Preparation and properties of polypropylene composites reinforced with wheat and flax straw
fibres - Part I : Fibre characterization
Authors: PR Hornsby, E Hinrichsen and K Tarverdi
Wolfson Centre for Materials Processing, Brunel University, Uxbridge, Middlesex, UK
The microstructure, thermal and mechanical properties of flax and wheat straw fibres have been
examined with a view to using these natural fibres as reinforcing additives for thermoplastics. In this
regard, the fibres were characterized prior to incorporation into the polymer, using a range of
techniques, including SEM, image analysis, thermogravimetric analysis and micro-mechanical tensile
testing, at room and elevated temperatures. The thermal and mechanical properties obtained have
been discussed in relation to the measured composition and structural form of the fibres.
For many years, additives of organic origin have been combined with polymers in order to modify
their properties. For example, the presence of starch in thermoplastics can impart biodegradable
characteristics to the material, finding widespread application for this purpose . Wood flour has
long been used as an additive to thermosetting phenolic resin compositions to enhance impact
performance and reduce shrinkage on cure . When contained in thermoplastics, it functions more
as a matrix diluents creating a wood-like effect to the material, but frequently with an adverse effect
on toughness, depending on the extent of interfacial bonding between filler and matrix . There
are also many reports on the use of naturally occurring fibres, including jute, sisal, etc., as reinforcing
fibres for polymers [4-6]. Significant improvements in mechanical properties have been achieved
with the reinforcing efficiency of fibres further enhanced by chemical modification to promote
bonding at the fibre-matrix interface . In this regard, the use of fibres from arable waste sources
offer significant potential as low-cost reinforcements for polymers. Some limited work has already
been reported with such materials in combination with various polymer matrices [7-9]. In this paper,
and in Part II , the use of natural fibres derived from comminuted wheat and flax straw waste as
functional additives for polypropylene is discussed. Factors influencing their effectiveness as
reinforcements are highlighted, with emphasis on the influence of shear and thermal conditions
experienced during melt compounding on the structure and properties of the fibres and also the
effects of interfacial bonding between the fibres and matrix. Part I considers the microstructure,
mechanical and thermal properties of these fibrous additives, prior to incorporation into the
2. Experimental Procedure
Flax fibres were supplied by the Silsoe Research Institute, UK, after retting and decortications
procedures, in lengths of approximately 30-40cm. To facilitate handling, treatment and subsequent
compounding with polymer, the fibres were first pulped by passing through a co-rotating twin-screw
extruder. Wheat straw fibres were obtained in hammer-milled form from Compak Systems Ltd UK,
and were sieved to remove very coarse fibres, as well as particulate components, to give a mean
length of about 2.5cm.
2.2 Measurement of fibre strength
The tensile strength and elastic modulus of flax and wheat straw fibres were measured using a
Rheometrics RSA mechanical analyser, using a static strain sweep programme. This procedure
involved gluing individual fibres to a paper frame with an epoxy adhesive and attaching this to thin
film grips on the RSA analyser. Before testing, the edges of the supporting paper were cut so that no
load was carried by the frame. Specimens were stressed instantaneously to cause failure before
reaching a strain limit of 3%, with data points recorded at each 0.1% strain increment. The cross-
sectional area of single flax fibres was assessed using a calibrated graticule eyepiece on a reflectance
light microscope, taking an average of five measurements. For the purpose of these experiments, it
was assumed that the fibres were perfectly round and of uniform structure throughout their
thickness. In the case of wheat straw samples, sections were cut from the walls of the tubular stem
by slicing in the longitudinal direction and their width and thickness measured using a Vernier screw
In order to establish the extent of possible specimen size effects on fibre properties, samples were
examined at 10, 15 and 20mm gauge lengths. Fibres were also tested at 25, 100 and 200°C within
the temperature-conditioning unit, surrounding the specimens. Samples were exposed to these
temperatures for up to 4 min to allow thermal equilibration before determination of their
Prior to testing, fibres were conditioned at 10%-15% relative humidity (at 23°C) according to BS 3411
2.3 Thermal properties of fibres
The thermal stability of flax and wheat straw fibres was determined using a Perkin-Elmer TGS-2
thermo-gravimetric analyser. Tests were undertaken in both nitrogen and air atmospheres between
30and730°C, at a heating rate of 5°Cmin-1. The arithmetic mean from three determinations was used
to obtain values for moisture level, inorganics content and onset of fibre decomposition. Samples of
pure lignin (Aldrich Chemical Co. Inc.) and α-cellulose (Svenska Traemjoels-Fabrikerna AB) were also
analysed for comparison. Some measurements on the flax and wheat straw fibres were also carried
out under isothermal conditions in an air environment at 190, 210, 230 and 250°C, by heating for 30
min at each temperature.
2.4 Fibre microstructure and chemical composition
Scanning electron microscopy (SEM) was used to examine the microstructure of the natural fibres
before combination with polymer. These were sprinkled on to double-sided adhesive tape attached
to aluminium stubs, then coated with gold. Energy dispersive X-ray analysis was also employed to
obtain information about the chemical composition of the fibres.
3. Results and discussion
A principle aim of this work was to analyse the structure and properties of wheat and flax straw
fibres with a view to assessing their potential as reinforcing additives for polypropylene. Hence
emphasis has been given towards characterizing these fibres prior to compounding with the polymer
in order to establish a basis for examining changes induced by subsequent melt mixing and injection-
A schematic diagram of the cross-section of wheat straw fibres is shown in Fig. 1. The fibrous form of
this material used in this work is itself a composite material, comprising cellulose microfibrils bound
within an amorphous matrix of lignin and hemicellulose. Macroscopically, a straw stem comprises
internodes separated by nodes, which are the hard points at which the leaves are attached to the
Fig. 1: Schematic Diagram of the structure of what straw (after ).
The internodes are formed as concentric rings leaving a void or lumen in the centre. As seen in Fig. 1,
the outermost ring is a cellulose-rich dense layer (termed the epidermis) which has a concentration
of silica on the surface. Beneath this is a layer of lignin containing 25%-27% by volume of
sclerenchyma, or vascular bundles.
Fig. 2 shows the epidermal and inner sclerenchyma structures of wheat straw sections taken from
the wall of the stem. The cellular form of the sclerenchyma and parenchyma is clearly evident in Fig.
2b, and in Fig. 2c the epidermis is observed. The small holes on the surface are stomata through
which the plant cells can breathe.
Fig. 2: Microstructure of wheat straw. (a) General view of fibres. (b) Cellular structure of central
region. (c) Epidermis of fibres.
The structure of linseed flax is very similar to wheat straw but in addition has a pericyclic ring
situated just beneath the epidermis layer containing around 30 bundles of macro-fibres, which are
the commercially important extract used in the textile industry . These fibre cells typically have a
width between 25 and 125µm and a length of 8-30mm. These are joined together in the form of a
chain of fibre cells from the bottom of the stem to the seed-bearing top of the plant, resulting in
bundle lengths of between 300 and 900 mm. The form of linseed flax fibres used in this investigation
is shown in Fig. 3. Some single-fibre cells are apparent as well as several joined in bundles. The fibres
have a polygonal cross-sectional geometry (Fig. 4), but can contain several hundred dislocations
along their length, running transversely across the fibres. These so-called internodes may occur
when the fibre cells unite , but could also result from mechanical damage to the fibre walls.
Intercellular material cementing together individual fibres (Fig. 5) consists of lignin hemicelluloses
and pectin. Fig. 6 shows the end of a single flax-fibre cell demonstrating the relatively dense nature
of this material.
Fig. 3: Linseed flax fibres, as-received.
Fig. 4: Cross-sectional geometry and structural defects in flax fibres.
Fig. 5: Intercellular binding of flax fibres.
Fig. 6: End view of linseed flax fibre.
Linseed flax fibres were also subjected to a pulping treatment prior to introducing them into the
polymer. This involved passing the fibres through a co-rotating intermeshing twin-screw extruder in
the dry state, and causing the formation of shorter fibres, generally being more ribbon-like in shape
(Fig. 7). Considerable amounts of debris also resulted from this process.
Fig. 7: Flax fibres and associated debris after pulping through a twin-screw extruder.
Table I gives an indication of principal elements, other than carbon and hydrogen present in the two
fibre types under study, as determined by microprobe analysis. It is apparent that the wheat straw
contains considerably more silicon (in the form of silica) than the flax, which is located mainly in the
epidermis. Furthermore, the flax straw contained more calcium, but less sulphur and potassium than
the wheat straw.
Table I: Elemental composition of a wheat straw and linseed flax straw fibres.
In the context of the present investigation, the thermal properties of the natural fibres are of
particular relevance, because processing of these materials in combination with polypropylene
requires exposure to temperatures approaching 200°C during both compounding and injection-
Fig. 8 shows isothermal weight-loss curves for flax and wheat straw fibres obtained in air at
increasing temperatures. Weight-loss results are compared in Fig. 9 at an exposure time of 10 min.
These results clearly illustrate that, in air, linseed flax has much greater thermal stability than wheat
straw, although both species exhibit a progressive increase in the rate of degradation as the
exposure temperature is raised. After 30 min heating at 250°C, both fibre types were found to be
extremely friable, reflecting probable structural deterioration.
Fig. 8: Isothermal gravimetric analysis of (a) linseed flax and (b) wheat straw in air: ( ) 190°, (∆)
210°, (◊) 230°, (Ο) 250°.
Fig. 9: The effect of temperature on the weight loss from ( ) flax straw and ( ) wheat straw fibres
(10 min exposure, in air).
Figs 10 and 11 show TGA curves for flax and wheat straw in nitrogen and air environments obtained
at a heating rate of 5°C min-1. Also included are results for lignin and α-cellulose.
Fig. 10: Thermogravimetric analysis of natural fibres and lignin in air, at a heating rate of 5°C min-1.
() α-cellulose, () lignin, ( ) wheat straw, ( )flax.
Fig. 11: Fig. 10: Thermogravimetric analysis of natural fibres and lignin in nitrogen, at a heating rate
of 5°C min-1. ( ) α-cellulose, ( ) lignin, ( ) wheat straw, ( )flax.
Table II summarizes observations from these experiments including the magnitude of the initial
weight loss (attributed to combined water), the residue remaining after heating to 725°C and the
onset of temperature for the main decomposition event.
Table II: Summary of thermal analysis results from natural fibres and lignin.
It is evident that both wheat and flax straw contain a significant level of water, which is released at
temperatures well below that required when melt compounding with polypropylene, confirming the
need to pre-dry these fibres prior to processing, or to employ devolatilization procedures during
compounding to remove evolved steam. The thermal degradation of flax straw is very similar to α-
cellulose in both air and nitrogen atmospheres, whereas the thermal behaviour of wheat straw lies
between lignin and α-cellulose. Hence the lower thermal stability of wheat straw relative to linseed
flax can be attributed to the higher lignin content of the former material.
Thermal decomposition of cellulose has been widely studied on account of its combustibility .
Various products are derived on degradation, in particular levoglucosan, together with gaseous
products, including H2O, CO, CO2, H2 and various volatile saturated and unsaturated hydrocarbons.
Carbonaceous residues are also prominent in the 300-500°C temperature range. When tested in air,
a-cellulose was found to leave no residue at temperatures above 500°C, indicating that all products
of decomposition have been oxidized, where as in nitrogen nearly 10% of carbonaceous residue
remains (Figs 10 and 11). Residue remaining after analysis of flax and wheat straw in air therefore
reflects the extent of inorganic matter in these materials. The value of 7% obtained from wheat
straw is consistent with reported silica levels typically contained in this fibre .
It is necessary to consider the mechanical properties of naturally occurring fibres under conditions
which relate, as closely as possible, to their thermal history experienced during polymer processing.
To this end, property measurements were made at test temperatures of up to 200°C. In addition, the
effect of fibre gauge length on properties was assessed. Results are presented in Tables III and IV
and analysed to assess the statistical significance of differences between means (represented as
percentage error values). Very high standard deviations were observed in all cases, which is to be
expected due to the inhomogeneous form of naturally occurring fibres and the predominance of
defects within their structures. However, it was surprising to find that, in general, results obtained
from wheat straw fibres were less variable than flax straw, because microstructurally the latter
appeared to have a more homogeneous structure. As mentioned earlier, because of difficulties
encountered during measurement of fibre diameter, assumptions were made during the
determination of cross-sectional area, which may also have contributed to the apparent variability of
Table III: Mechanical properties of wheat and flax straw fibres. The effects of temperature and
sample gauge lengths on properties.
Table IV: Mechanical properties of wheat and flax straw fibres. Statistical analysis of mechanical
It can be concluded from these results that under all test conditions, the linseed flax straw was
significantly stronger and stiffer than wheat straw. The tensile strength and elastic modulus of flax
straw decreased noticeably with increasing temperature up to 200°C, which occurred to a lesser
extent with wheat straw. There was no clear relationship between specimen gauge length and
elastic modulus for either fibre type tested; however, the tensile strength of flax straw was found to
be less dependent on gauge length than wheat straw, which increased markedly as gauge length was
Results obtained from the thermogravimetric analysis of these fibres provide some insight into the
relationship between elastic modulus and temperature. Lignin which is known to contribute to the
stiffness of the fibres begins to degrade at around 214°C. Hence heating the fibres to 200°C would be
expected to cause some softening, leading to a lowering in modulus.
The effects of moisture must also be considered. Although the fibres were pre-conditioned in a
desiccators prior to testing, it was expected that they would still contain some constitutional tightly
bound water. Furthermore, due to their hygroscopic nature, some moisture absorption immediately
prior to testing was unavoidable. Hence it can be assumed that fibres tested at 25°C contain some
residual water which, to some extent, will evaporate at 100°C. In the case of flax fibres, it is known
that the presence of water increases stiffness and strength . It was observed that, especially with
flax straw fibres, tensile strength decreased with increasing test temperature. This effect has also
been reported with sisal fibres and was explained in terms of differences in coefficients of thermal
expansion between cellulose, hemicellulose, lignin and pectin . As a consequence, at elevated
temperatures the mismatch in expansion between structure-forming components of the fibre
creates regions of stress concentration and possibly voids or flaws, resulting in weakening of the
fibre. Thermal degradation of fibre components will exacerbate this problem.
There appears to be a distinction between the effects of specimen gauge length on strength and
modulus for the two fibre variants, because results suggest that flax fibres are less sensitive to gauge
length than wheat straw. It is expected that the longer the fibres, the higher will be the probability
of finding defects with a critical size to initiate failure. However, it is known that in flax fibres, the
average cell length is 8-30mm compared to about 1.4mm for wheat straw fibre cells. Hence, in a
given specimen length there are more potential weak points at cell interfaces in wheat than in flax
straw. Furthermore, from microstructural examination it can be seen that the cell wall in flax fibres
contains many microfibrils which would limit the tendency towards total failure under stress from
the presence of surface or even small internal defects.
Curves showing typical force/strain behaviour of the fibres under investigation are shown in Fig. 12.
Many of the flax fibres gave two force peaks before total failure with the slope of the curve towards
the second peak being lower than the first. This is typical of the failure mode of multiphase
composite materials and is a reflection of the layered wall structure within this material.
Fig. 12: Force-strain curves for ( ) flax and ( ) wheat straw fibres.
It was also apparent that force-strain curves for several of the flax samples showed curvature
towards higher force levels with increasing strain (Fig. 13). This phenomenon, which was not seen
with wheat straw, is indicative of strain-hardening, which has also been reported with other natural
fibres, such as coir . It is conceivable that microfibrils present in the cell walls which are aligned
off –axis in the unstrained fibre, progressively become reoriented in the axis of tension as stress is
applied, resulting in increased fibre stiffening.
Fig. 13: Force-strain curve for linseed flax fibre showing strain-hardening effect.
SEM and EDAX have revealed significant microstructural differences between the flax and wheat
straw fibres investigated. The form of wheat straw used in this work appeared much more cellular
than linseed flax, which showed greater fibrous character and a higher aspect ratio. The flax fibres
had a well-defined cross-sectional geometry and contained features resembling dislocations, clearly
evident along their length. The surface region, or epidermis, of the wheat straw was particularly rich
in silica, estimated by TGA to be around 7% by weight.
TGA demonstrated that both fibre types are prone to significant weight loss on exposure to
temperatures up to 200°C. Whilst this is, in part, due to release of adsorbed or constitutionally
bound water, prolonged heating results in structural breakdown and extensive degradation of the
organic phase. Linseed flax was found to be significantly more thermally stable than wheat straw,
which is a consequence of the greater lignin content of the latter.
Micromechanical tests showed that flax was significantly stiffer and stronger than wheat straw,
although these properties declined noticeably with increasing temperature. There was no clear
relationship between gauge length and elastic modulus for either of the fibre types; however, the
tensile strength of wheat straw fibres increased as gauge length was reduced.
1. G.L.G. Griffin and K. Tarverdi in "Proceedings from 5th International Symposium on
Biodeterioration“ Ed. By T.A. Oxley and S. Barry, Aberdeen University, 1981 UK (Wiley-
Interscience, 1983) pp. 557-67.
2. J.A. Brydson “Plastics Materials”, 5th Edn (Butterworths, London, 1989).
3. G.R. Lightsey, P.H. Short and V.K.K. Sinha, Polym. Engng Sci. 17 (1977) 305.
4. S. Varghese, B. Kuriakose, S. Thomas, C.K. Premalatha and A.T. Koshy, Plastics, Rubber
Compos. Process. Applic. 20(1993)93.
5. A.C. Karmaker and G. Hinrichsen, Polym. Plast. Engng 30 (1991) 609.
6. R.G. Raj, B.V. Kokta, G. Grouleau and C. Daneault, Polym. Plast. Technol. Engng 29 (4) (1990)
7. N.M. White and M.P. Ansell, J.Mater.Sci. 18 (1983) 1549.
8. W. Michaeli and P. Hock in “Proceedings from Society of Plastics Engineers ANTEC”, Vol. 39,
New Orleans (Society of Plastic Engineers, Brookfield, CT, 1993) pp. 1332-37.
9. F. Rettenmail, Eur. Pat. 379599 (1990).
10. P.R. Hornsby, E. Hinrichson and K. Tarverdi, J. Mater. Sci. in press.
11. H.G. Gilbertson, Outlook Agric. 19 (4) (1990) 243.
12. R.H. Peters, “Cotton and Flax” in “Textile Chemistry”, Vol. 1 (Elsevier, 1963).
13. R.M. Aseeva and G.E. Zaikov, “Combustion of Polymer Materials” (Hanser Munich, 1985).
14. A.R. Staniforth, “Cereal Straw” (Clarendon Press, Oxford, 1979).
15. T. Foelster and W. Michaeli, Kunststoff e83 (1993) 687.
16. C. Navin and Sarhashmi, J. Mater. Sci. 28 (1993) 6724.
17. A.G. Kulkarni, K.G. Satyanarayana, K. Sukumaran and P.K. Rohatgi, ibid. 16 (1981) 905.
Acknowledgements: The authors are grateful to the EPSRC, DTI and Cookson Plantpak Ltd, for the
support of this work through the LINK Crops for Industrial Use programme.