NC State
Vishtal, A., and Retulainen, E. (2014). "Boosting the extensibility potential of fibre networks: A review," BioRes. 9(4), 7951-8001.


Production of paper-based packaging is growing at the present moment and has great future prospects. However, the development of new packaging concepts is creating a demand for an improvement in the mechanical properties of paper. Extensibility is one of these properties. Highly extensible papers have the potential to replace certain kinds of plastics used in packaging. Extensibility is also important for the sack and bag paper grades and for runnability in any converting process. This paper reviews the factors that affect the extensibility of fibres and paper, and discusses opportunities for improving the straining potential of paper and paper-like fibre networks. It is possible to classify factors that affect extensibility into three main categories: fibre structure, interfibre bonding, and structure of the fibre network. Extensibility is also affected by the straining situation and the phase state of the polymers in the cell wall. By understanding the basic phenomena related to the elongation, and by combining different methods affecting the deformability of fibre network, extensibility of paper can be raised to a higher level.

Download PDF

Full Article

Boosting the Extensibility Potential of Fibre Networks: A Review

Alexey Vishtal a,b and Elias Retulainen a,*

Production of paper-based packaging is growing at the present moment and has great future prospects. However, the development of new packaging concepts is creating a demand for an improvement in the mechanical properties of paper. Extensibility is one of these properties. Highly extensible papers have the potential to replace certain kinds of plastics used in packaging. Extensibility is also important for the sack and bag paper grades and for runnability in any converting process. This paper reviews the factors that affect the extensibility of fibres and paper, and discusses opportunities for improving the straining potential of paper and paper-like fibre networks. It is possible to classify factors that affect extensibility into three main categories: fibre structure, interfibre bonding, and structure of the fibre network. Extensibility is also affected by the straining situation and the phase state of the polymers in the cell wall. By understanding the basic phenomena related to the elongation, and by combining different methods affecting the deformability of fibre network, extensibility of paper can be raised to a higher level.

Keywords: Extensibility; Fibres; Bonding; Network; Deformation; Polymers; Papermaking

Contact information: a: VTT Technical Research Centre of Finland , Koivurannantie 1, P.O. Box 1603, FI-40101 Jyväskylä, Finland; b: Tampere University of Technology, Laboratory of Paper Converting and Packaging Technology, P.O. Box 589, FI-33101 Tampere, Finland;

* Corresponding author:


Extensibility is one of the undeservedly disregarded properties of paper. It is considered to be important mainly for sack and bag paper grades (Hernandez and Selke 2001). Typically, papermakers and converters mainly operate with tensile strength, bending stiffness, tear strength, etc. However, recent trends in the development of paper-based packaging materials indicate that, in the production of advanced 3D shapes, extensibility of paper is gaining a new, key role that trumps the importance of other mechanical properties (Kunnari et al. 2007; Östlund et al. 2011; Post et al. 2011; Vishtal et al. 2013; Svensson et al. 2013; Larsson et al. 2014). In 3D-forming processes with fixed paperboard blank, extensibility is the main parameter determining the depth of the shapes produced and the formability of such paper in general (Vishtal et al. 2013). Novel paper-based materials with high extensibility are broadening the area of utilization for paper, and may help to bring the paper industry back onto a growth trajectory in some grade categories.

Despite the importance of extensibility, most of the paper-based materials that are produced at present have poor extensibility. Most paperboards have extensibility in the range of around 1 to 4% and 3 to 6% in MD and CD, respectively. Such values are obviously too low for the development of many new applications. The question is how far and by which means the extensibility potential of paper can be further boosted?

The extensibility of paper has been specifically addressed by only a few researchers (Steenberg 1947, 1949; Brecht and Erfurt 1960; Dodson 1970; Seth 2005). It has been commonly studied together with the other stress-strain properties of paper. Therefore, an extensive summary of the previous fundamental research together with the analysis of the methods for the improvement of extensibility can be a useful addition to the current knowledge in this field.

This paper aims to review extensibility as a mechanical property of paper, and discusses the factors that control it. The influence of the straining situation is also considered. The overarching goal is to find tools to increase the extensibility of paper to a qualitatively new level. The methods that can be used for the improvement of extensibility are comprehensively reviewed, with the focus on the industrial applicability of such methods.


Stress-strain Behaviour of Paper

The extensibility of paper can be defined as the ability of paper to increase its linear length, due to elastic, viscoelastic, and plastic deformations under the action of external mechanical forces. The most widely studied type of mechanical deformation in paper is tensile deformation, where extensibility is determined as the strain at break value of the stress-strain curve (Fig. 1.).

Fig. 1. A typical stress-strain curve for isotropic restrained-dried paper made from softwood pulp

The strain at break value of paper is determined at the point of maximum tension of the stress-strain curve (Fig. 1). Usually, at this point fracture of paper occurs. In this respect, strain at break value is closely associated with the tensile strength of paper, and factors that affect the tensile strength are likely to control the strain at break value as well. However, strong paper may be brittle, while extensible paper usually exhibits a decrease in tensile strength with increase in extensibility.

The extensibility of paper is defined by the fracture point. By considering the stress-strain curve of paper it is possible to suggest that extensibility is dependent on the same factors that are in charge for the shape of the curve and the fracture point of paper. However, the shape of the stress-strain curves may vary to a great extent in accordance with the fibre network properties and the straining conditions (sample dimensions, temperature and moisture content, strain rate, etc.). The influence of the straining conditions can be demonstrated via the stress-strain curve for non-immediate rupture of paper depicted in the Fig. 2.

Fig. 2. A stress-strain curve of paper with a non-immediate rupture (redrawn from Goldschmidt and Wahren 1968)

The development of fracture in paper under tensile stress depends on the amount of elastic strain energy stored in the test sample (area W in Fig. 2), and the amount of energy needed for the propagation of the fracture line to complete the break. If the test specimen is sufficiently long and the elastic strain energy is large, then a brittle, catastrophic failure occurs. If the amount of elastic energy stored in the test strip is low or the energy needed for the propagation of the fracture line is high, for example, due to a short testing span, high fracture toughness or high moisture content, additional energy (area R) has to be delivered into the sample by further straining so as to complete the break (Robertson 1959; Goldschmidt and Wahren 1968; Kurki et al.2004). It can be assumed that, in many converting and forming processes, the apparent testing span is actually shorter than in the standard tensile test.

Fig. 3. A stress-strain curve of rewetted paper (55% dryness) prepared from hydroxypropylated pulp (material produced in Vuoti et al. 2013)

From the example in Fig. 3, we can observe that the fracture of paper does not always occur at the point of maximum load. For example, wet paper can often be strained much further, and only a part of the tension is lost. Such behaviour is close to that of truly ductile materials.

This raises the question of how the extensibility of paper should be defined – at the point of maximum tension or at the point where a significant part of the tension is lost?

This is especially relevant for various converting processes, including 3D-forming. In these cases, extensibility should not be evaluated only through the maximum tensile strength of paper, but rather through the typical loading situation in accordance with the end-use of the paper. For the sack paper grades, extensibility is usually evaluated in connection with TEA (Tensile Energy Adsorption) i.e. gain/loss in strength or strain is compared with gain/loss in TEA.

The deformation of paper is usually divided into elastic, viscoelastic, and plastic components. In forming 3D structures, a large plastic deformation is preferred in order to avoid the springing back and deflexion of shapes (Vishtal and Retulainen 2012). With increase in overall deformation, the extent of all strain components also increases. However, the relative share of plastic deformation increases and the relative share of elastic deformation decreases with increase in overall deformation (Brecht et al. 1971; Skowronski and Robertson 1986). The development of the plastic and elastic components in overall deformation of paper is shown in Fig. 4, where x-axis represents overall strain in %-points and y-axis is for %-points of elastic or plastic deformation. Elastic deformation is defined as the immediately recovered component while plastic deformation component is approximated to be the rest of the deformation.

Fig. 4. The amounts of the apparent elastic and plastic deformations in overall tensile deformation of lightly creped paper for both MD and CD (redrawn from the data of Brecht et al. 1971)

When using only two components, apparent elastic and apparent plastic deformation (which also includes the viscoelastic deformation), it is possible to see that, at overall strains of 2 to 3% points, the amount of apparent plastic and elastic deformations are equal, and after that the apparent plastic component increases linearly. The elastic strain component is present at all deformation levels, but the plastic component starts to appear only at 0.2 to 0.3% strain. At strains higher than 3% the apparent plastic deformation is the dominating component, and it increases linearly. It is also interesting that, despite the different paper structures and different strains at break, the plastic deformation of several papers grades even in machine and cross machine directions was found to follow the same curve (Brecht et al. 1971).

Factors Affecting the Shape of Stress Strain Curve and Final Failure

Qualitatively, the non-linear shape of the stress-strain curve and plastic deformation of paper has been shown to depend mainly on the properties of the fibres in paper. It should be noted that the properties of fibres in paper are different from those of individual fibres because during the papermaking process the fibre properties are modified by wet straining and drying stresses. Ebeling (1976) stated that in well bonded paper “the plastic region in the load/elongation curve is not caused by the breakage of fibre-fibre bonds but is connected to the significant irreversible intra-fibre deformation on molecular and supramolecular level” (Ebeling 1976). The term “efficiency factor” has been used when analysing the shape of the stress strain curve (Seth and Page 1981; Coffin 2009, 2012). When isotropic paper made from a certain fibre material gives higher extensional stiffness, the material has a higher efficiency factor. The efficiency factor is closely related to the load-bearing activity of the material (Lobben 1975). When a fibre network has a higher load-bearing activity, the efficiency factor is also increased and the slope of the stress-strain curve is steeper. Seth and Page (1981) showed with well-bonded sheet that in cases where the slope of stress-strain curve and the initial load bearing activity of the material is changed, for example due to beating, wet pressing, or bonding agent, the stress-strain curves normalised by the efficiency factor (that is calculated from the elastic moduli) can be transposed to an apparently single curve. Bonding plays only a minor role; this is based on the statement that the stress-strain curve of paper is related to the stress strain curve of fibres through the factors that take into account the orientation of fibres and the efficiency of stress distribution of fibres. This suggests that, with a certain pulp, the straining behaviour of the material that actually bears the load is the same, although the amount of the load bearing material is initially different, and only the end point of the stress-strain curve, i.e. the fracture point, varies. The load-bearing activity (and efficiency factor) of the material, however, can change during straining, when the bonding level is low. Then even the stress-strain curves normalised by efficiency factor do not superimpose when transposed (Seth and Page 1981).

Skowronski and Robertson (1983) have concluded that, in addition to elastic, viscoelastic, and plastic behaviour, activation, deactivation, and failure phenomena are also needed in order to explain the tensile behaviour (including stress-strain cycling and relaxation) of paper. Activation of paper under tension can be related to the straightening of fibre segments and re-arrangement of the fibrillar material in the fibre wall, especially in dislocated and microcompressed areas. Straining of wet paper before drying is also known to be an efficient method to activate the load-bearing ability of fibre material (Parsons 1972).

During the straining of paper internal fractures take place. The most fractures are micro-failures and related to debonding. Debonding, i.e. partial or complete fracture of fibre-fibre bonds, is known to take place during straining of paper (Page et al. 1962). However (Ebeling 1976; Seth and Page 1981), the plastic behaviour of paper is not caused by the debonding, but is related to the irreversible intrafibre deformations. Although fibre bonding and debonding have only a minor effect on the shape of the stress-strain curve with well-bonded papers, in less bonded papers the debonding eventually plays a more important role in reducing the efficiency factor and altering the shape of the stress-strain curve. Debonding creates stress concentrations that may lead to the initiation of the final failure (Helle 1965). The first mechanism initiating the final macroscopic failure of paper is either a burst of bond failures or fiber failures (Alava and Niskanen 2006). This conclusion has also gained support from recent stress-strain simulations of paper (Borodulina et al. 2012). Additionally, other factors such as structural heterogeneity or unevenness (bad formation, holes, etc.) can also cause uneven stress distributions and lead to premature initiation of the final fracture (Nazhad et al. 2000).

To conclude this section, it can be claimed that especially in well-bonded sheets the general shape of the stress-strain curve is mainly determined by the properties of fibres in the network and their activity to bear load; but the end point of the curve is also determined by factors that trigger the final failure. Therefore the factors that affect the extensibility of paper are related to properties of fibres, fibre bonds, and network structure.


Numerous factors affect the extensibility of paper. In his excellent review paper Seth (2005) stated that the extensibility of paper depends on two principal components: fibres and interfibre bonding. However, from the point of controlling the extensibility of paper, we should also include the network structure as a separate factor.

It is known that the fibre network and properties of individual fibres in the fibre network are modified by wet straining and drying stresses (Retulainen et al. 1998, Wahlström and Fellers 2000). But factors that modify the fibre network also affect the properties of fibres and bonds. Extensible fibres might be connected by strong bonds, but yet the paper would not be extensible due to restrictions arising from the fibre network structure. This fact is illustrated by conventional papermaking process in which the wet draws, fibre orientation anisotropy, and restrained drying limit the extensibility in MD (machine direction) to the range 1 to 3%, while in CD (cross direction) paper can be strained two times more than in MD. The straining conditions have a major impact on the overall elongation of paper as well as on the share of elastic and plastic components of deformation.

Fig. 5. Factors that affect the extensibility of paper

The scheme in Fig. 5 shows the three factors that affect the extensibility of paper. This scheme emphasizes the role of straining conditions in the overall extensibility potential of paper. Even though these factors are presented individually, they are in close interaction with each other in papermaking and converting processes. For instance, changes in the structure of single fibres due to high-consistency treatment (causing curl, kinks, dislocations, and axial microcompressions) affect the formation of paper and properties of interfibre contacts and the deformation behaviour of the whole network. And, on the other hand, drying shrinkage of the fibre network affects individual bonds and fibres.

In order to boost the extensibility potential of paper to a maximum level, one should design and adjust treatments in such way that they would complement each other.

Properties of Single Fibres that Affect Extensibility

Fibres are the primary constituents and load-bearing elements of paper. They have a strong influence on all the mechanical properties of paper, and its extensibility is not an exclusion from this rule. Wood fibres are generally axially stiff and non-extensible. The typical strain of wood pulp fibres is about 3 to 6%, but juvenile wood fibres may have extensibility up to 20%-points (Hardacker and Brezinski 1973; Bledzki and Gassan, 1999). Also, certain non-wood and synthetic fibres have extensibility varying over a broad range, i.e. 50 to 800%. However, extensible fibres are not necessarily able to form an extensible paper. In order to fulfil this requirement, fibres should be able to form sufficiently strong bonds and network structure with even stress distribution (Seth 2005; Zeng et al. 2013). The bonding potential of fibres depends on morphological, chemical, and mechanical properties. The features of single fibres, which are of high importance for extensibility, are discussed below.

Fibrillar angle

The morphological features of fibres are the key factors determining their mechanical properties. Cellulose, the main chemical component of fibres, is stiff in an axial direction, with a theoretical modulus of the chain around 250 GPa (Vincent 1999). When the individual cellulose chains form a cellulose Iβ crystallite structure, the stiffness decreases to 140 GPa (Cintron et al. 2011). The stiffness is further decreased to around 55 to 65 GPa, when the cellulose Iβ nanostructures are assembled into microfibrils (Sun et al. 2014), and finally to 20 to 40 GPa for fibres (softwood latewood) (Page and El-Hosseiny 1983; Altain 2014). It should be noted that the different cellulose crystalline allomorphs have different stiffness; it is decreasing in the order of Cel I (140 GPa) ˃ Cel II (88 GPa) ˃ Cel III (87GPa) ˃ Cel IV (58GPa) (Nishino et al. 1995).

The decrease in stiffness and the corresponding increase in the extensibility of fibres in comparison with the cellulose molecule is partially attributed to the spring-like alignment of the microfibrils in fibres. This alignment is described by the microfibrillar angle (MFA) that is determined as the angle between the axis of fibre and the direction of the cellulose fibrils in the S2 cell wall layer (Barnett and Bonham 2004) (Fig. 6). The increase in elongation of fibres due to high MFA is explained by the untwisting of the spring-like structure, sliding of fibrils under shear forces, and higher flexibility of such fibres (Horn 1974; Page and El-Hosseiny 1983; Gurnagul et al. 1990; Martinschitz et al. 2008).

Fig. 6. Graphic representation of the MFA in fibre (Hearle 1963)

Fig. 7. The relation between MFA and extensibility of chemical pulp fibres of black spruce (redrawn from the data of Page and El-Hosseiny 1983)

Fibres with high MFA tend to have higher extensibility and lower stiffness than fibres with low MFA (Fig. 7), which is explained by unwinding of the structure of fibres with high MFA under the action of shear forces (Page and El-Hosseiny 1983). Juvenile softwood fibres have higher extensibility than latewood fibres, which is explained by the higher fibrillar angle of juvenile fibres (Wimmer 1992; Lindström et al. 1998; Reiterer et al. 1999; Ljungqvist et al. 2005; Donaldson 2008; Hänninen et al. 2011). For instance, extensibility of latewood fibres of Picea Abies is around 2%-points (MFA 5°), while for springwood fibres it is around 13%-points (MFA 50°) (Reiterer et al. 1998). It was also shown that the fibres with high MFA in Acacia mangium have lower glass transition temperature, which indicates certain differences in the composition and arrangement of wood polymers in the cell in comparison with latewood fibres (Tabet and Aziz 2013).

An additional factor contributing to the higher extensibility of paper made from springwood fibres is the higher longitudinal drying shrinkage (Dong 2009). Although fibres with a high fibril angle have generally lower axial stiffness, they do have better resistance to the shear forces than fibres with a low fibril angle (Satyanarayana et al. 1982; Page and El-Hosseiny 1983; Bledzki and Gassan 1999; Nishino et al. 2004). Interestingly, a trend analogical to the MFA of wood fibres can be observed with viscose fibres, in which the molecular orientation in major extent determines the elongation of the fibres (Lenz et al. 1994).

Fibre fraction with high MFA can be obtained by hydrocyclone separation, which allows separation of springwood and latewood fibre in an efficient way. This treatment was applied and found to be efficient both for softwood (Paavilainen 1992; Vomhoff and Grundström 2003) and hardwood pulps (Blomstedt and Vuorinen 2006). However, this method separates fibres according to their hydrodynamic properties, which are not always correlated with MFA. Utilization of the first-thinning wood (Kärenlampi and Suur-Hamari 1997) or tree species such as Juniperus communis (Hänninen et al. 2011) for pulping is also an option for obtaining extensible fibres with relatively high MFA. Recently, a method for production of cellulose nanofibrils filaments with controlled fibril alignment along the filament axis was proposed (Håkansson et al. 2014). In this case fibrillar alignment is artificially adjusted by process parameters such as flow velocity, flow acceleration and deacceleration, etc., which makes it possible to obtain fibres with desired stress-strain properties. Mechanical properties of the artificial fibres are in line with the natural wood fibres with the same degree of alignment (i.e. MFA) (Håkansson et al. 2014). This is an interesting approach allowing the design of mechanical properties of fibres for a certain applications.

Chemical composition of fibres

Chemical pulp fibres are composed of cellulose and hemicelluloses; lignin is present in very small amounts (less than 0.5%). However, unbleached (2 to 5%) and especially mechanical pulps (5 to 25%) have significantly higher lignin content (Alén 2000). The alternation in the relative share and internal structure of different natural polymers which constitute fibres is of great influence for the mechanical properties of paper (Spiegelberg 1966).

Cellulose is the stiffest chemical component of fibres. Basically, the elongation of cellulose takes place through two mechanisms: by elastic axial elongation of the cellulose molecules and by irreversible, time-dependent slippage between cellulose molecules (Altaner et al. 2014). The axial elongation of the cellulose molecule is, in addition to covalent bonds, also dependent on the intramolecular hydrogen bonds and intermolecular hydrogen bonds. The slippage between molecules is dependent on the intermolecular hydrogen bonds. In fibres, slippage is more likely to occur between fibrils and fibril bundles, which are held together by amorphous cellulose and hemicellulose.

Cellulose in papermaking fibres is present in two states, crystalline and amorphous, with a respective ratio of around 3:1 for bleached wood pulp (Ward 1950; Fiskari et al. 2001). In addition to fully amorphous and crystalline cellulose, the regions with not fully amorphous cellulose can be found, and they are typically regarded as the paracrystalline regions (Kulasinski et al. 2014).

Crystallinity of cellulose in fibres depends to a great extent on the origin and type of the processing utilized for fibres. Increase in the crystallinity of cellulose increases the strength and stiffness of the fibres; at the same time, it negatively affects their extensibility and flexibility (Ward 1950; Lee 1960; Parker 1962; Thygesen 2006). High stiffness and respective low extensibility of crystalline cellulose regions is likely to originate from the O3′H···O5 and O2H···O6′ intermolecular hydrogen bonds and their interaction with the covalent bonds (Altaner et al. 2014). The hydrogen bonds are also responsible for the moisture sensitivity of the cellulose molecules and their extensibility. The mechanical properties and structure of the amorphous cellulose is known to a much smaller extent than that of crystalline cellulose. Generally amorphous cellulose is characterized by the absence of long range order and greater disorder in the orientation of the cellulose chains (Fink et al. 1987; Muller et al. 2000). The stiffness of the crystalline and amorphous parts of the cellulose differs significantly (220 GPa for crystalline vs. 10.4 for amorphous) (Sun et al. 2014). This great difference means that the softer amorphous part mainly determines the extensibility of cellulose (Fig. 8).

An increase in the proportion of amorphous cellulose in pulp is accompanied by an equivalent increase in extensibility and a decrease in elastic modulus, as is schematically depicted in Fig. 8. The same assumption is valid for the regenerated cellulose fibres. For instance, a decrease in the degree of crystallinity of regenerated cellulose fibres (Lyocell®) from 0.63 to 0.5 improves the extensibility of the fibres from 11 to 17% (Lenz et al. 1994). It is important to note that the actual nature of crystallinity is different for cellulose I (native fibres) and cellulose II (regenerated cellulose). In the Cel I cellulose, chains are aligned parallel, meaning that the reducing ends are all facing the same direction. However, upon swelling and dissolution, resulting in the transformation to the Cel II form, they develop an antiparallel arrangement, which is more thermodynamically stable (Kim et al. 2006). The anti-parallel arrangement of the cellulose chains leaves more space for alignment upon straining, and thus explains the higher elongation and lower stiffness of regenerated cellulose in comparison with the native form.

Fig. 8. Schematic representation of the influence of the increase in amorphous cellulose content on the extensibility and strength of pulp relation between fibres (redrawn from the data of Page 1983)

The crystallinity of the cellulose can be reduced by means of several chemical treatments; for example, concentrated acid treatment (Ioelovich 2012), ZnCl2 impregnation (Patil et al. 1965), or ethylamine decrystallization (Parker 1962). Also, the structure of crystalline cellulose can be transferred to a less-ordered one by treatment in water under severe conditions (320°C, 25 MPa) (Deguchi et al. 2006, 2008). Electronic beam irradiation also can be used to reduce crystallinity of cellulose. For instance a dose of irradiation of 200 kGy has been found to reduce crystallinity of MCC, flax, cotton, and viscose by up to 12% (Alberti et al. 2005). At the same time, degradation of hemicelluloses and condensation of lignin is observed (Chung et al. 2012). The amount of energy needed to decrease the crystallinity by a certain value greatly varies in accordance with sample origin, type of the pre-treatment applied, moisture content, etc. (Driscoll et al. 2009). Irradiation has a considerable effect on the structure of cellulose; it causes chain scission and thus decreases DP (Saeman et al. 1952) and oxidizes cellulose by introduction of carboxyl and carbonyl groups (Morin et al. 2004; Bouchard et al. 2006; Henniges et al.2013). Decrease in crystallinity is likely to be caused by substitution of hydroxyl groups with the oxidized ones, along with a consequent weakening of intramolecular and intermolecular hydrogen bonding (Henniges et al. 2013).

However, reducing the cellulose crystallinity cannot be regarded as a feasible method for obtaining of highly-extensible fibres due to the associated costs of chemical treatments and rather poor selectivity in case of irradiation.

Xylans and mannans are the most common hemicelluloses in hardwood and softwood pulp fibres, respectively (Alén 2000). Hemicelluloses are amorphous polymers with a relatively low degree of polymerization (50 to 300) and elastic modulus (7 GPa) and significantly lower softening temperature. This is also reflected in the extensibility of hemicellulose; for instance strain at break of films prepared from arabinoxylan can be as high as 15% (Mikkonen et al. 2012).

Hemicelluloses improve the bonding potential of fibre and thus the extensibility of paper. According to Spiegelberg (1966) and Leopold and McIntosh (1961), high hemicellulose content in chemical pulp fibres is favourable to the extensibility and strength, while Helmerius et al. (2010) have not observed any decrease in the elongation of paper even after removal of 60% of the xylan from birch pulp. Obermanns (1934) in his pioneering work has claimed that there is a certain optimum for hemicellulose content in respect to the strength of paper, which then depends on the origin of pulp. Henriksson et al.(2008) showed that the MFC films with high hemicellulose content had the highest tensile strength and strain, which was attributed to decreased porosity of such films. Hemicellulose removal has also been found to relate to hornification (loss of the swelling ability due to drying) of fibres (Oksanen et al.1997). The extensibility of fibres is often related to their swelling ability (WRV) and the corresponding shrinkage potential of fibres. Hemicellulose-rich pulps have a higher swelling ability. However, hemicellulose removal by hot water extraction has been shown to increase the WRV of fibres and the elongation of paper (Saukkonen et al. 2011). The explanation may be that, in this case, the fibres were not hornified and dried after the hemicellulose removal. Removal of the hemicelluloses makes dried and rewetted fibres stiffer, which results in a reduced amount of fibre-fibre contact and lowers the density of dry paper (Spiegelberg 1966). At the same time, the tensile elastic recovery of alkaline-extracted birch fibres decreases with the removal of xylan, i.e. deformation of fibres tends to be more plastic and come from rearrangements of cellulose microfibrils (Spiegelberg 1966). Cottrall (1954) reported that the mannan is more effective in the strength improvement of pulp than xylan, which is explained by higher amount of available non-acetylated hydroxyl groups per unit of mannose.

However, when it comes to the question of how low or high content of hemicelluloses in fibres affects the extensibility of fibre, there is no straightforward answer. It is likely that the hemicelluloses do not directly influence the elongation of fibres themselves but favour the elongation of paper.

The influence of lignin content on the extensibility of fibres and paper is of concern with mechanical, chemomechanical, semichemical, and unbleached pulps. It was found that selective removal of lignin from wood fibres improved their elongation by around 20%; notably this effect was obtained already when 25% of total lignin was removed from fibre (Zhang et al. 2013). Further delignification did not improve extensibility or the tensile strength of fibre. The effect of lignin removal might be associated with the rearrangement of the microfibrillar structure due to slippage of fibrils in fibres, besides the fact that lignin is actually a stiff and non-extensible polymer in dry state (Zhang et al. 2013). High lignin content also negatively affects the extensibility of paper; mechanical pulps have much lower elongation than chemical pulps (Hatton 1997). In mechanical pulps, lignin is still present in the cell wall of fibres, which also negatively affects the fibre-fibre bonding and flexibility of fibres, and as a consequence the density of fibre network. Unbleached chemical pulp fibres (both kraft and sulphite) are capable of forming fibre-fibre bonds as strong as bleached ones (Mayhood et al. 1962; McIntosh 1963; Fischer et al. 2012). Hartler and Mohlin (1975) claimed that the maximum bond shear strength between fibres occurred at lignin contents of 7% for unbleached kraft and 10 to 12% for unbleached sulphite. In the practice of sulphite cooking, it is sometimes observed that lignin may adsorb on the surface of fibres restricting the formation of effective bonds, and thus negatively affect extensibility (Paasonen and Koivisto 1970; Koljonen et al. 2004). Additionally, the area of bonding and the strength of bonds should be considered. Stiffer fibres tend to make bonds with a smaller area; however, when fibres are pressed together with sufficient force, the specific strength bond can be as high as the more flexible bleached fibre.

Lignin as well as the hemicelluloses and the amorphous part of the cellulose soften under the action of elevated temperature, and the temperature-induced component of extensibility is thus higher in lignin- and hemicellulose-rich pulps (Waterhouse 1985; Salmén 1979; Back and Salmén 1989). However, water is not an effective plasticizer for the kraft lignin and thus it is not softened by moisture alone. The lowest softening temperature of moist lignin is around 80 ºC, and it is reached at around 10% moisture content (Scallan 1974; Kunnari et al. 2007). To summarise, fibres with high lignin content are not likely to demonstrate high extensibility.

Fibre length and fibre strength

The length, microfibrillar angle, strength, and coarseness of individual papermaking fibres have been found to correlate with each other (Karenlampi and Suurhamari 1997). Therefore, these properties may affect extensibility of fibres and the extensibility of paper prepared from such fibres. The effect of the fibre strength on extensibility is much greater than the effect of fibre length (Kärenlampi and Yu 1997). These effects are illustrated in Fig. 9.

The higher the strength of the fibres, the higher its extensibility paper (Fig. 9 “A”). In this case the fibre strength was varied by weakening them by acid vapour treatment. Therefore, this assumption is valid only in the case of strong bonds between weak fibres. In the case of strong fibres, the strength and extensibility of paper would be determined by the strength of the fibre bonds. The influence of fibre length on the extensibility of paper in zero-span test (Fig. 9 “B”) is almost negligible; the strain at break is increased only slightly. But, on the other hand, there is a considerable increase in the work needed to complete the fracture after the stress maximum. The influence of the fibre length on the extensibility of paper is likely to be more evident in the case of wet paper or weakly bonded fibres. The simulation study of Kulachenko and Uesaka (2012) showed that increase in the fibre length from 1.5 to approximately 3 mm doubled the elongation to breakage of the paper.

Mechanical treatments and fibre morphology

Fibres experience mechanical stresses and deformations in many operations on their way from the wood yard to the paper machine (Rauvanto 2010). These stresses may cause both positive and negative changes in fibre morphology, in relation to fibre extensibility (Ljungqvist et al. 2003). High-consistency (ca. 30% dry solids content) mechanical treatments are known to create deformations in fibres. The deformations are local structural changes in the fibre wall and MFA. Visually, they appear in the form of dislocations, microcompressions, curls, twists, folds, kinks, etc.




Fig. 9. The influence of the fibre strength of spruce pulp (measured as zero-span tensile strength of paper, number stands for Nm/g) on the elongation of paper (A) and the influence of the fibre length (number stands for the length in mm) of spruce fibres on its elongation (B) (redrawn from the data of Kärenlampi and Yu 1997).

Microcompressions (telescoping axial buckling along the fiber axis), curls, and dislocations (irregularities in fibres origination from the jams or bends of fibre) contribute to the improvement of extensibility of fibres. They have a definite effect on the elongation of wet paper (Barbe et al. 1984) and they may increase the elongation of dry paper (but they reduce the tensile stiffness and elastic modulus). These deformations occur due to the action of compressive forces oriented in the axial direction of the fibre (Dumbleton 1972; Page et al. 1976; Page and Seth 1980b; Barbe et al. 1984; Mohlin et al. 1996; Gurnagul and Seth 1997; Joutsimo et al. 2005; Seth 2005; Kunnari et al. 2007). Kinks in fibres have not been observed to have any effect on extensibility of paper. However, they impair the tensile strength (Kibblewhite and Kerr 1980). The influence of different fibre deformations on the shape of stress-strain curve of fibre is illustrated in Fig. 10.

Fig. 10. The stress-strain curves of the fibres with different deformation types (redrawn from the data of Page and Seth 1980)

Page and Seth (1980) have stated that gently curled fibres impart high extensibility to paper, maintaining the stiffness. The influence of fibre deformations on the extensibility of paper have been discussed in more detail elsewhere (Mohlin et al. 1996; Seth 2005; Zeng et al. 2012, 2013).

High consistency treatments are known to cause deformations that reduce the straightness and increase the extensibility of single fibres. However, extensible fibres do not necessarily result in improved extensibility of paper, and in addition they usually lead to a decreased elastic modulus and tensile strength of paper (Seth 2005). Low consistency refining straightens fibres and does not increase fibre level extensibility, but induces fibre swelling that increases paper shrinkage, which can have a considerable effect on the extensibility of paper, as shown later.

Combined high- and low-consistency mechanical treatment is a well-known method for improving the extensibility of paper. It unites creation of fibre deformations in the high consistency treatment with the straightening of curled fibres in low-consistency refining that improves the stress transfer ability of the fibre network and promotes bonding between fibres. This combination provides high extensibility to paper, while at the same time maintaining the dewatering resistance of the furnish at low level (Arlov and Hauan 1965; Jackson 1967; Ljungqvist et al. 2005; Sjöberg and Höglund 2007; Pettersson et al. 2007; Gurnagul et al. 2009).

Axial microcompressions can be observed in fibres that still appear to be straight (Fig. 11). Changes in fibre morphology are caused by deformations occurring to some extent in all pulp and papermaking processes in which shear and compressive forces are involved (Forgacs and Mason 1958, 1959; Seth 2005; Salmén and Hornatowska 2014). Industrially produced pulp generally gives paper with higher elongation than laboratory-cooked pulp, due to the higher amount of deformations in the industrially made pulp (Ljungqvist et al. 2003; Duker et al. 2007).

Comparison of different natural and synthetic fibres

Based on the previous discussion, it is clear that chemical and structural properties of fibres have a definite effect on the rupture elongation of the fibres. In natural fibres, the content of cellulose, hemicelluloses, and lignin varies, but what is even more important is that they have a different cell wall structure and dimensions, which may have a detrimental influence on extensibility of fibres.

Synthetic fibres are generally more homogeneous in terms of morphology, but their chemical composition and mechanical properties can vary over a wide range in accordance with the origin of such fibres. The mechanical, structural, and chemical properties of selected natural and synthetic fibres are shown in Table 1.

Table 1. Mechanical, Structural, and Chemical Properties of Selected Natural and Artificial Fibres

Some synthetic (spandex) and natural (coir) fibres have a notably high extensibility potential. These fibres, however, are not capable of creating a strong bonded network due to their limited ability to form hydrogen or covalent bonds. The differences in mechanical behaviour between natural and synthetic fibres are more evident when the stress-strain curves are compared (Fig. 12).

Fig. 11. Variations in the fibrillar orientation due to microcompressions (solid line circle) and dislocations (dashed line circle) in softwood latewood fibres after combined high and low consistency refining. Image taken with polarized light microscopy (courtesy of S. Heinemann, VTT)