NC State
BioResources
Ketola, A., Strand, A., Sundberg, A., Kouko, J., Oksanen, A., Salminen, K., Fu, S., and Retulainen, E. (2018). "Effect of micro- and nanofibrillated cellulose on the drying shrinkage, extensibility, and strength of fibre networks," BioRes. 13(3), 5319-5342.

Abstract

Elongation is an important property of many packaging board and paper grades. Paper with high extensibility could provide an alternative for oil-based packaging materials. Micro- (CMF) and nanofibrillated (CNF) cellulose are known to increase the strength of a paper, but their effect on the drying shrinkage and elongation is not well-studied. In this work, paper was reinforced with fibrillated material. Added fibrillated material increased the drying shrinkage, which was generally proportional to the increase of paper elongation before breakage. Results differed depending on the fibrillated material and how it was added to paper (wet-end addition or spray application). The papers were dried unrestrained in order to achieve the highest elongation potential for the paper. Spray application of CMF increased elongation by 13%, while wet-end additions increased elongation by 20% and also strength by 10%, but only with high dosages. Spray application of oxidized-CNF improved elongation by 33%, while wet-end applications increased only strength by 20%. Thus, boosting the drying shrinkage with fibrillated cellulose is one potential way to increase elongation and 3D formability of paper.

 


Download PDF

Full Article

Effect of Micro- and Nanofibrillated Cellulose on the Drying Shrinkage, Extensibility, and Strength of Fibre Networks

Annika E. Ketola,a Anders Strand,b Anna Sundberg,b Jarmo Kouko,a Antti Oksanen,a Kristian Salminen,a Shiyu Fu,c and Elias Retulainen a

Elongation is an important property of many packaging board and paper grades. Paper with high extensibility could provide an alternative for oil-based packaging materials. Micro- (CMF) and nanofibrillated (CNF) cellulose are known to increase the strength of a paper, but their effect on the drying shrinkage and elongation is not well-studied. In this work, paper was reinforced with fibrillated material. Added fibrillated material increased the drying shrinkage, which was generally proportional to the increase of paper elongation before breakage. Results differed depending on the fibrillated material and how it was added to paper (wet-end addition or spray application). The papers were dried unrestrained in order to achieve the highest elongation potential for the paper. Spray application of CMF increased elongation by 13%, while wet-end additions increased elongation by 20% and also strength by 10%, but only with high dosages. Spray application of oxidized-CNF improved elongation by 33%, while wet-end applications increased only strength by 20%. Thus, boosting the drying shrinkage with fibrillated cellulose is one potential way to increase elongation and 3D formability of paper.

Keywords: Elongation; Drying shrinkage; Tensile strength; Fibre network; Extensibility; Fibre surface; Fibre-fibre joints; Bio-based products; Fibrillated cellulose; Packaging paper

Contact information: a: VTT Technical Research Centre of Finland Ltd, P.O. Box 1603, FI-40101 Jyväskylä, Finland; b: Åbo Akademi University, Johan Gadolin Process Chemistry Center, Porthansgatan 3, FI-20500 Åbo/Turku, Finland; c: State Key Laboratory of Pulp and Paper Engineering, South China University of Technology, Guangzhou, Guangdong 510640, China;

*Corresponding author: elias.retulainen@vtt.fi

INTRODUCTION

Plastics are widely used in many packaging applications, but the growing problems with disposal and waste handling have raised a need for alternative, more environmentally friendly materials such as cellulose fibre-based biomaterials. In order to compete with oil-based packaging materials, the fibre-based materials must have improved properties such as a more extensible and formable structure. Cellulose is inherently a very strong, stiff, and crystalline material, which limits the extensibility and formability properties of fibres (Ward 1950; Thygesen 2006).

However, extensibility can be affected by modifying properties of individual fibres through mechanical refining, which introduces deformations, internal and external fibrillation in fibres affecting their flexibility, tensile stiffness, shrinkage, and bonding behaviour (Jackson 1967; Mohlin et al. 1996; Seth 2005; Zeng et al. 2013; Khakalo et al. 2016). Fibre network properties are affected by stresses during wet straining and the drying process (Retulainen et al. 1998; Wahlström and Fellers 2000), and free shrinkage of paper during drying is known to promote the fibre network’s elongation properties (Fujiwara 1956; Page 1971; Htun and de Ruvo 1981; Htun et al. 1989; Waller and Singhal 1999). Elevated temperature and moisture can be used to soften the fibres, which decreases stiffness and enhances the elongation potential of paper (Goring 1963; Salmén and Back 1977; Back and Salmén 1982; Caulfield 1990; Haslach 2000).

In addition to fibre properties, the fibre-fibre interactions in the fibre network have important effects on the mechanical properties, extensibility, and mouldability of paper (Borodulina et al. 2012). Hydrogen bonding and van der Waals forces are responsible for the adhesion-related interactions between fibres (Persson et al. 2013). The fibre-fibre interactions can be affected by additives, such as fines, starch, and polyelectrolytes. Generally, additives improve the adhesion of fibres (Pelton 1993). Fines generally have a positive effect on strength and elongation properties of paper (Retulainen et al. 1993; Luukko 1998; Seth 2003; Taipale et al. 2010).

In recent years, micro-, and nanofibrillated materials have been under intensive research due to their interesting properties, including a large specific surface area and the ability to bind large amounts of water and form a partly gel-like structure. Micro- and nanofibrillated materials are produced by mechanically delaminating and fibrillating cellulose fibres (Herrick et al. 1983; Turbak et al. 1983; Heux et al. 1999; Andersen et al. 2006). Fibres can be also pre-treated chemically (Saito and Isogai 2004) or enzymatically (Pääkkö et al. 2007) before homogenisation, leading to finer and more homogeneous material. Several studies have been done where fibrillated material has been added in a paper furnish to improve a paper’s mechanical properties (Ahola et al. 2008; Eriksen et al. 2008; Taipale et al. 2010; Gonzales et al. 2012; Hassan et al. 2015; Tarrés et al. 2016). In all of these studies, fibrillated material increased the density, strength, and strain of the paper, as well as reducing its porosity and air permeability.

Ahola et al. (2008) tested a bi-layer system of cellulose nanofibrils and cationic polyelectrolyte in bleached pine kraft pulp, resulting in a significant increase in both wet and dry tensile strength of paper. Eriksen et al. (2008) studied microfibrillated cellulose in TMP paper and found increases in tensile index and air resistance. Taipale et al. (2010) observed that microfibrillated cellulose added to kraft pulp increased the tensile strength of a paper but also decreased the drainage rate. However, by adjusting the pH and conductivity of the wet-environment in an optimal way, the strength properties could be improved without a massive decrease in drainage. Tarrés et al. (2016) observed a significant increase of paper strength when they first added enzymatic cellulose nanofibres in bulk suspension and then TEMPO-oxidised nanofibres as a coating. This kind of combination is a good alternative for CNF bulk addition without the negative effects in drainage rates, and the amounts of needed chemically modified CNF could be reduced.

Many studies have used CNF as an alternative to excessive beating of pulp (Gonzales et al. 2012; Delgado-Aguilar et al. 2015; Hassan et al. 2015; Tarrés et al. 2016). In addition to fibrillated material as a paper additive, CNF can be used in other packaging applications (Hubbe et al. 2017), such as paper coatings (Fukuzumi et al. 2009), printing (Hamada and Bousfield 2010), intelligent packaging (Bardet and Bras 2014), and composites (Hult et al. 2010; Ridgway and Gane 2012).

While fibrillated material has advantages as a paper additive, it also has drawbacks. In addition to the decreased drainage rate, the production of fibrillated cellulose requires energy, hazardous chemicals, and highly sophisticated equipment (Saito and Isogai 2004; Pääkkö 2007; Eriksen et al. 2008), and it can be unaffordable for industrial production (Delgado-Aguilar et al. 2015; Espinosa et al. 2016). However, intensive research continues in this field for finding more environmentally friendly and efficient production and application methods for micro- and nanofibrillated material.

In previous studies the effects of fibrillated material have been evaluated using restrained dried sheets. However, in order to find the most cost-effective approach for enhancement of paper properties, especially on the extensibility of paper, the effects of micro- and nanofibrillated cellulose on the paper mechanical properties, should be evaluated using unrestrained drying. The application of microfibrillated cellulose may offer a more cost-efficient alternative to nanofibrillated material.

In this study, cellulose microfibrils (CMF) and TEMPO-oxidized cellulose nanofibrils (oxidized-CNF) were used to modify the fibre network and fibre-fibre interaction, and determine the effects on paper elongation and strength. Two different techniques for addition of CMF and oxidized-CNF were used: wet-end addition directly to the pulp suspension, and spray application onto wet sheets. The used pulp was refined to SR 25, and the sheets were dried unrestrained to determine the elongation potential of the fibre network. In addition, two wet pressing pressures (50 kPa and 350 kPa) were applied, and the retention of oxidized-CNF material in sheets was determined. Mechanical properties, i.e., strain at break, tensile strength, and tensile stiffness, as well as paper shrinkage during drying were measured.

EXPERIMENTAL

Materials

Cellulose microfibrils (CMF) and oxidized cellulose nanofibrils (oxidized-CNF)

CMF was prepared from never-dried birch kraft pulp by mechanical disintegration with a microfluidiser (Vartiainen and Malm 2016). The dispersed pulp (1.7% consistency) was first pre-refined with a grinder (Supermasscolloider MKZA10-15J, Masuko Sangyo Co., Japan) at 1500 rpm, followed by treatment with a fluidizer (Microfluidics M-7115-30, VTT, Espoo, Finland). The CMF was produced after five passes at an operating pressure of 1800 bar. No chemical modification was applied.

Oxidized-CNF was prepared from never-dried birch kraft pulp by TEMPO-mediated oxidation and fluidisation. TEMPO-mediated oxidation was performed as previously described (Saito and Isogai 2004). Fibres were suspended in a water solution of TEMPO and sodium bromide. NaClO solution was added to the suspension (5 mmol to 15 mmol per gram of fibres), and the pH was adjusted to 10 at room temperature with NaOH. The reaction was considered complete when the pH remained stable at 10. After oxidation, fibres were washed thoroughly with deionised water followed by treatment with the microfluidiser M7115-30 (2 passes).

Retention aids

Commercially available wet-end grade cationic starch Raisamyl 50021 with a degree of substitution 0.035 was received from Kemira (Finland). Cationic polyacrylamide (CPAM) Fennopol K3400R with molecular weight of 6 to 7 mg/mol and charge density of 1 meq/g was received from Kemira (Finland), and a dry powder anionic long chain copolymer of acrylamide and acrylic acid (anionic micropolymer, AMP) Perform™ SP7200 was received from Solenis. The dry cationic starch was diluted in deionized water (conductivity < 1 µS/m) so that the final concentration was 1%, and it was heated for 30 min at 90 °C under constant stirring. The dry CPAM and AMP were first diluted in deionized water so that the final concentrations were 0.3% and 0.5%, respectively. Solutions were kept under constant stirring until fully dissolved and then further diluted to 0.01% (CPAM) and 0.05% (AMP) solutions for actual use.

Pulp

The fibres used in the experiments were bleached softwood kraft pulp from a pulp mill in central Finland. The pulp was refined using conical fillings and a specific refining energy of 135 kWh/ton to SR 25 (at 3.6% consistency) with a Prolab refiner (Valmet, Finland) at Åbo Akademi University, Turku, Finland. Table 1 shows the morphological properties and fines content of ProLab pulp determined using Metso FibreLab and a Lorentzen & Wettre STFI FibreMaster analyser. Measuring systems were based on picture analysis in which single fibres were investigated in a water-fibre suspension.

Table 1. Morphological Properties of Refined Pulp Used in Experiments. Analyses Included Length-Weighted Fibre Length, Fibre Width, Fibre Curl, Fines, Shape Factor, Kink/mm and Kink Angle of the Pulp Fibres

Methods

Handsheet preparation: the wet-end additions of CMF and oxidized-CNF

Figure 1a shows the handsheet preparation scheme for the experimental series of wet end additions of CMF and oxidized-CNF. Handsheets were prepared according to a modified SFS-EN ISO 5269-1 (2005) standard using a Lorenzen & Wettre laboratory sheet former without white water recirculation. The aimed target basis weight was 60 g/m2. The pH of the pulp suspension was adjusted to 7.5 with NaOH and the conductivity to 1000 µS/cm with NaCl.

To enhance the retention of CNF and CMF, retention aids were used. The approach of the used retention system was modified from a commonly used dual system (Eklund and Lindström 1991). Cationic dry strength agents were used to bind anionic material and flocculate the fibre suspension, while anionic copolymer was added to bridge the primary flocs together and increase the shear resistance of the flocs.

First, cationic starch (1.5% per dry fibre) was added to the suspension that was stirred for 30 min before addition of retention aids and fibrillated material. After starch addition, 0.02% (per dry fibre) of CPAM and 0.02% (per dry fibre) of AMP and CMF or oxidized-CNF were added to the suspension. A 10 s delay was used between the additions in order to imitate the addition system in real paper machines. Oxidized-CNF was added in amounts to obtain papers with 3%, 5%, 7%, and 14% CNF content. CMF additions were 1%, 7%, 20%, and 35%. Dilute suspensions (0.3% to 0.5%) of CMF and oxidized-CNF were activated (fibril bundles opened) before addition by vigorous stirring with a laboratory disperser. The sheets were wet pressed at 50 kPa or 350 kPa for 5 min, after which the blotting boards between the sheets were changed to dry ones and pressing was continued another 2 min.

Handsheet preparation: Spray application of CMF and oxidized-CNF on wet fibre network

Figure 1b shows the handsheet preparation scheme for experimental series of spray application of CMF and oxidized-CNF. Handsheets were prepared like previously described but only with 1.5% (per dry fibre) of cationic starch and 0.01% (per dry fibre) of CPAM. Handsheets were wet-pressed with 50 kPa for 5 min to remove extra water and then CMF or oxidized-CNF solutions were sprayed on the wet sheets. CMF or oxidized-CNF solutions were sprayed at 0.3% to 0.5% consistency, and also contained cationic polyacrylamide 0.01% (per dry fibre). An electrospray gun was used for spraying the solutions, and the amount of CMF or oxidized-CNF added was controlled by weighing the sheets during the spraying procedure. Oxidized-CNF was added in amounts to obtain papers with 3%, 5%, and 7% CNF content. CMF addition was 7%. The solutions were sprayed onto the top-side of the sheet. CMF and oxidized-CNF penetration into the fibre network was enhanced by using vacuum suction after spraying. The sprayed sheets were wet pressed again with 350 kPa for 5 min, after which the blotting boards between the sheets were changed to dry ones and pressing was continued another 2 min.

Restrained and unrestrained drying

The handsheets were dried unrestrained between wire fabrics with a gap of 1 mm to 3 mm, allowing free shrinkage of the handsheets without excessive cockling or curling. The handsheets were dried and conditioned according to ISO 187 (1990) at 23 °C and 50% relative humidity prior to paper property measurements. For comparison, part of the handsheets (Fig. 1a) without (reference) and with wet-end addition 5% of oxidized-CNF were also restrain-dried via standard plate drying.

Fig. 1. The preparation schemes for handsheets made using refined fibres. a) Wet-end additions of CMF and oxidized-CNF; b) Spray applications of CMF and oxidized CNF

Analyses and Determinations

CMF and oxidized-CNF characterisation

The carboxylic content was determined by electric conductivity titration according to the method described by Saito and Isogai (2004). Apparent viscosity measurements were modified from the method described by Kangas et al. (2014). The CMF sample was diluted to 1.5% and the oxidized-CNF to 0.8% concentration with Milli-Q water and dispersed with an Ultra-Turrax disperser at 14000 rpm for approximately 2 min. Viscosity measurements were performed in a 250 mL Pyrex beaker, and each sample was left to settle for a minimum of 30 min at room temperature after the dispersion. This allowed the samples to regain their initial viscosity. The temperature of the sample was adjusted to 20 ± 1 °C, and a vane spindle V-73 was used for the measurements. The shear viscosity was measured at 300 measuring points at 0.5 rpm and 10 rpm. The relative viscosities were measured twice for each sample. The samples were gently mixed between the measurements.

A Nikon Eclipse Ci light microscope (VTT, Espoo, Finland) was used to image fibrillated material. A scanning electron microscope (SEM, Merlin, South China University of Technology) was used to characterise fibre morphology on a micrometre scale. An atomic force microscope (AFM, Bruker NanoScope V, South China University of Technology) was used to characterise fibre morphology. Fibril dimensions were roughly estimated by using ImageJ software (ImageJ freeware, USA). Fibrillated material was freeze-dried prior the imaging with SEM and AFM.

An X-ray diffraction device (XRD, Bruker D8 Advance, South China University of Technology) with a CuKα radiation (Kα1; 0.15406 nm and Kα2; 0.15444 nm), generated at 40 kV and 40 mA, was used to obtain the crystallinity of cellulose fibrils. XRD spectra were recorded from 5° to 60° at a scan rate of 1 °/min. Fibrillated material was freeze-dried prior to the analysis. Crystallinity index (CI) of CMF and oxidized-CNF was calculated from the x-ray diffraction patterns according to the Segal method in Eq. 1 (Segal et al. 1959),

 (1)

where I200 is the intensity of the 200 peak (2θ = 22.6°), and Iam is the intensity minimum between 200 peaks and 110 peaks (2θ = 18.7°).

A thermogravimetric analysis (TGA, Q500-1198, South China University of Technology) was used to analyse the thermal stability of fibrillated material. Fibrillated material was freeze-dried prior to the analysis. Temperatures for the analysis were run from 25 °C to 700 °C at a heating rate of 10 °C/min. Nitrogen atmosphere was used to prevent thermoxidative degradation.

Retention of oxidized-CNF in handsheets (Biofibril)

The amount of oxidized-CNF retained in the handsheets was evaluated based on an analytical method for determination of carbonyl ratio and/or concentration of oxidised nanofibrillar cellulose in a sample. Oxidised nanofibrillar cellulose in the sample was enzymatically hydrolysed into oxidised cellobioses. The cellobioses were then analysed and quantified to reveal the amount of oxidised nanofibrillar cellulose in the sample. High-performance anion-exchange chromatography with pulsed amperometric detection (HPAEC-PAD, Espoo, Finland) was used for determination of the concentration of oxidised nanofibrillar cellulose (Laukkanen et al. 2014). No analytical determination method could be used for approximation of retained CMF in the handsheets.

Paper shrinkage during unrestrained drying

Paper shrinkage during unrestrained drying was determined by measuring the perimeter before and after drying between four holes made with a standardised metallic punching plate on the wet papers. Shrinkage (%) was calculated according to Eq. 2,

 (2)

where Pw is the perimeter (mm) of the square-shaped wet handsheet, and Pd is the perimeter (mm) of the dry handsheet.

Mechanical and optical properties of the handsheets

Paper technical tests were determined according to ISO-standards (Table 2).

Table 2. Standards and Devices Used for Mechanical and Optical Characterisation of Handsheets

(*VTT, Jyväskylä, Finland)

Wet web strength (Impact test rig)

The wet web strength and strain at break were measured with the Impact test rig (VTT, Jyväskylä, Finland). The Impact test rig is a fast strain tensile tester, which uses a strain rate of 1 m/s, and is used for studying paper runnability by measuring the tensile and relaxation properties of both wet and dry paper samples. More detailed descriptions of the Impact test rig can be found in the literature (Kurki et al. 2004; Kouko et al. 2006).

2D formability strain measurement

A 2D-formability testing device (VTT Jyväskylä, Finland) was used to measure the formability of the handsheets. The 2D-formability tester measured the breaking strain and force of the sample formed with a double curved press. The formability strain of the samples was measured as an average value of seven samples at 23 °C, 60 °C, 90 °C, and 120 °C. Detailed descriptions of the procedure are found elsewhere (Vishtal et al. 2013; Vishtal and Retulainen 2014).

RESULTS AND DISCUSSION

Properties of Micro- and Nanofibrillated Cellulose Material Used in the Study

The determined properties and viscosity parameters of CMF and oxidized-CNF are listed in Table 3. The determined carboxylic content in CMF was 0.02 mmol/g to 0.05 mmol/g, i.e., very low, while the content in the oxidized-CNF was 1 mmol/g. The carboxylic contents were similar to previously reported values for wood cellulose and chemically oxidised cellulose (Isogai et al. 2011). Oxidized-CNF also had higher viscosity and higher yield stress than CMF. This indicated that oxidized-CNF consisted of more homogeneous, finer fibril material than CMF. The high degree of fibrillation in the oxidized-CNF sample resulted in a more homogeneous suspension, and higher internal friction between the small particles led to a higher viscosity value (Kangas et al. 2014). In rheology, yield stress describes the minimum stress required to achieve a structured flow of a certain fluid. Higher yield stress of oxidized-CNF meant that higher pressure needed to be applied to the material to make it flow.

Table 3. Properties and Viscosity Parameters of CMF and Oxidized-CNF

*Brookfield RVDV-III+, V73 vane spindle, 1.5%, T=21 °C.

**Brookfield RVDV-III+, V73 vane spindle, 0.8%, T=21 °C

Fig. 2a) X-ray diffraction patterns of cellulose microfibrils (CMF) and cellulose nanofibrils (CNF); b) Thermogravimetric analysis (TGA) curves of cellulose microfibrils (CMF) and cellulose nanofibrils (CNF). Weight loss observed under 100 °C is considered to be water.

Figure 2a shows the X-ray diffraction patterns of CMF and oxidized-CNF materials. X-ray diffraction patterns of CMF and oxidized-CNF were very similar, which meant that the oxidation had not changed the crystal structure of the CNF material (Isogai et al. 2011). However, oxidized-CNF had a lower crystallinity index (CI 54.2%) compared to CMF (CI 64.5%). The lower crystallinity of oxidized-CNF was an indication of more flexible and deformable material with higher water absorption capability.

Fig. 3. Light microscopy picture of (a) Congo red-stained CMF and (b) oxidized-CNF. Scanning electron microscopy (SEM) images of freeze dried samples of (c) CMF and (d) oxidized-CNF. Atomic force microscopy (AFM) images of freeze dried samples of (e) CMF and (f) oxidized-CNF.

The thermal degradation curves of CMF and oxidized-CNF are shown in Fig. 2b. CMF started to degrade at 270 °C in a nitrogen atmosphere, which is a typical degradation temperature for cellulose (Fukuzumi et al. 2010; Isogai et al. 2011). Oxidized-CNF started to degrade at a much lower temperature of 220 °C. TEMPO-oxidation has been shown to decrease the thermal degradation point of cellulose in other studies (Fukuzumi et al. 2010; Isogai et al. 2011). The replacement of cellulose hydroxyl groups with carboxylate groups in TEMPO-oxidation and decrease in crystallinity seemed to decrease the thermal stability of oxidized-CNF.

Different microscopy techniques were used to characterise the morphology. Some large fibril bundles could be distinguished by light microscope in the CMF suspension (Fig. 3a), while no clear fibrils were seen in the oxidized-CNF suspension (Fig. 3b). Scanning electron microscopy (SEM) and atomic force microscopy (AFM) images of CMF and oxidized-CNF showed clearly that the fibril content in the CMF was more heterogeneous with and larger in size than in the oxidized-CNF (Fig. 3c, 3d, and Fig. 3e, 3f). Rough estimation of dimensions showed values of 15 nm to 30 nm for CNF and 15 nm to 105 nm for CMF.

Retention of Fibrillated Cellulose

The retention of the oxidized-CNF proved to be a challenge. The retention of wet-end and spray applications of oxidized-CNF in the handsheets was determined according to the previously described method. With wet-end addition of oxidized-CNF, the retained amounts equalled 0.5% to 0.8% of the weight of the sheet (w/w%) (Fig. 4). Increasing the wet-end addition of oxidized-CNF to 14% did not increase the amount of retained CNF. The wet pressing pressure did not affect the oxidized-CNF retention either. Cationic dry strength agents were used to bind anionic material and flocculate the fibre suspension, while anionic copolymer was added to bridge the primary flocs together.

Fig. 1. The determined amount of oxidized-CNF retained in sheets (bars; oxidized-CNF retention). The amount of added oxidized-CNF (dots; oxidized-CNF addition) in sheets.

The application of retention polymers, i.e., 1.5% (per dry fibre) of cationic starch, 0.02% (per dry fibre) of cationic polyacrylamide, and anionic copolymer, had only a minor effect on the retention of the oxidized-CNF. However, notably higher retained amounts of oxidized-CNF, from 1.1% to 4.1% of the weight of the sheet (w/w%), were achieved when the CNF suspension was sprayed onto wet handsheets (Fig. 4).

Effect of Spray and Wet-End Addition of Fibrillated Cellulose on Paper Properties

Density and tensile index

Figure 5 shows the effects of wet-end and spray applications of CMF and oxidized-CNF on tensile index and density. A clear correlation between the tensile index and the density was observed. Wet-end applications of CMF and oxidized-CNF increased density and tensile index of the handsheets, but the effect of CNF (Fig. 5a.) was more notable than the effect of CMF (Fig. 5b).

Fig. 5. (a) Tensile index (Nm/g) as a function of density (kg/m3) of oxidized-CNF-reinforced sheets. Linear fit describes the effect of density on the tensile index of reference sheets (green line) and wet-end added CNF sheets (blue dashed line). Blue arrows show the direction of the effect of wet-end addition of CNF on reference sheets; (b) Tensile index (Nm/g) as a function of density (kg/m3) of CMF-reinforced sheets. Linear fit describes the effect of density on the tensile index of reference sheets (green line). The average values with their confidence level intervals (95%) are shown. Green dots represent reference sheets, blue dots CNF-reinforced sheets and red dots CMF-reinforced sheets

A linear fit was used to describe the effect of density on the tensile index of reference sheets (green line) and wet-end added CNF sheets (blue dashed line). Blue arrows were included to show the direction of the effect of wet-end addition of CNF on reference sheets. Oxidized-CNF increased the tensile index of the restrain-dried sheets by 15%. CMF and CNF increased the tensile index of the unrestrained sheets by 10% and 20%, respectively. The same trend was also observed with 50 kPa wet-pressed CMF and oxidized-CNF handsheets.

As shown in Figs. 6a and 6b, light scattering and air permeability of CMF and CNF-reinforced (wet-end) sheets were clearly lower than with reference sheets. Reduced light scattering can be seen as an indication of increased fibre-fibre contacts. The results indicated especially that oxidized-CNF not only increased the number of interfibre bonds, but it also increased the strength of the bonds. It is also worth mentioning that the high additions of CMF (20% and 35%) increased the drainage resistance notably during sheet making.

Fig. 2. (a) The effect of wet-end and spray applications of CMF and oxidized-CNF on air permeability (Bendtsen method, mL/min, 10 cm2) of the handsheets. (b) The effect of wet-end and spray applications of CMF and oxidized-CNF on light scattering of the handsheets. Green bars represent reference sheets, blue bars CNF-reinforced sheets and red bars CMF-reinforced sheets. The average values with their confidence level intervals (95%) are shown

Spray application of CMF and oxidized-CNF on the fibre network did not have a clear effect on density or tensile index (Fig. 5a and 5b square dots), but decreased air permeability and light scattering notably (Fig. 6a and 6b). The spray treatment in itself had no noticeable effect on most of the measured paper properties. However, air permeability increased considerably, from 222 mL/min to 450 mL/min. This was probably due to the effect of sprayed water that left the sheet wetter after wet pressing.

Elongation, shrinkage and tensile stiffness

Results of the effects of wet-end and spray applications of CMF and oxidized-CNF on strain at break, shrinkage, and tensile stiffness of the handsheets are shown in Fig. 7. Results showed a noticeable correlation between shrinkage and strain at break (Fig. 7a). As shrinkage and strain increased, tensile stiffness decreased (Fig. 7b). The unrestrained-dried reference sheets had notably higher strain at break (9.3% strain) than restrain-dried reference sheets (4.4% strain), but lower density and tensile index as expected, according to the literature (Fujiwara 1956; Page 1971; Htun and de Ruvo 1981; Htun et al. 1989; Waller and Singhal 1999).

Fig. 3. (a) Strain at break (%) as a function of shrinkage (%) Linear fit describes the effect of shrinkage on the strain at break; (b) Tensile stiffness index (kNm/g) as a function of light shrinkage (%). The black line is added to guide the eye. The average values with their confidence level intervals (95%) are shown. Green dots represent reference sheets, blue dots CNF-reinforced sheets and red dots CMF-reinforced sheets. Ref = reference sheet, restrained = dried under restrain, unrestrained = dried unrestrained

Wet-end addition of CMF improved shrinkage and strain at break with increasing CMF addition level, while tensile stiffness of the unrestrained handsheets decreased. The 20% addition of CMF increased shrinkage by 73% and strain at break by 20%. Spray application of CMF increased shrinkage by 10% and strain at break by 13%, but decreased tensile stiffness by 8%.

Wet-end addition of oxidized-CNF did not have a considerable effect on the strain or tensile stiffness of the restrain-dried sheets. Oxidized-CNF had no considerable effect on shrinkage and strain of the unrestrained sheets either, but it increased tensile stiffness by approximately 18%. Sprayed oxidized-CNF, however, increased the shrinkage by 45% and stain at break by 33%, while stiffness decreased by 20% to 25%. Oxidized-CNF addition by spraying may increase the strain at break more effectively, but also reduce the subsequent strain at break.

A positive effect on shrinkage was observed when lower wet pressing (50 kPa) and fibrillated material addition were combined. This can be seen with both CMF and oxidized-CNF, but with CMF the effect was much greater; the shrinkage increased by 57% and strain at break by 44%. Part of this was probably due to the higher moisture content after wet pressing (dry solid content of the handsheets shown in Fig. 8).

Initial wet strength

Wet tensile index and wet strain at break as a function of dry solid content (DSC) of selected CMF- and oxidized-CNF-reinforced sheets are shown in Fig. 8. Measurements were done in two dryness levels achieved by using two different wet pressing pressures (50 kPa and 350 kPa).

Fig. 4. Initial wet strength index (a) and strain at break (b) as a function of dry solid content of CMF- and CNF-reinforced sheets. Wet tensile index was calculated by dividing wet tensile strength with dry grammage of the sheets. (a) Exponential fit describes the effect of DSC on the wet tensile strength and (b) linear fit describes the effect of DSC on the wet strain at break. The average values with their confidence level intervals (95%) are shown

Wet tensile properties of paper has been shown to be dependent on the dry solid content of the paper (Kurki et al. 2004), and exponential fits describe well the effect of dryness level on the wet tensile strength (Retulainen and Salminen 2009; Erkkilä et al. 2013). Thus, exponential fit was also used here to describe the effect of DSC on the wet tensile strength (Fig. 8a). The effect of DSC on strain at break of the wet handsheets is described using linear fit (Fig. 8b).

There was improvement in wet tensile index and strain at break in the presence of CMF (20% addition). Tensile index increased from 12.7 Nm/g to 15.7 Nm/g and strain at break from 6.4% to 8.2% at the fixed dry solid content of 50%. Oxidized-CNF (3% addition) did not have any noticeable effect on wet web strength of the handsheets at the fixed dry solid content of 50%, but strain of the CNF sheets increased slightly from 6.4% to 7.4%. Figure 8 also shows that after lower wet pressing (50 kPa) the dry solid content of the CMF- and CNF-reinforced sheets (DSC 37% and 39%, respectively) was lower than in reference sheets (DSC 42%). After higher wet pressing (350 kPa), the dry solid content of the CMF- and CNF-reinforced sheets (DSC 56% and 58%, respectively) was slightly higher than in reference sheets (DSC 55%).

Formability in a two-dimensional test

The formability of CMF- and oxidized-CNF-reinforced sheets is shown in Fig. 9. Unrestrained sheets had higher 2D-formability than restrain dried sheets. 2D-formability strain was found to correlate well with strain at break of the paper sheets at 23 °C (Fig. 9a, R2=0.8681). This was expected, since the sample experienced similar stresses in a 2D-former as in a conventional tensile tester, but only in the vertical direction. CMF-reinforced sheets had the highest formability strain at all temperatures.