Page 1
BIOCOMPOSITES BASED ON REGENERATED CELLULOSE FIBER
AND BIO-MATRIX
Sunil Kumar Ramamoorthy, Chanchal Kundu, Kayode Adekunle, Mikael Skrifvars
School of Engineering, University of Boras, Sweden
Abstract
Wood pulp based regenerated cellulose fibers like Lyocell and viscose which are from
natural origin have high and even quality; used to develop superior composites with good
properties. In this project, Lyocell and viscose fibers were reinforced in chemically modified
soybean based bio-matrix, acrylated epoxidized soybean oil (AESO) by compression molding
technique. The composites are characterized for mechanical performance by tensile, flexural
and impact tests, viscoelastic performance by dynamical mechanical thermal analysis (DMTA)
and morphological analysis by scanning electron microscopy (SEM). In general, Lyocell
composites had better tensile and flexural properties than viscose based composites. The same
goes with elastic and viscous response of the composites. Hybrid composites were formed by
fiber blending; on addition of Lyocell to viscose based composites improved the properties. The
amount of Lyocell and viscose fibers used determined the properties of hybrid composites and
the possibility of tailoring properties for specific application was seen. Hybrid composites
showed better impact strength. Morphological analysis showed that the viscose composites had
small fiber pull out whereas Lyocell composites had few pores. Hybrid composite analysis
showed that they had uneven spreading of matrix; delamination occurred on constant heating
and cooling.
To overcome the above mentioned issue and to reduce the water absorption, surface
modification of the fiber was done by alkali treatment and silane treatment. The effect of
treatment is done through swelling, water absorption and morphological analysis tests. The
properties could be increased on proper modification of the fibers. The results show the good
potential of these composites to be used in automotives and construction industries.
Background and Requirements
Researchers developed several biocomposites from natural fibers which are used in
automotives and construction. The potential of these composites is high and the possibility of
replacing synthetic fibers is more. The main drawbacks of the natural fibers being used in
composites are hydrophilic nature and quality variation which is due to plant maturity, place
dependent, harvesting method etc. Few industries like automotives need smell-free products
and this adds to disadvantages of the natural fiber reinforced composites. Several authors have
addressed the problems associated with water absorption by chemical modification of the
surface of the fiber. But the quality variation is difficult to control as various factors come into
consideration.
Page 2
Wood pulp based regenerated cellulose fibers like Lyocell and viscose which are from
natural origin have high and even quality; used to develop superior composites with better
properties.2,3 These types of composites are not explored as much as natural fiber composites.
Some researchers have been trying to explore the possibilities of using these fibers as
reinforcements in structural composites. As the fibers come from natural origin; it would be
beneficial for the environment. Green composites could be formed when these fibers are
reinforced in bio-matrix.
However the processes of producing these regenerated fibers are not completely
environmentally friendly. Ongoing research activities to produce these fibers in environmentally
friendly way have many challenges. It is necessary to have closed loop cycle and minimize the
toxic chemicals.4,5 The same should be done with bio-matrix as small amounts of synthetic
materials are used in producing these resins. The amount of work done on natural fibers
composites is enormous and these fibers are the most environmentally friendly reinforcements.
On considering that these fibers are not produced globally and limited to certain parts; an
alternative is required to avoid transportation and import of these fibers. One of the main
challenges in natural fiber composites is that these fibers give irregular results due to
unevenness of the fibers.
Materials and Methods
Two types of regenerated cellulose fibers such as Lyocell and viscose were used as
reinforcements. Lyocell fibers were supplied by Lenzing AG, Austria and it has specific gravity
1.5 gm/cm3, linear density 1.7 dtex, average fiber length of 38 mm and diameter of 13.4 µm.
Non-woven Lyocell mats were formed by carding and needling as reported by Adekunle1 with a
surface weight of 525 g/m2. The viscose fiber non-woven mats were supplied by Suominen
Nonwovens Ltd, Finland, and they had a surface weight of 60 gm/m2 and a sheet thickness of
0.66 mm. Bio-based thermoset resin AESO (Acrylated Epoxidized Soybean Oil) derived from
soybean oil was used as matrix, it is commercially available as Tribest S350-01 EXP supplied
by Cognis GmbH, Germany. The cross linking initiator (tertiary-butyl peroxy benzoate) was
supplied by Aldrich Chemical Company, Wyoming, IL, USA. Sodium hydroxide pellets, 3-
aminopropyl-triethoxysilane and absolute ethanol were supplied by sigma aldrich.
Fiber Treatments
Alkali Treatment
Regenerated cellulose fibers were pre-dried at 105 ˚C for 2h before immersing in sodium
hydroxide (NaOH) solution with three different concentration, 4wt%, 5 wt% and 6 wt% and
stirred continuously at 25 ˚C. The fibers were treated for 24h, 48h and 72h. After the treatment,
fibers were washed thoroughly with distilled water for neutrality. The pH is checked periodically
using litmus paper. Then the fibers were dried in room temperature for 4h followed by oven
drying for 3h at 105 ˚C. The similar treatment was done at 50 ˚C.
Page 3
Silane Treatment
APTES (3-aminopropyl-triethoxysilane) was used as silane coupling agent to treat the
regenerated cellulose fibers. APTES was added to ethanol-water mixture (8/2 volume ratio) to
make three different concentration, 2wt%, 4wt% and 6wt%. Pre-dried regenerated cellulose
fibers (105 ˚C, 2h) were immersed in three different concentration solutions and stirred
continuously at 25 ˚C. The treatment was done for three time intervals, 24h, 48h and 72h.
Fibers were washed thoroughly with distilled water after treatment and pH is checked for
neutrality. Then the fibers were dried in room temperature for 4h followed by oven drying for 3h
at 105 ˚C. The similar treatment was done at 50 ˚C.
Composite Preparation
Acrylated epoxidized soybean oil was used as a matrix to make composites. Viscosity of
resin is the main factor influencing fiber impregnation. The viscosity of AESO was reduced by
heating in oven at 60˚C for 5 minutes. AESO was then blended with initiator for high
temperature curing. Tert-butyl peroxybenzoate was used as a free radical initiator (2 wt %) and
was mixed well with AESO to give a homogeneous solution.
The treated and untreated fiber mats were cut to 20cm×20cm dimension. The fibers were
impregnated with the blended resin and the fiber-resin ratio was taken as weight fraction.
Composites were made with different weight fraction of regenerated cellulose fiber by
compression molding. This method was adopted to fabricate the composites. Curing was done
with heat and pressure for 5 min. Pressure (40 bar) was used to make composites and at a
temperature between 160oC to 170oC on hot press from Rondol Technology, Staffordshire, UK.
Specimens were cut according to ISO standard by using laser cutting technology (GCC
LaserPro Spirit).
Characterization
Mechanical performance is characterized by tensile, flexural and impact tests. Dynamic
mechanical analysis was done see the viscoelastic properties. Water absorption test was done
to see the hydrophilic nature of the composites. SEM images were made to see analyze the
morphological properties.
Tensile Test
The tensile test was carried out based on ISO 527 using universal Tinius Olsen H10KT
testing machine and QMat software. The rate of loading was 10 mm/min and the load range
was 5 kN and 10 kN. Atleast 10 specimens were analyzed for each sample. The dumb-bell
shaped specimens were cut from the laminates with laser cutting machine with overall length of
150 mm which includes parallel-sided portion (60 mm). The widths at ends and at parallel-side
are 20 mm and 10 mm respectively. Gauge length was 50 mm and initial distance between
grips was 115 mm.
Page 4
Flexural Test
The three point flexural test was adopted to determine the flexural stress (σf) and flexural
modulus (Ef). The tests were performed based on ISO 14125 standard using a Tinius Olsen
H10K-T UTM (universal testing machine). At least 5 specimens were tested for every sample.
The specimen dimension was 80×15mm (length×width) while the thickness varied depending on
the sample. The load range was 5 kN and 10 kN and the rate of loading was 10 mm/min. The
outer span was 64 mm and the displacement range was 10 mm.
Impact Test
This test is performed based on ISO 179 to obtain the charpy impact strength of un-notched
specimens. Atleast 10 specimens were tested for each sample using Zwick test instrument, and
mean impact resistance was determined. The specimens were tested flatwise.
Charpy impact strength = [energy absorbed/cross-sectional area]
Dynamic Mechanical Thermal Analysis
Dynamic mechanical thermal analysis (DMTA) was done to determine the viscoelastic
properties of the composites with Q series TA instrument supplied by Waters LLC, Newcastle,
DE, USA. Dual cantilever clamp was used to mount the specimens. The specimen’s dimension
was 35 × 10 × 2 mm3 and the temperature ranges from 30˚C to 150˚C at frequency 1 Hz.
Water Absorption Test
Water absorption test was carried out on the samples to determine the dimensional stability
of the composites. Four specimens were examined for each sample and the average was taken.
The specimen dimension was approximately 36×12 mm. The specimens were dried in an oven
for 24 hr at 60oC. Then the specimens were kept in desiccators in order to cool down to room
temperature and the weights of these specimens were denoted as Wo. Then specimens were
immersed in distilled water at room temperature. The amount of water absorbed was measured
every 24 hours for 10 days. The specimen was taken out of the water each time, and the
surface wiped dry and weight recorded as W. The percentage water absorption (WA %) was
then calculated using the formula below.
WA% = [(W-Wo)/Wo] × 100
Scanning Electron Microscopy
Morphological analysis was done by studying the cross section of the fractured specimens
by environmental scanning electron microscope (ESEM), FEI Quanta 200 F. The equipment
was run at low vacuum and high voltage (5-10 Kv). This is done to see the fiber-matrix interface
and pores.
Page 5
Swelling
Treated fibers were subjected to swelling measurement. Swelling of the fibers were noticed
by measuring the diameter of the fibers through microscope. Fibers were treated for 30 minutes
before measuring the diameter. Several measurements were made and mean was taken.
Weight Loss
Weight loss was checked for fibers treated for 30 minutes, 12 hours and 24 hours. Dried
fibers were weighed and noted. Then fibers were treated to respective time and dried before
checking the weight loss.
Results and Discussion
Tensile Test
Table 1 shows that the tensile strength of Lyocell fiber reinforced composites was higher
than the viscose fiber reinforced composites and the hybrid composites, which indicated that the
Lyocell reinforced composite was the toughest and strongest. Composite consisting 60 wt%
Lyocell fiber had tensile strength of 135 Mpa. Whereas for the same fiber content, the tensile
strength was approximately 96 MPa and 117 MPa in the viscose fiber reinforced composite and
hybrid composite respectively. This is because Lyocell fiber on its own has higher tensile
strength (750 MPa) compared to viscose fiber (310 Mpa). The general trend of an improvement
of tensile strength of all composites was associated on increasing Lyocell fiber. For Lyocell fiber
reinforced composite, the tensile strength increased from 113 MPa to 135 MPa with 20% fiber
increase and the same trend was observed for viscose fiber reinforced and hybrid composite.
The uniformity in the results could be due to the consistency in impregnation, curing condition,
sheets alignment (all fiber sheets were aligned in 00).
The tensile moduli of the Lyocell and viscose fiber composites increased with an increase in
fiber content from 40 wt% to 50 wt%, but surprisingly there was no increment in the modulus
when the fiber content was increased from 50 wt% to 60 wt% in the hybrid composite. An
interesting trend was observed in the hybrid composite when 20 wt% of Lyocell was hybridized
with 30 wt% viscose fiber because the tensile modulus increased, whereas the tensile strength
was lower in the hybrid composite with 25 wt% Lyocell and 25 wt% viscose.
The percentage elongation was quite good for viscose fiber reinforced composites which
was approximately 2.5 %. This was expected due to the higher percentage elongation of
viscose. For Lyocell fiber reinforced composites, it was less than 2% irrespective of fiber
content. It was expected, due to the morphology of regenerated cellulose fiber. For the hybrid
composites, the values were more or less in between 2 to 2, 5 % for all compositions.
Page 6
Table 1: Tensile Properties of the Composites
Product Composition
Tensile Strength (MPa)
Tensile Modulus (GPa) Elongation %
Lyocell/Resin-40/60 113.3 14.17 1.95
Lyocell/Resin-50/50 124.15 16.77 1.74
Lyocell/Resin-60/40 135.42 16.64 1.89
Viscose/Resin-40/60 77.48 9.39 2.44
Viscose/Resin-50/50 91.6 12.68 2.56
Viscose/Resin-60/40 95.94 12.1 2.6
Lyocell/Viscose/Resin-20/20/60 94.22 11.28 2.31
Lyocell/Viscose/Resin-25/25/50 99.9 11.24 2.27
Lyocell/Viscose/Resin-20/30/50 111.42 13.17 2.15
Lyocell/Viscose/Resin-30/30/40 117.07 15.07 2.5
Flexural Test
The hybrid composite had the highest flexural strength compared to Lyocell and viscose
fiber reinforced composites. Composite consisting 60 wt% fiber content (30 wt% Lyocell & 30
wt% viscose), the flexural strength was approximately 140 Mpa; whereas the flexural strength
was 127 MPa and 92 MPa for Lyocell and viscose fiber reinforced composites respectively with
same fiber content (Table 2). So, the effect of hybridization was quite significant in case of
flexural strength of composites. For the viscose fiber reinforced composite, the flexural strength
followed a downward trend with the increase of fiber content; it reduced from 101 MPa to 92
MPa for 20 wt% fiber increase. But for other cases, there was noticeable increase in flexural
strength with fiber content increase.
The Lyocell-reinforced composite had the highest flexural modulus of about 7 GPa for 60
wt% fiber content. Hybridized composite and viscose fiber reinforced composite had flexural
modulus of 6 GPa and 5 GPa respectively for the same fiber content. However the effect of
hybridization was negligible, increasing the viscose fiber content did not have effect on the
flexural modulus of the composite.
Page 7
Table 2: Flexural Properties of the Composites
Product Composition
Flexural Strength (MPa) Flexural Modulus (GPa)
Lyocell/Resin-40/60 108.49 6.46
Lyocell/Resin-50/50 118.81 6.17
Lyocell/Resin-60/40 127.04 6.79
Viscose/Resin-40/60 101.18 5.16
Viscose/Resin-50/50 98.29 5.2
Viscose/Resin-60/40 92.92 5.28
Lyocell/Viscose/Resin-20/20/60 91.83 4.48
Lyocell/Viscose/Resin-25/25/50 107.78 5.47
Lyocell/Viscose/Resin-20/30/50 124.83 5.54
Lyocell/Viscose/Resin-30/30/40 140.08 5.96
Impact Test
Table 3 shows the impact resistance properties of the composites which indicates the
amount of energy absorbed in the cross sectional area of a material. The Lyocell fiber
composites showed impact resistance between 40 and 50 kJ/m2. Viscose and the hybrid
composites have impact resistance between 45 and 50 kJ/m2. It was noticed that the increase in
fiber content imparted higher impact resistance in composites materials. It resulted in longer
average fiber pull-out lengths, and therefore caused higher impact strength. In contrast, higher
fiber-matrix adhesion results in shorter average pull-out lengths and make the material brittle
and that ultimately induce lower impact resistance in the material. However these results are
expected because natural fibers are quite different from regenerated cellulose fibers due to their
morphology. In previous studies the impact behavior of regenerated cellulose fiber reinforced
composites is quite different from natural fiber reinforced composites.
Page 8
Table 3: Impact Properties of the Composites
Product Composition Impact Strength (kJ/m2)
Lyocell/Resin-40/60 42.35
Lyocell/Resin-50/50 38.92
Lyocell/Resin-60/40 49.9
Viscose/Resin-40/60 44.69
Viscose/Resin-50/50 47.95
Viscose/Resin-60/40 45.14
Lyocell/Viscose/Resin-20/20/60 35
Lyocell/Viscose/Resin-20/30/50 45.93
Lyocell/Viscose/Resin-25/25/50 44.14
Lyocell/Viscose/Resin-30/30/40 52.23
Dynamic Mechanical Thermal Analysis
Table 4, The storage modulus (E') is a measure of elastic response of a material, and
Lyocell fiber composite had the highest storage modulus and variation in fiber content wt% had
effect on results. Similar trend was observed for viscose fiber reinforced composites and the
hybrid composites. The hybrid composites seemed to have the least storage modulus which
could be due to delamination during constant heating and deformation for about 1 hour in the
equipment and the possibility of mismatch in the hybrid composite structure. In this case,
Lyocell fiber and viscose fiber were combined, so a micro-structural analysis of a transverse
section of the specimen could give a better explanation.
Page 9
Table 4: Viscoelastic Properties of the Composites
Product Composition E' (MPa) E" (MPa) Tg tanδ
Lyocell/Resin-40/60 4805.33 233.23 82.86 0.049
Lyocell/Resin-50/50 3956.33 207.5 83.95 0.052
Lyocell/Resin-60/40 4672 237.8 82.5 0.051
Viscose/Resin-40/60 3761.67 196 82.38 0.052
Viscose/Resin-50/50 3599.33 186.97 81.95 0.052
Viscose/Resin-60/40 3542.33 176.27 81.22 0.05
Lyocell/Viscose/Resin-20/20/60 3428.33 201.17 81.32 0.059
Lyocell/Viscose/Resin-25/25/50 2885.33 162.57 81.25 0.056
Lyocell/Viscose/Resin-20/30/50 3252 189.3 84.87 0.058
Lyocell/Viscose/Resin-30/30/40 3386.67 177.57 85.41 0.052
The loss modulus indicates a materials response to a viscous behavior. Loss modulus of
Lyocell fiber composites were superior and was between 200 to 250 MPa for different fiber
content. So, from that perspective the Lyocell fiber composites exhibit the best viscoelastic
properties compared to others. For viscose fiber reinforced composites, the loss modulus was
nearly 200 Mpa.
The Tg values were measured from the tanδ curve and it was in between 800C to 850C for
Lyocell and viscose fiber reinforced composites. But it was a bit higher in hybrid composites with
50 wt% and 60 wt% fiber content. Though the value of Tg will vary depending on which
parameter used to detect the transition. It also depends on the experimental parameters such
as frequency of oscillation, temperature ramp rate and sample dimensions and it is expected
that Tg should be measured on a material which is not under mechanical stress.
The structural or material damping of a composite material could be analyzed using DMTA.
Tanδ is the ratio of the loss modulus (E") to storage modulus (E') or it could be defined as the
ratio of the energy lost to the energy retained during a loading cycle. And the values of tanδ
were measured at 350C in this study. The most significant result was obtained from the hybrid
composites with a tanδ of approximately 0.06. This result indicates that the hybridization has
optimized the good structural damping properties in the composite materials and that could be
considered for automotive application.
Page 10
Scanning Electron Microscope
Alkali and silane treatments made the surface of fiber rougher. At higher concentration or
longer treatment exposure, treatment had adverse affects due to fibrillation. Fibrillation occurred
also at increased treatment temperature.
Figure 1: Untreated Fiber Figure 2: Alkali-treated
Figure 3: Fibrillated Fiber(at higher Concentration Figure 4: Fibrillated Fiber(at longer time)
Figure 5: Dispersed Fiber in Composite
Page 11
Figure 1 shows the SEM image of untreated fiber which has smoother surface than alkali
treated fiber, figure 2. Rougher surface in figure 2 helps in improving the properties. Optimizing
the treatment temperature, time and concentration is important as higher treatment conditions
gives adverse effects like fibrillation. Figure 3 and 4 shows the fibrillation in fibers which has
extreme treatment conditions. Figure 5 shows the dispersion of fibers in matrix.
Figure 6b shows the pores in composites and this could affect the properties of the
composites. This should be addressed to have better composites. Uneven spreading of matrix
was seen in hybrid composites which could be due to hybrid structure.
Figure 6. SEM images ; (a) Viscose-matrix interface, (b) Lyocell-matrix interface, (c)
Viscose fiber pull-out in hybrid specimen and (d) Hybrid composite specimen
Page 12
Swelling
Swelling increased on increasing the concentration of treatments and reaches highest
swelling at 13 wt% concentration. This is seen in both alkali and silane treatments, 51.8% and
9.4% were corresponding swelling %. This is quite similar with natural fibers as alkali treatment
gives higher swelling.
Table 5: Swelling and Weight Loss % of the Composites
Weight Loss
Weight loss is seen after treatment of fibers on different treatment concentration for different
treatment time at room temperature. Highest weight loss was noticed at 10 wt% for alkali and
silane treatments, table 5.
Summary and Next Step
Composites made out of regenerated cellulose fibers had good properties. Treatments
improved the surface roughness. Swelling and weight loss was noticed on various treatment
concentrations. It was noticed that treatment time, temperature and concentration affects the
fiber. It is necessary to improve the properties by reducing the pores and have optimized
treatments. It is also required to look into hybrid structure to have better properties.
References
1. Adekunle KF. Bio-based Composites from Soybean Oil Thermosets and Natural Fibers. PhD Thesis,
Department of chemical and biological engineering, Chalmers University of Technology, Sweden,
(2011).
2. George L. Handbook of composites. 2nd
ed. Chapman & Hall, (1998).
3. Allin SB. Polymer Science and Technology, 2nd Edition (Joel R. Fried). Journal of Chemical
Education (2004); 8: 809.
4. Borbély É. Lyocell, The new generation of regenerated cellulose. Acta Polytechnica Hungarica
(2008); 5: 11-18.
5. Carrillo F, Colom X and Cañavate X. Properties of regenerated cellulose lyocell fiber-reinforced
composites. Journal of Reinforced Plastics and Composites (2010); 29: 359-371.