Introduction

In papermaking, the swelling phenomenon is mostly related to the ability of basic pulp components, namely cellulose fibres and fines, to absorb and bind water. The swelling is a direct outcome of water penetration into the capillary system of plant-based fibre cell wall with simultaneous interaction between polar liquid, i.e. water, and hydrogen bonds formed by hydroxyl (–OH) and carboxyl (–COOH) groups of cellulose and hemicelluloses, which are the constituents of fibre cell wall. In the presence of water, hydrogen bonds are replaced by “water bridges” (Hietanen and Ebeling 1990; Retulainen 1997), which is evidenced by increased material volume. Partial solubility of fibre cell wall hemicelluloses also contributes to swelling (Luukko and Maloney 1999). The swelling phenomenon starts at the moment cellulose-based pulp has been immersed into water. This is so-called free swelling. Research conducted by Olejnik (2012) showed that free swelling time was approximately 90–110 min, and an increase in the amount of absorbed water was usually very small—approx. 0.02–0.06 g of water per 1 g of b.d. pulp. Both swelling time and swelling degree depend on many parameters, among other things, on chemical composition of aqueous environment, pulp type, its consistency and refining degree (Wultsch and Weissmann 1959; Scallan and Grignon 1979; Scallan and Tigerström 1992; Gharehkhani et al. 2015). Results of research conducted, among others, by Klemm et al. (1998) confirmed that swelling degree of cellulose-based pulp increased in alkaline environment. Nevertheless, it should be remembered that pulp refining produces the most extensive swelling, which is obtained in relatively short time. The phenomenon of cellulose fibre swelling is of great importance in paper technology. An increase in swelling degree of cellulose fibres results in an increase in these fibres flexibility and, therefore, paper strength properties are improved (Przybysz and Czechowski 1985; Luukko et al. 1997; Joutsimo and Asikainen 2013). Water retention value (WRV) is the most common parameter used for the evaluation of the swelling effect in case of natural cellulose materials. However, some authors make statements about the importance of monitor also the rate (kinetics) of swelling (Botkova et al. 2013).

In addition to cellulose fibres, so-called fines fraction is present in fibrous pulp. According to the commonly accepted definition, fines are the cellulose pulp fraction (particles detached from cellulose fibres) that passes through a screen with a diameter of the round holes of 76 μm (Htun and DeRuvo 1978, T261 cm-94 TAPPI Standard method (TAPPI 1994)) or through a 200 mesh wire (Seth 2003). Depending on the objective of conducted research and the test methods applied, different authors use broader definition of fines fraction, identifying already as fines the particles passing through 150 or even 100 mesh wire (Ferreira et al. 1999; Seth 2003; Kang and Paulapuro 2006). This corresponds to the particles of diameter below 0.2 mm. Fines fraction has been recognized as a separate component of pulp since its properties differ considerably from those of actual cellulose fibres. A number of research (Laivins and Scallan 1996; Luukko and Maloney 1999; Seth 2003) have shown that fines exhibit, among other things, low dewatering potential, very high specific surface area, and significantly larger swelling ability, compared with cellulose fibres.

There are two types of chemical pulp fines, the primary fines and the secondary fines. Primary fines are present at low level in unrefined pulps and they have no major impact on papermaking process. Refining creates secondary fines, which are a valuable component of papermaking pulp, as they affect paper properties significantly (Gharehkhani et al. 2015). This fines fraction fills spaces between fibres within the paper, which reduces its porosity and increases total bonded surface area in paper. This results in increased paper strength properties (Retulainen et al. 1993; Hubbe 2002). Additionally, the presence of secondary fines fraction contributes to dissipation and more uniform distribution of stresses between fibres in paper (Waterhouse and Bither 1992), which affects its dimensional stability. Furthermore, production process itself is also greatly affected by fines fraction. Water absorbed during swelling of both cellulose fibres and fines cannot be removed mechanically, i.e. during paper web formation or pressing. Therefore, high level of swollen fines in paper web will lead to worsening of web dewatering in wet end, lower web consistency after press section and, hence, higher energy requirement during drying process of such a paper web (Giertz 1980; Przybysz and Czechowski 1985; Cedric et al. 2008; Chen et al. 2009).

It should be mentioned that, in case of fines, the term ‘swelling’, is probably not fully correct since swelling phenomenon is usually related to the internal sorption of water resulting in dimension change of the swollen material. Fines are very small, thus, their internal/external surface area ratio is lower than that of the fibre wall. Hence, it is very likely that high hydratation level of fines is a result of water absorbed both internally and externally. Therefore, in case of fines, the term ‘swelling’ should be considered as a general hydratation level. However, it is common in scientific literature that fines hydratation is most often characterized by WRV parameter and called swelling (Luukko 1999b; Luukko and Maloney 1999; Seth 2003; Mayr et al. 2017). Therefore, the same nomenclature has been used in the presented paper.

Secondary fines fraction has larger water sorption ability than primary fines. Research of Laivins and Scallan (1996) showed that the amount of water absorbed by secondary fines of chemical pulps (Softwood and Hardwood) might vary in the range of approximately 1–3 g/g (g of water per 1 g of b.d. pulp). Without any doubt, the type of pulp and its prior processing play an important role here. Swelling degree of fines fraction originating from mechanical pulps may amount to about 1.7 g/g (Luukko and Maloney 1999), whereas for secondary pulps, this value may be even lower than 1 g/g (Mancebo et al. 1981). Higher swelling degree of chemical pulp fines is probably an outcome of lower lignin content and better developed capillary system accessible to water (Luukko 1999a). Fines swelling process upon refining of pulp is difficult to assess, among other things, due to the fact that swelling itself requires certain time, during which a new portion of fines is generated simultaneously.

Knowledge about the possibility of effective control over fines generation and swelling during refining process are nowadays of key importance. Beside its significance in papermaking in papermaking process (e.g. impact on paper structure and mechanical properties and dewatering properties of the pulp), more and more often, the fines fraction is recognized as raw material for the production of novel materials e.g. microfibrilled cellulose (MFC) and nanocellulose (Matsuda et al. 2011; Pääkkö et al. 2007; Subramanian et al. 2008; Kalia et al. 2011). Possibility of application of microfibrillated material from natural cellulose fibres has been also researched by Taniguchi and Okamura (1998).

Among different production methods used for this purpose, Osong et al. (2013) listed many examples where refining process was employed. Due to the binding and filling properties of secondary fines fraction, some paper mills try to use it as an additive to the lower quality pulps in order to improve mechanical properties of produced paper. Therefore, knowledge of the impact of refining conditions onto the properties of generated fines may be significant to both modern papermaking process and designing of new cellulose-based materials (Retulainen et al. 2002). In the literature, it is still difficult to find more detailed information on this subject. The research by Luukko and Maloney (1999) showed that swelling degree of secondary fines increased with an increase in energy consumed during refining. Moreover, they determined that swelling degree was more extensive for higher specific energy consumption during refining. The result of work by Laivins and Scallan (1996) showed practically the constant linear correlation between swelling degree of fines and that of cellulose fibres in a given pulp. Swelling degree of fines fraction was twice as high as that of fibre fraction. Most researchers concentrated mainly on the effect of fines on paper properties (Przybysz and Czechowski 1985; Retulainen et al. 1993; Luukko and Paulapuro 1999).

As a result of the above analysis, series of pulp refinings was carried out. The main objective of the presented study was to determine the impact of the most important refining parameters (e.g. net refining energy, rotor speed, and pulp consistency) which could be used to control the fines development and their swelling degree. Correlation between swelling of chemical pulp fines and refining energy consumption was of special interest. Another objective was to evaluate the effect of refining forces on fines swelling during refining process. Intensity of the refining process was evaluated using commonly accepted factor SEL (Specific Edge Load), given by the Eq. (1):

$$SEL = \frac{{P_{net} }}{CEL},\;{\text{J/m}}$$
(1)

where Pnet is the effective refining power (W) and CEL is the cutting edge length (m/s). CEL is defined as (2):

$$CEL = Z_{R} \cdot Z_{S} \cdot l \cdot \frac{n}{60},\;{\text{m/s}}$$
(2)

where ZR is the number of rotor bars, ZS is the number of stator bars, l is the bar effective length (m), n is the rotational speed of refiner rotor (rpm), and 60 is the conversion divider (s/min).

SEL factor—introduced by Wultsch and Flucher (1958) and later supplemented by Brecht and Siewert (1966)—defines the amount of effective refining energy transferred by the edges of refining elements to the refining zone. SEL is still considered as a useful parameter when analysing refining process (Koskenhely and Paulapuro 2005; Desarada 2010). The main disadvantage of SEL is that it does not take into consideration the obvious impact of pulp consistency and its flow rate in the refining zone. That is why, in the industrial conditions, SEC (Specific Energy Consumption) is used additionally:

$$SEC = \frac{{P_{net} }}{{Q_{m} \cdot \frac{{C_{F} }}{100} \cdot \rho }},\;{\text{kWh/t,}}$$
(3)

where Pnet is the effective refining power (net power) (kW), Qm is the pulp volumetric flow rate through the refining zone (m3/h), CF is the consistency of refined pulp (%), and ρ is the pulp density (ton/m3). SEC expresses the amount of refining energy transferred to the refined pulp during a single pass through the refining zone and it can also be considered as a measure of refining intensity.

Materials

Commercial bleached kraft softwood (pine) pulp was used in the experiments. Pulp was delivered in the form of dry sheets. The average moisture content was 7.4%. Initial pulp properties are shown in Table 1.

Table 1 Initial pulp properties
Full size table

Methods

The refining process was carried out in a pilot plant described in the publication by Olejnik (2013). The plant was equipped with an Escher-Wyss conical refiner R1L working as a semi-continuous system (refined pulp passed through the refiner zone several times). Parameters of the refiner filling are given in Table 2.

Table 2 Parameters of the refiner filling
Full size table

SEL and SECSP factors were used to control the refining intensity. Specific energy consumption for single pulp pass (SECSP) through the refining zone in a given experiment was also kept constant. This factor was further denoted by the SECSP abbreviation, whereas total specific energy consumption, i.e. a number of the passes multiplied by the single pass specific energy consumption, was denoted by the SEC abbreviation. Each experiment started with the determination of the no-load power (measured for the refiner working with the pulp and maximum refiner gap clearance). A heat exchanger was used to cool the refined pulp so its temperature did not exceed 35 °C. A single volume of refined pulp was 130 dm3. The refining sequences were carried out in accordance with the scheme presented in Table 3.

Table 3 Refining parameters for each experiment
Full size table

Water retention values (WRV) for pulp and fibres were measured using a centrifugal method according to the SCAN-C 102 XE standard. The test sample was centrifuged under a specific centrifugal force (3000 g) for a specific time (15 min). The measurements were carried out with High Speed Centrifuge MPW-350 produced by MPW Med. Instruments Company. The coefficient of variation for all WRV measurements was less than 3%.

The fines fraction was defined as all the material passing through a 150 mesh wire. Consequently, the fibres fraction was retained on the 150 mesh wire. The total amount of the removed fines fraction was determined by the gravimetric method, comparing the pulp weight before and after a screening in a Bauer–McNett apparatus. The coefficient of variation for all the measurements of fines fraction amount was less than 6%.

Calculation of fines swelling

Swelling degree of fines fraction was calculated based on the measured values of: pulp WRV, fines WRV, and fines content in the examined pulp. The calculations were done after converting the well-known Eq. (4) into the Eq. (5):

$$WRV_{pulp} = \left( {1 - x_{fines} } \right) \cdot WRV_{fibers} + x_{fines} \cdot WRV_{fines}$$
(4)
$$WRV_{fines} = \frac{{WRV_{pulp} - WRV_{fibers} \cdot \left( {1 - x_{fines} } \right)}}{{x_{fines} }}$$
(5)

where WRVpulp—water retention value of the whole pulp, g/g, WRVfibres—water retention value of fibres, g/g, WRVfines—water retention value of fines, g/g, xfines—fines content in the pulp.

The same method of fines WRV calculation has been presented and used by Mayr et al. (2017).

Results and discussion

Curves in Fig. 1 illustrate fines content increase in bleached kraft pulp as a result of refining conducted for different net power and constant rotational speed of the refiner rotor. An increase in net power led to the increase in total refining intensity, expressed by SEL value. Upon refining, at a given specific energy consumption SEC, the higher the refiner net power was, the higher was the amount of generated fines fraction. These results confirm findings by Mou et al. (2013) that the increase in specific energy consumption caused the increase in fines content in refined pulp.

Fig. 1
figure 1

Changes in the amount of bleached kraft pulp fines as a function of the net power (constant parameters: rotational speed n = 900 rpm, pulp consistency CF = 3%, pulp volumetric flow rate through the refiner Qm = 1.51 dm3/s)

Full size image

Additionally, more rapid fines swelling was observed when refining intensity was higher (Fig. 2). These results corroborate the findings by Luukko and Maloney (1999). Presented here results indicate that—for a given SEC value—the higher the net power was, the higher was the fines WRV. However, it should be mentioned that such a dependence was observed for SEC value below 300 kWh/t. For higher values of specific energy consumption SEC, differences in swelling degree of fines were smaller and smaller, i.e. impact of refining intensity was less and less significant. For SEC values over 300 kWh/t, swelling degree of fines reached a specific boundary value which amounted to approx. 3–3.2 g/g, regardless of applied net power Pnet. The obtained maximum values of fines swelling correspond to those obtained by Laivins and Scallan (1996).

Fig. 2
figure 2

Changes in the swelling degree of bleached kraft pulp fines as a function of the net power (constant parameters: rotational speed n = 900 rpm, pulp consistency CF = 3%, pulp volumetric flow rate through the refiner Qm = 1.51 dm3/s)

Full size image

In the next part of the research, the effect of rotational speed of the refiner rotor on generation of fines fraction and its swelling degree changes was investigated. For constant effective refining power (net power), lower rotational speed results in higher SEL value and higher refining intensity. According to Fig. 3—for a constant net power—the lower the rotational speed was, the higher was the amount of generated fines (for a given specific energy consumption SEC). These results confirm direct relationship between the increasing refining intensity and the increasing amount of fines fraction.

Fig. 3
figure 3

Changes in the amount of bleached kraft pulp fines as a function of the refiner rotational speed for constant net power (constant parameters: Pnet = 1.2 kW, pulp consistency CF = 3%, pulp volumetric flow rate through the refiner Qm = 1.51 dm3/s)

Full size image

When analyzing the changes in swelling degree of fines induced by rotational speed of rotor at constant net power, it was found that fines WRV increased more rapidly for higher rotational speeds (Fig. 4). Moreover, in contrary to hereinbefore analysed effect of net power, noticeable differences in the final swelling degree values were observed here. For the lowest rotational speed of n = 600 rpm (the highest refining intensity), the highest increase in the amount of fines of the lowest swelling degree was obtained. On the other hand, for the highest rotational speed of n = 1500 rpm (the lowest refining intensity), the lowest amount of fines but of the highest final swelling degree was generated. In general, the highest increase in the amount of fines corresponded with the lowest swelling degree of fines.

Fig. 4
figure 4

Changes in the swelling degree of bleached kraft pulp fines as a function of the refiner rotational speed for constant net power (constant parameters: Pnet = 1.2 kW, pulp consistency CF = 3%, pulp volumetric flow rate through the refiner Qm = 1.51 dm3/s)

Full size image

The obtained results may indicate that changes in swelling degree of fines during refining are affected differently by refiner net power and by changes in shearing and friction forces caused by the rotational speed of the refiner rotor. The presented results show that, despite lower refining intensity, higher rotational speed leads to more rapid increase in swelling degree of fines and to higher boundary values of swelling degree of fines.

In order to determine the significance and the credibility of the refining intensity defined by SEL factor, the experiments were conducted in which this factor was kept constant. However, constant value of SEL was achieved thanks to the proper selection of refining parameters i.e. net power and rotational speed. Pulp consistency and its volumetric flow rate through the refining zone were constant. According to the SEL factor theory, refining processes should run similarly when this factor is kept constant. Hence, for a given specific refining energy consumption, the changes in both the fines amount and WRV should be comparable. Figure 5 shows an increase in the amount of fines for this case but those differences were small. It can be stated that the presented results confirm the validity of the SEL theory assumptions. However, the changes in swelling degree of fines (Fig. 6) reveal that, despite keeping constant value of SEL factor, changes in the rotational speed and changes in the net power result in condition changes in the refining zone. Accordingly, starting with SEC factor of approx. 100–150 kWh/t, certain differences in fines WRV were observed. In general, higher swelling degree of fines were obtained for higher rotational speeds and for higher net power. Taking into consideration hereinbefore described effect of net power (Fig. 2) and that of rotational speed (Fig. 4) on the swelling degree of fines, one may assume that the swelling effect would be influenced more significantly by the shearing and friction forces, resulting from higher rotational speed. Considering the fact that most of modern industrial refining systems are run at constant rotational speed, the refiner operators are deprived of important technological parameter which would improve refining process controllability.

Fig. 5
figure 5

Changes in the amount of bleached kraft pulp fines for constant SEL = 2.85 J/m obtained via suitable ratio of rotational speed and net power (other constant parameters: pulp consistency CF = 3%, pulp volumetric flow rate through the refiner Qm = 1.51 dm3/s)

Full size image
Fig. 6
figure 6

Changes in the swelling degree of bleached kraft pulp fines for constant SEL = 2.85 J/m obtained via suitable ratio of rotational speed and net power (other constant parameters: pulp consistency CF = 3%, pulp volumetric flow rate through the refiner Qm = 1.51 dm3/s)

Full size image

The next investigated issue was related to the determination of swelling degree of fines at constant refining intensity, expressed as constant value of SEL (n = 900 rpm = const., Pnet = 1.2 kW = const.), and at constant value of specific refining energy consumption in the refining zone (SECSP = 7.3 kWh/t) obtained via suitable ratio between pulp volumetric flow rate and its consistency—the following product was kept constant: CF * Qm = 4,54 (kg/s). Figure 7 presents changes in the amount of fines as a function of the refining energy consumption for a discussed case. Even though, the curves lie close to each other, it can be clearly noticed that, for a given energy consumption, the amount of fines increased with an increase in pulp consistency and with a decrease in flow rate in the refining zone. Confirmation of different conditions prevailing in the refining zone, despite keeping constant values of SEL and SECSP, can be found in Fig. 8. An increase in the swelling degree of fines was higher for higher pulp consistencies and for lower flow rates through the refining zone. It is a well-acknowledged fact that fibre flocculation is more intensive for higher pulp consistencies and lower pulp flow rates. It is also widely known that typical dimensions of refiner bars and the refining gap clearance are much larger than the diameter of fibres. Therefore, fibre flocs rather than individual fibres are subjected to refining. Comparison of the results in Figs. 7 and 8 indicates that both the friction forces between fibres and the total surface are on which these interactions occur (i.e. surface of contact between pulp particles) are of crucial importance for an increase in the amount as well as for a growth in the swelling degree of fines. It is obvious that both of these parameters will be higher for longer dwell time in refining zone and for higher refined pulp consistency. Taking into account that refining zone length was 180 mm, diameters of feed and outlet pipes were 80 mm, and pulp was incompressible liquid, the dwell time in the refining zone for each trial could be easily calculated (Table 4). It can be clearly seen that the dwell time for the pulp of consistency of 1% is 4 times shorter than that for the pulp of consistency of 4%.

Fig. 7
figure 7

Changes in the amount of bleached kraft pulp fines for constant SEL (n = 900 rpm = const., Pnet = 1.2 kW = const.) and constant SECSP in the refining zone (7.3 kWh/t) obtained via suitable ratio between pulp volumetric flow rate and its consistency

Full size image
Fig. 8
figure 8

Changes in the swelling degree of bleached kraft pulp fines for constant SEL (n = 900 rpm = const., Pnet = 1.2 kW = const.) and constant SECSP in the refining zone (7.3 kWh/t) obtained via suitable ratio between pulp volumetric flow rate and its consistency

Full size image
Table 4 Analysis of pulp dwell time in refining zone
Full size table

Curves in Fig. 8 also show that - for refinings at pulp consistencies of 2 and 3%—differences in swelling degree of fines were small. The differences, yet, are clearly visible for consistencies of 1 and 4%. Results presented in Figs. 7 and 8 comply with data obtained by other researchers (Page 1985; González et al. 2012) who found that the longer refining time and higher shear rate increase the amount of fines in refined pulp. This may be supplemented by additional finding that the increase of pulp dwell time in refining zone and higher pulp consistency cause the increase in both fines content and their swelling level.

In order to determine the effect of pulp consistency on the swelling degree of fines more thoroughly, the refining series at constant refiner operating conditions were carried out. In this trial series, the rotational speed was n = 900 rpm, the effective refining power (net power) was Pnet = 1.2 kW, the pulp volumetric flow rate was Qm = 1.51 dm3/s. For such operating conditions, the pulp dwell time in the refining zone for all studied cases was constant and equal to 0.6 s. It was found that the higher the consistency of refined pulp was, the larger was the increase in the amount of fines (Fig. 9). The largest differences were observed for the lowest pulp consistencies (1 and 2%). Furthermore, the higher the consistency of the refined pulp was, the more rapid was the growth in the swelling degree of fines (Fig. 10) and the higher was the final value of the swelling degree. Those differences were observed despite the constant pulp dwell time in the refining zone (Qm = const.). The obtained results indicate that the pulp consistency is the key parameter responsible for the increase in the WRV of generated fines and that the flow rate plays a minor role here.

Fig. 9
figure 9

Changes in the amount of bleached kraft pulp fines as a function of pulp consistency for constant refiner operating conditions (n = 900 rpm = const., Pnet = 1.2 kW = const., Qm = 1.51 dm3/s = const.)

Full size image
Fig. 10
figure 10

Changes in the swelling degree of bleached kraft pulp fines as a function of pulp consistency for constant refiner operating conditions (n = 900 rpm = const., Pnet = 1.2 kW = const., Qm = 1.51 dm3/s = const.)

Full size image

Data, presented in Fig. 11, confirm the decisive effect of the refined pulp consistency on the swelling degree of fines. This figure compares the two sets of the results of fines WRV changes. Within a given set, the pulp consistency was kept constant but the flow rate varied. The pulp dwell times in the refining zone were equal to 0.6 and 0.2 for 1% consistency, and 0.6 and 0.8 for 4% consistency.

Fig. 11
figure 11

Comparison of the flow rate effect on the changes in the swelling degree of fines for two different pulp consistencies: 1 and 4%

Full size image

The flow rate impact was observed only for refinings at lower consistency (1%). It was determined that—in that case—longer dwell time in the refining zone resulted in lower increase in fines WRV. However, in the case of 4% consistency refinings, neither the flow rate nor the pulp dwell time in the refining zone (for range studied) affected the swelling degree of fines.

Conclusion

On the basis of the presented research results, it may be concluded that cellulose fines generation and swelling phenomena which occur during pulp refining process are neither easy to describe precisely in terms of refining parameters, nor can be neglected. It has been found that the amount of the fines generated during refining process was affected by effective refining power (net power) and rotational speed of the refiner rotor. At a given refining energy consumption, the higher the effective net power and the lower the rotational speed were, the higher was the amount of fines. This means that the amount of fines was increasing when the refining intensity, expressed as SEL factor, was also increasing. Obtained results indicated that the final swelling degree of fines could be different, depending on the refining operating conditions used. During the conducted investigation, it was determined that—depending on the refining parameters used—the final swelling degree of fines ranged from 2.2 to 3.4 g/g. It was found that the swelling degree of fines was mostly influenced by the following refining parameters: rotational speed of the refiner rotor, refined pulp consistency, and—for very low consistency (1–2%)—also pulp volumetric flow rate through the refiner. For a given refining energy consumption, the swelling degree of fines increased with an increase in pulp consistency and rotational speed.

This means that the shearing and friction forces, related to these two parameters, have the crucial impact on the swelling phenomenon of fines. Therefore, greater attention should be paid to the rotational speed of the refiner rotor and the refined pulp consistency as, beyond doubt, these parameters contribute to more effective control over the entire refining process (in terms of the final cellulose-based material properties).