Characterization of the cellulosic cell
tensile stress at 48 h and tensile strain at 96 h. Further wall. Wiley-Blackwell, New Jersey, pp 53–64
GenStat (2003) GenStat for Windows. Release 7.2, 7th edn.
studies could be undertaken to determine at what VSN International Ltd, Oxford
concentration of NaOH significant effects are seen in George J, Ramana K, Sabapathy S, Bawa A (2005a) Physico-
mechanical properties. Evidence from electron mechanical properties of chemically treated bacterial
microscopy is consistent with fibre alignment being (Acetobacter xylinum) cellulose membrane. World J
Microbiol Biotechnol 21:1323–1327. doi:10.1007/S11274-
the dominant feature of uniaxial extension for 005-3574-0
bacterial cellulose pellicles. As this mechanism is George J, Ramana KV, Sabapathy SN, Jagannath JH, Bawa AS
not available for biaxial deformation, modulus and (2005b) Characterisation of chemically treated bacterial
fracture stress values are an order of magnitude (Acetobacter xylinus) biopolymer: some thermo-mechan-
ical properties. Int J Biol Macromol 37:189–194. doi:
higher than those measured in this study under 10.1016/j.ijbiomac.2005.10.007
uniaxial tension. These findings are expected to Hamad WY (1998) On the mechanisms of cumulative damage
provide useful baseline information to those employ- and fracture in native cellulose fibres. J Mater Sci Lett
ing bacterial cellulose in novel applications, partic- 17:433–436. doi:10.1023/A:1006555722188
Hestrin S, Schramm M (1954) Synthesis of cellulose by Ace-
ularly those interested in using bacterial cellulose in a tobacter xylinum 2. Preparation of freeze-dried cells
hydrated and sterile state. capable of polymerizing glucose to cellulose. Biochem J
58:345–352
Acknowledgments This research was funded through the Hsieh YC, Yano H, Nogi M, Eichhorn SJ (2008) An estimation
Australian Research Council’s Discovery scheme (DP0665467 of the Young’s modulus of bacterial cellulose filaments.
and DP058067). The authors would like to thank Dr. Darren Cellulose 15:507–513. doi:10.1007/S10570-008-9206-8

123
Cellulose (2009) 16:1047–1055 1055

Iguchi M, Yamanaka S, Budhiono A (2000) Bacterial cellu- with well oriented fibril on PDMS substrate. Polym J
lose—a masterpiece of nature’s arts. J Mater Sci 35:261– 40:137–142. doi:10.1295/polymj.PJ2007180
270. doi:10.1023/A:1004775229149 Ross P, Mayer R, Benziman M (1991) Cellulose biosynthesis
Ishikawa A, Okano T, Sugiyama J (1997) Fine structure and and function in bacteria. Microbiol Rev 55:35–58. doi:
tensile properties of ramie fibres in the crystalline form of 0146-0749/91/010035-24
cellulose I, II, IIII and IVI. Polymer 38:463–468. doi: Svensson A, Nicklasson E, Harrah T, Panilaitis B, Kaplan DL,
S0032-3861(96)00516-2 Brittberg M, Gatenholm P (2005) Bacterial cellulose as a
Jonas R, Farah LF (1998) Production and application of potential scaffold for tissue engineering of cartilage.
microbial cellulose. Polym Degrad Stabil 59:101–106. Biomaterials 26:419–431. doi:10.1016/j.biomaterials.2004.
doi:10.1016/S0141-3910(97)00197-3 02.049
Kersters K, Lisdiyanti P, Komagata K, Swings J (2006) The Tokoh C, Takabe K, Sugiyama J, Fujita M (2002) Cellulose
Family Acetobacteraceae: the Genera Acetobacter, Aci- synthesized by Acetobacter xylinum in the presence of
domonas, Asaia, Gluconacetobacter, Gluconobacter, and plant cell wall polysaccharides. Cellulose 9:65–74. doi:
Kozakia. In The Prokaryotes. Springer, New York, pp 10.1023/A:1015827121927
163–200 Touzel JP, Chabbert B, Monties B, Debeire P, Cathala B
Klemm D, Heublein B, Fink HP, Bohn A (2005) Cellulose: (2003) Synthesis and characterization of dehydrogenation
fascinating biopolymer and sustainable raw material. polymers in Gluconacetobacter xylinus cellulose and
Angew Chem Int Ed 44:3358–3393. doi:10.1002/anie. cellulose/pectin composite. J Agric Food Chem 51:981–
200460587 986. doi:10.1021/jf020200p
Nakayama A, Kakugo A, Gong JP, Osada Y, Takai M, Erata T, Vincent J (1990) Structural biomaterials. Princeton University
Kawano S (2004) High mechanical strength double-net- Press, New Jersey
work hydrogel with bacterial cellulose. Adv Funct Mater Wharton DA (1991) Freeze substitution techniques for pre-
14:1124–1128. doi:10.1002/adfm.200305197 paring nematodes for scanning electron microscopy.
Nishi Y, Uryu M, Yamanaka S, Watanabe K, Kitamura N, J Microsc (Oxf) 164:187–196
Iguchi M, Mitsuhashi S (1990) The structure and Whitney SEC, Brigham JE, Darke AH, Reid JSG, Gidley MJ
mechanical properties of sheets prepared from bacterial (1995) In vitro assembly of cellulose/xyloglucan net-
cellulose 2. Improvement of the mechanical properties of works—ultrastructural and molecular aspects. Plant J
sheets and their applicability to diaphragms of electro- 8:491–504. doi:10.1046/j.1365-313X.1995.8040491.x
acoustic transducers. J Mater Sci 25:2997–3001. doi: Yamada Y, Hoshino K, Ishikawa T (1997) Taxonomic studies
10.1007/BF00584917 of acetic acid bacteria and allied organisms.11. The phy-
Nishino T, Takano K, Nakamae K (1995) Elastic modulus of logeny of acetic acid bacteria based on the partial
the crystalline regions of cellulose polymorphs. J Polymer sequences of 16S ribosomal RNA: the elevation of the
Sci B Polymer Phys 33:1647–1651. doi:0887-6266/95/ subgenus Gluconoacetobacter to the generic level. Biosci
111647-05 Biotechnol Biochem 61:1244–1251
Putra A, Kakugo A, Furukawa H, Gong JP, Osada Y, Uemura
T, Yamamoto M (2008) Production of bacterial cellulose

123