The reduction of energy consumption by industrial processes has become imperative due to rising energy costs and efforts toward decarbonisation. The continuous manufacturing of paper is energy intensive due in part to the water removal process required to convert pulp slurries to valuable paper products. This necessitates the development of energy conservation techniques, while simultaneously ensuring the quality of the product. A pilot-scale test unit was developed to quantify the effects of dwell time, vacuum pressure, and refining energy on the achievable pulp mass concentration or dryness level of three pulp types utilised in paper machines. Pulp dry matter was investigated as a means of gauging vacuum consumption and hence energy utilisation in paper machines, which could potentially reduce utility consumption of the overall drying process. A novel approach to simulate the pulsating high vacuum zone in the forming section of a paper machine was implemented, allowing the development of statistical correlations to explore vacuum dewatering conditions that may lead to energy efficiency. Bleached hardwood, mechanical/groundwood and recycled pulp were characterised to determine the effects of refining energy on fibre morphology and their drainage behaviour in pulp slurries. A dryness level of 21.8% at -55 kPa gauge was achieved for bleached hardwood, whereas lower values of 19.8% and 18.3% were observed for recycled and mechanical pulps, respectively. This behaviour was attributed to the differences in drainability and morphology of the pulps due to their respective unique properties, further exaggerated by refining.
Research Article
Experimental investigation of pulp high vacuum dewatering by suction boxes
https://doi.org/10.21203/rs.3.rs-4724106/v1
This work is licensed under a CC BY 4.0 License
You are reading this latest preprint version
Pulp refining
Pulp characterisation
Vacuum dewatering
The pulp dewatering process in paper machines is energy-intensive (Rezk et al. 2013). Filtrate removal occurs in three distinct stages, each with varying levels of energy utilisation. The forming section draws filtrate from pulp slurries through gravity and vacuum. This is followed by the press section where mechanical force is used to further dewater pulps. The last stage, referred to as the dryer section, utilises evaporation to achieve the desirable product. Although the dryer section removes the least amount of water, it is responsible for most of the energy applied for dewatering as it becomes increasingly difficult to remove filtrate from pulp slurries (Sjöstrand 2017). The opposite is true for the forming section which removes the most amount of water thereby making it imperative to maximise filtrate removal in this section. Vacuum consumption in the forming section may be monitored by assessing the dryness level of pulps leaving the forming section, thereby inferring energy consumption. Improving dryness levels after vacuum dewatering in the forming section will reduce energy usage in the dryer section.
Fibre web formation through drainage of pulp suspension in the high vacuum dewatering zone of the forming section of a paper machine has been studied by various authors (Åslund 2008; Pujara et al. 2008a, b; Rahman et al. 2018; Sjöstrand et al. 2020). In the first zone of the forming section of a paper machine, water is removed through low vacuum filtration to accelerate gravitational dewatering. The paper product is subsequently exposed to higher levels of vacuum to further induce dewatering in the second zone, which is applied in suction pulses to produce fibre mats with an outlet mass concentration of 18 to 25% (Belle and Odermatt 2016); this process is referred to as high vacuum dewatering. High vacuum dewatering occurs at pressures ranging between -15 and ‑40 kPa gauge, although vacuum pressures as low as -65 and -70 kPa gauge can be implemented at high-speed paper production (Åslund 2008; Åslund and Vomhoff 2008). The vacuum pulses are achieved by employing slotted suction boxes that act as vacuum chambers to administer a pressure differential over each slot as the wet paper product or pulp slurry moves over these boxes, thus drawing in air to displace filtrate from the pulp slurries. The combined duration of individual pulses is referred to as the dwell time, which is dependent on paper speed through the machine and the slot dimension. Slurries are transported across suction boxes by a filter medium called the forming fabric, which is permeable to water and air while retaining valuable fibres and additives (Bajpai 2018). Vacuum pressure and dwell time are usually the focal point of studies that aim to predict conditions in the high vacuum dewatering zone.
Experimental simulations of the high vacuum dewatering process can be categorised into static and dynamic setups. Montgomery (2008) modified a conventional hand sheet former to study the effect of vacuum dewatering of pulp on retention and filler migration. The simple filtration setup was modified to allow the collection of data during the mat formation process, thus providing a static setup to determine the impact of different vacuum pressures on the retention of fibres, fines, and filler. However, such static setups are incapable of accurately simulating the vacuum pulsation effects that are applied in industrial paper machines. More complicated dynamic setups have been previously explored to better simulate the vacuum pulsation effects, achieve more accurate dwell times or paper machine speeds and attain fibre orientations in pulp slurries, comparable to industrial formers. For example, Räisänen et al. (1995) reported on the application of a pilot-scale moving belt drainage tester in which pulsation effects were achieved by using a slotted conveyor belt rotating around a suction box that supplied vacuum to hand sheets. The authors developed a mathematical model by modifying a wet press model by Jönsson and Jönsson (1992) in which they replaced a press pulse with a suction pulse to better explain the results obtained. It was concluded that a vacuum pulse had a more prominent effect on the dryness of pulp when compared to airflow where experimental data was comparable with predictions provided by the model. Mitchell et al. (2002) improved the design of the moving belt drainage tester by including a novel mixer that created turbulence in pulp slushes using compressed air, which pushed out fine jets of water for improved fibre orientation. In an alternative mechanical design, Pujara et al. (2008a, b) simulated pulsations through a rotating disc with a slot width of 0.0127 m whereby the dwell time was controlled by varying its rotational speed. Regression analysis of their results revealed a linear dependency on the airflow and sheet grammage by the achievable solids content whereby a plateau is eventually reached for all vacuum pressures at maximum dwell time; the plateau of pulp dewatering rate is a well-known phenomenon (Ramaswamy 2003). Nilsson (2014) utilised a similar configuration, employing a linearly driven plate with a single slot beneath a sample holder to administer a pulse. They concluded that air mass flux is directly related to vacuum level, especially for low-grammage pulp sheets. It would be beneficial to incorporate the effects of fibre structure when empirically analysing pulp slurry drainage behaviour during high vacuum dewatering by a dynamic experimental simulation like those previously discussed.
Three setups were evaluated to choose the most appropriate dynamic configuration for the required conditions. These include the moving belt drainage tester by Räisänen et al. (1995) which utilised a perforated belt to create pulses, a method that can be prone to friction as the belt loosens over time with usage (Pujara 2008). The formation of the sheet in this setup was achieved on the drainage tester, thereby requiring large storage of removed filtrate, as sheet formation is achieved at a very low pulp mass concentration or dilute state. This could also lead to excessive air leakage into the storage unit as it was prone to flooding (Mitchell et al. 2002). The dynamic setup utilised by Nilsson (2014) only produced a single vacuum suction pulse during each run where the plate had to be moved back to its original position for more pulses, which prevents the simulation of multiple exposures as is the case in industrial formers. The third setup evaluated was designed by Pujara et al. (2008a, b). This design requires a complicated sealing mechanism to prevent excessive friction between the rotating disc and vacuum chamber, which presents significant challenges in fabrication and is likely to have limited robustness against failure.
A novel laboratory suction box layout is presented to achieve the dynamic conditions that exist in industrial formers. Pulses were administered by employing a spindle with five slots encased in a vacuum chamber. The chamber had a single slot that would align with each spindle slot to administer vacuum pulses to hand sheets. The configuration simplified the design of a mechanical seal, as there was no direct contact to cause friction which was the case in the design utilised by (Pujara et al. 2008a, b). Early stages of gravity drainage were achieved by using a hand sheet former thus preventing the need for large storage needed for removed filtrate. The novel configuration could achieve pulses in the order of a few milliseconds where machine speeds of up to 400 m/min could be simulated at vacuum pressures as low as -60 kPa gauge. Three vacuum reservoirs were included to reduce air leakage during dewatering. The novel dynamic configuration critically contributed towards enabling the collection of dewatering data of pulps under various vacuum dewatering conditions by simulating the forming section in industrial formers.
The relationships that may exist between the structural makeup of individual fibres in pulps, their behaviour in slurries and the achievable mass concentration or consistency during vacuum dewatering by suction boxes in the high vacuum dewatering zone were analysed. There are limited studies where authors statistically relate the changes in water-fibre interactions and morphology of pulps due to refining, to their drainage behaviour in the high vacuum zone when developing correlations to predict solids retention and improve energy efficiency. This issue is addressed in the current study. Three pulp types were evaluated, namely, bleached hardwood, mechanical or groundwood as well as recycled pulp as each represents a specific category of pulping processes implemented in the South African Pulp and paper industry.
Novel laboratory suction box design constraints
A dynamic setup with a novel configuration was constructed through an analysis of previous configurations reported in the literature, to simulate the conditions that exist in suction boxes. Table 1 below provides a summary of requirements that were fulfilled by the novel configuration, followed by various design factors considered.
Table 1 The design requirements and constraints of the novel laboratory suction box
|
Variable/Parameter |
Test Range |
|
Manipulated variables |
|
|
Vacuum pressure |
-10 to -60 kPa gauage |
|
Spindle speed |
400 m/min |
|
Air flux |
≤110 m3/min.m2 |
|
Singe pulse dwell/suction time |
≥ 6 ms |
|
Controlled variables |
|
|
Grammage oven-dry |
170 g/m2 |
|
Incoming pulp solids content or mass concentration |
4% to 7% |
|
Measured output |
|
|
Outlet solids content or pulp mass concentration (%) |
|
Pulse-generating mechanism of novel laboratory suction box
Pulse generation was achieved by implementing a slotted spindle residing beneath the sample holder (Figure 1). The spindle is inside a vacuum chamber. It has five slots, each with a width and length of 38.2 and 150 mm, respectively whereby the latter was chosen to match the sample diameter. Each pulse was the result of the spindle slot aligning with the vacuum port opening or slot directly below the sample holder during rotation. A pressure transmitter was placed inside the vacuum port to capture the pressure drops occurring during pulse generation. A solenoid valve bank regulated the distribution of vacuum from the reservoirs to the chamber within which the spindle was contained. The valves could also be used to generate much longer pulses. The desired dwell time was achieved by rotating the spindle at a specified velocity using a motor. A standard Programmable Logic Controller (PLC) with a high-speed card along with a dashboard or a screen for display was utilised to monitor and control the operating speed during each test. A separate high-speed data acquisition was used to log high-speed pressure readings as it was beyond the operating parameters of the standard PLC.
Vacuum reservation mechanism
Three vacuum reservoirs each with a capacity of 50 L were included in the setup to provide enough storage for the required vacuum conditions. Each reservoir was fitted with a pressure gauge to monitor the vacuum pressure. A manifold connected the reservoirs to the vacuum chamber housing the spindle, vacuum source as well as the solenoid valve bank. The system could be evacuated to the desired vacuum pressures using an ejector, which was specifically chosen as it required less maintenance due to the lack of moving parts and easy detection of leakages (Kent 2018). Previous configurations have shown signs of excessive air leakage during evacuation and dewatering of pulps. This occurrence seemed prominent in the studies done by Pujara et al. (2008a, b) and Räisänen et al. (1995), as previously discussed.
Control philosophy
Test conditions were controlled where pulse duration as well as the dwell time were specified, after which the motor would rotate at the appropriate speed to generate the desired pulse frequency. The reservoirs would be evacuated to a specific vacuum pressure to allow the administration of vacuum pulses to the hand sheets thus removing filtrate. The experimental setup is shown in Fig. 1 below.
Fibre characterisation
Three pulp types were studied as specified in Table 2. All samples were collected directly from the pipeline of a paper mill during stock preparation at different refining points.
Table 2 Pulp sample specification
|
Pulp type |
Pulping process |
Wood species/grade |
|
Bleached hardwood pulp |
Kraft |
Eucalyptus |
|
Groundwood pulp |
Mechanical |
Spruce |
|
Recycled pulp |
None |
New corrugated kraft waste |
The pulp mass concentration of the collected slurry was computed as specified in ISO 4119:1995. The concentration for each characterisation method was obtained using Eq. (1) below.
C1V1 = C2V2 Eq. (1)
Where C and V represent the pulp mass concentration and volume of slurry, respectively. Subscripts 1 and 2 signify the sample before and after dilution to the desired pulp mass concentration for the specific characterisation method.
Pulp samples in the form of slurry were prepared per ISO 16065-2:2014 to achieve a mass concentration and volume of 0.1% and 100 mL, respectively. This was done to determine fibre morphological properties such as fibre length, shape factor and fines content. These properties were obtained at each refining stage of the pulps using data that was collected from the Lorentzen & Wettre fibre tester (ABB 2020).
The pulp freeness was tested following a procedure described in the technical information document ISO 5267-1:2001. Freeness is a measure of the drainability of pulp and can be acquired using the Schopper Riegler (°SR) method. It is important to note that the higher the value in °SR, the lower the drainability of the pulp whereas the opposite is true for low °SR values.
The water retention value is a parameter that indicates how well water is bound in fibres (Sundblad 2015). It is measured as the ratio of mass of water per unit mass of dry fibre. A centrifuge was utilised to determine the parameter as illustrated in ISO 23714:2014, where a pulp pad with a grammage of 1700 g/m2 at a diameter of 30 mm was placed in the equipment to be centrifuged at a force of (3000±50) g for 30 min ±30 s at a temperature of (23±3) °C where g is the gravitation acceleration. The parameter was computed as shown in Eq. (2).
Where m1 and m2 represent the pad mass after centrifugation and oven drying, respectively.
Hand sheet formation and vacuum dewatering
Hand sheets were formed to a pulp mass concentration of approximately 7% according to the procedure in TAPPI T-205:2006 where sheets with a diameter of 150 mm and an approximate mass of 3 g were prepared, which is the recommended mass for pulps suitable for board manufacturing. The chosen grammage to achieve the desired mass was 170 g/m2 as shown in Table 1 above. Formed sheets were placed on the sample holder above the chamber housing the spindle, after which they were dewatered at room temperature. The dewatered hand sheets were weighed before being placed in an oven to dry overnight at 105 °C ± 2 °C. The oven dry mass was measured to enable computation of the outlet mass concentration according to Eq. (3). Test conditions during vacuum dewatering of pulp hand sheets are provided in Table 3 below.
Where mc, mf and mi represent the outlet pulp mass concentration (%), oven-dry mass and sample mass after vacuum dewatering in grams, respectively.
Table 3 Vacuum dewatering test conditions
|
Parameter |
Value |
|
Temperature (°C) |
Room (21-28) |
|
Vacuum pressure (kPa gauge) |
-19, -37, -55 |
|
Dwell time [Single Pulse time] (ms) |
30[6], 70[14], 110[22], 135[27], 250[50] |
|
Sample diameter (mm) |
150 |
|
Grammage (g/m2) |
170 |
Pulse generation
The pulse frequency produced by the novel laboratory suction box was compared to an ideal pressure profile generated from data characterising the high vacuum dewatering zone of a paper machine. This was to ensure that hand sheets to be dried using the dynamic configuration would observe vacuum suction pulses like that in industrial formers thus validating its ability to effectively simulate the process. Fig. 2 below shows an idealised pressure profile generated based on specifications for the vacuum pressure and suction box configuration of an industrial former in a South African paper mill. The depicted pressure profile was generated at vacuum pressures as low as -13.9 kPa gauge. It provides information on vacuum pulses that are expected at minimum and maximum paper speeds of 2.2 and 4.7 m/s, respectively.
Consecutive pressure drops are equivalent to the number of slots each suction box has and signify suction pulses. This means that there were five pulses exhibited during vacuum dewatering over each suction box. The duration or dwell time of a single slot was computed to be 6 ms for a maximum speed of 4.7 m/s whereas 12 ms was observed for a minimum speed of 2.2 m/s. Therefore, it is desirable to achieve such dwell times. Atmospheric pressure is recorded during deadtime when pulp slurry is not exposed to vacuum, during which the concentration gradient drives water from the forming fabric to the partially dried slurry, i.e., rewetting (Åslund 2008). Three suction boxes are presented in the pressure profile.
Actual pressure profiles achieved during drying of the hand sheets using the laboratory suction box, were compared to the preferred ideal profile in Fig. 2 above. Reflected in Fig. 3 is a pressure profile at a vacuum pressure of -55 kPa gauge for a dwell time of 30 ms. Five pulses were specified during all tests, as is the case in the ideal profile thereby resulting in a 6 ms individual vacuum pulse for a total dwell time of 30 ms. Fig. 3 shows that more than five pulses were observed out of the test boundary, which is due to the spindle rotating before the vacuum chamber was fully evacuated to the desired pressure of -55 kPa gauge. Residual pulses also occurred as the spindle continued rotating during which conditions were still below atmospheric pressure because of the slow closing time of the solenoid valves with respect to the spindle speed. There were more than 5 suction pulses or pressure drops, which means that pulp samples were exposed to a vacuum for far longer than desired, although latter pulses occurred at pressures closer to atmospheric conditions thereby still providing acceptable results. Shorter pulses or dwell times proved to be challenging to imitate, however, a pressure drop of only 2.8% was recorded at the last pulse when compared to the first pulse, in the test boundary.
Fig. 4 below shows improved vacuum dewatering performance for a test completed with the suction box set to a total dwell time of 135 ms. The individual pulses in the figure were relatively better defined. Furthermore, there were no residual suction pulses unlike in Fig. 3. This indicates that the suction box performs better at longer dwell times. There was a slight departure from vacuum condition between consecutive pulses due to air leaking into the pressure vessels, which is demonstrated by a slight increase in the absolute pressure after each suction pulse. This is especially true for the last vacuum pulse, where a pressure drop of 50% was recorded when compared to the first suction pulse. This could have been caused by the solenoid valve bank closing before the completion of the last pulse by the spindle.
Pulse generation using the solenoid valve bank proved to be successful, as shown in Fig. 5 below. The valve bank could accurately simulate five vacuum pulses like those depicted in Fig. 4 above. Hand sheets were dried for 250 ms with each pulse lasting for 50 ms. The first pulse occurs during evacuation of the sample holder thereby resulting in pressures closer to atmospheric conditions. Consecutive pulses are administered at the appropriate vacuum pressure with slight pressure drops due to minor air leakage into the vessels. The pressure drop observed between the second and last pulse was 12%, which is much lower than that observed in Fig. 4 above.
The dynamic configuration is capable of simulating pulsation effects by suction boxes as shown by the consecutive pressure drops in the pressure profiles discussed above.
Pulp characterisation results
Table 4 provides the average values of characterisation parameters of the three pulps to be dewatered. Samples of each were collected at different refining stages after which morphological traits were recorded thereby showing the transition of fibre structure of the pulps because of the refining process. It is important to note that the different pulp types were refined at different levels of energy due to their different structural properties. Fig. 6 below shows the effect of refining energy on the freeness of all three pulps.
Table 4 Fibre characterisation data
|
Pulp type |
Refining intensity (kWh/ton) |
WRV (g/g) |
Freeness (°SR) |
Fines content (%) |
|
Bleached hardwood pulp |
0 |
1.23 |
25 |
21.5 |
|
60 |
1.40 |
32 |
22.7 |
|
|
113 |
1.56 |
37 |
25.0 |
|
|
Mechanical pulp |
0 |
1.14 |
48 |
74.1 |
|
44 |
1.16 |
50 |
74.0 |
|
|
Recycled pulp |
0 |
1.34 |
29 |
43.5 |
|
70 |
1.39 |
37 |
46.3 |
Table 4 shows that the mechanically extracted pulp has the lowest drainability of 50 °SR. This can be attributed to its high fines content of 74.1 % after only a single refining stage at 44 kWh/ton. Fines reduce drainage by reducing media permeability due to their high surface area (Olejnik et al. 2017). The behaviour is graphically demonstrated in Fig. 6, where mechanical pulp exhibited the highest °SR values of all three pulps. Bleached hardwood pulp only generated 25% of fines after the second refining stage at 113 kWh/ton, resulting in a much faster dewatering, reflected in the average drainability of 37 °SR provided in Table 4. Mechanical pulps are exposed to abrasive forces during fibre separation (Kerekes et al. 2023), which may lead to excessive external fibrillation during refining thereby producing more fines(Retulainen et al. 1993). This is better reflected in the drainability of the two virgin pulps where bleached hardwood has the highest drainage rate as represented by low average °SR values of 25, 32 and 37 (see Table 4) at each consecutive refining stage. The opposite is true for groundwood or mechanical pulp which has an °SR value of 50 after refining at 44 kWh/ton, a decline of only 4% from its unrefined state. However, bleached hardwood pulp observed a decrease of over 28% after refining at 60 kWh/ton which further decreased to 48 % at 113 kWh/ton when compared to its unrefined state. Bleached hardwood pulp is the most susceptible to the effects of refining as reflected by drastic changes in its drainability after each stage. This is attributed to its low lignin contents which is common for chemical pulps, often resulting in decreased resistance to refining (Małachowska et al. 2020). Recycled pulp produced a fines content of 46.3 %. Its average drainage rate of 37 °SR is much better than that observed for mechanical pulp and identical to bleached hardwood pulp. However, Fig. 6 shows that the pulp released relatively less filtrate when compared to bleached hardwood pulp with slightly higher °SR values. Trends observed for recycled pulp were like those in virgin pulps where refining reduced its drainage rate. Therefore, it can be deduced that refining has an adverse effect on the drainage rate of pulps. A high drainage rate may be an indication of rapid filtrate removal during high vacuum dewatering by suction boxes.
Water retention values of all three pulps reveal that refining energy is positively associated with increased filtrate holding capacity as shown in Table 4 and Fig. 7. The parameter provides better visualisation of internal fibrillation in pulps (Motamedian et al. 2019), thereby indicating that mechanical pulp is more resistant to internal fibrillation when compared to bleached hardwood. This is shown in Fig. 7 below where linear trendlines are included to show the direct positive dependency of WRV on refining.
Bleached hardwood pulp has low lignin content (Brancato 2008) thereby making it less resistant to internal fibrillation, which is reflected by its high-water retention value of 1.56 g/g at a refining energy of 113 kWh/ton. The opposite is true for mechanical pulp whose WRV only increased from 1.14 to 1.16 g/g. Mechanical pulp had the lowest water retention value of 1.16 g/g at the highest refining stage when compared to bleached hardwood with a value of 1.56 g/g (see Table 4 above). An increase in the WRV of recycled pulp from 1.34 to 1.39 g/g was also observed. It is expected for the pulps to retain more water because of internal fibrillation brought about by the refining process.
Hypothesis testing through Analysis of Variance (ANOVA) was employed to assess the existence of significant relations that may exist between refining energy and water-fibre morphological traits. Relationships were rendered null if P > 0.05 thereby implying that no significant interactions were found. Freeness was found to be significantly affected by the increase in refining energy for all three pulps. Pearson’s coefficient further revealed that the relationship is positive thereby implying that drainability as expressed in °SR is inversely related to the effect of refining. The water retention value of bleached hardwood and recycled pulp was greatly affected by refining where P<0.05. The water retention value of mechanical pulp was not significantly affected by the changes in refining energy as the null hypothesis was accepted with P>0.05, which is due to its high lignin contents as previously discussed.
Vacuum dewatering experimental data
Vacuum dewatering of the three pulps was assessed by regressing an exponential decay to the experimental data to better visualise the diminishing effect of dwell time on pulp mass concentration as shown in Fig. 8a-c, i.e., a plateau in the dewatering rate of pulps, which is an effect discussed in multiple high vacuum dewatering studies (Ramaswamy 2003). Each pulp was tested using samples collected at the last refining stage.
The exponential relationships shown in Fig. 8a-c for vacuum pressures of -19, -37 and -55 kPa gauge, respectively, could be regressed using an exponential decay function, shown in Eq. (4) to best represent the plateau in the dewatering rate of pulps concerning dwell time.
Bleached hardwood pulp achieved the highest mass concentration values of 14.1% and 21.8% at -19 and -55 kPa gauge as seen in Fig. 8a and c. However, Fig. 8b shows that recycled fibre achieved a dryness level of 19.2% at - 37 kPa gauge. This value is higher than that achieved by bleached hardwood, i.e., 18.7% at -37 kPa gauge. Both pulps have a drainability rate of 37 SRo at their last refining stage, as reflected by similar vacuum dewatering behaviour. This is further reflected by comparable although lower pulp mass concentration values of 13.81% and 19.8% at pressures of -19 and -55 kPa gauge for the recycled pulp. Mechanical pulp had the lowest drainability of 50 SRo because of extreme external fibrillation, leading to the highest fines content of 74% (see Table 4 above). This resulted in poor vacuum dewatering where the lowest values of 12.1%, 15.8% and 18.3% were achieved at - 19, -37 and -55 kPa gauge, respectively. Experimental data shows that all three pulps observed a dewatering plateau for dwell times ranging from 30 to 135 ms, which is shown in Fig. 8a-c . Highest pulp mass concentration values of 21.8% (bleached hardwood pulp), 19.8% (Recycled pulp) and 18.3% (mechanical pulp) were recorded at a pressure for -55 kPa gauge and a dwell time of 250 ms for all three pulps as shown in Fig. 8a-c. Therefore, lower vacuum pressures coupled with longer dwell times resulted in dryer hand sheets and therefore high mass concentration for all three pulps. This was especially true for pulps that were not severely fibrillated during refining, i.e., bleached hardwood and recycled pulp.
The effect of dewatering time and maximum pulp concentration constants on the vacuum dewatering behaviour of the three pulps (see Eq. (4)) was statistically analysed with respect to fibre characteristics. The water retention value and freeness were utilised to evaluate the extent of fibrillation endured by the pulps because of refining. This is because change in these water-fibre parameters are the results of pulp internal and external fibrillation after refining (Abitz and Luner 1989; Gu et al. 2018). Analysis of covariates (ANCOVA) was utilised to prove that there is an effect on the vacuum dewatering constants of all three pulps by the freeness and water retention value when vacuum pressure is included as a covariate where P<0.05.
The results are graphically presented in Figs. 9-12 below for visualisation of the effect of freeness on the dewatering time constants of all three pulps thus exploring how drainabillity of pulp affects the pulp dry matter achieved during vacuum dewatering.
Bleached hardwood and recycled pulp achieved the same maximum pulp mass concentration constant of 20% at the lowest vacuum pressure of -55 kPa gauge as shown in Fig. 10. It could be attributed to the two pulps having the same drainability or freeness of 37 °SR. However, the virgin bleached pulp has the lowest dewatering time constant of 17 ms, thereby indicating quick dewatering when compared to recycled pulp with double the dewatering time constant of 34 ms at this pressure. This is demonstrated in Fig. 9. According to the results, bleached hardwood reaches a dewatering plateau at a relatively faster rate. At -19 kPa gauge, the pulps achieve similar plateau pulp mass concentrations of 13% (bleached hardwood) and 14% (recycled pulp), with visibly different dewatering rates of 43 ms and 39 ms for bleached hardwood and recycled pulp, respectively. A maximum pulp mass concentration constant of 17% was achieved by both pulps at -37 kPa gauge with the virgin pulp observing a relatively lower dewatering time constant of 21 ms, much quicker than the 35 ms reported for recycled pulp. Mechanical pulp showed signs of resistance to filtrate removal as proven by an °SR value of 50. Therefore, it had the highest dewatering time constants of 169 ms (-19 kPa gauge), 47 ms (-37 kPa gauge) and 38 ms (-55 kPa gauge) which resulted in the respective low plateau or maximum pulp mass concentration constants of 13%, 15% and 16%. From the discussion above, high drainability and low vacuum pressures may be associated with high pulp mass concentration for all three pulps. This is graphically shown in Figs. 9 and 10.
Figs. 11 and 12 show the relationship between pulp water retention value and the vacuum dewatering time constants. Although the previous discussion showed that bleached hardwood and recycled pulp had the fastest dewatering rate at all vacuum levels, they observed the highest water retention values of 1.56 g/g and 1.39 g/g when compared to mechanical pulp whose water retention was quantified at 1.16 g/g, all of which is provided in Table 4. Therefore, it may be implied that the two variables have an indirect proportionality. Pulp water retention value is a representation of internal fibrillation undergone by pulps during the refining process and is often used to monitor pulp swelling and thereby fibre flexibility (Singh 1996; Olejnik et al. 2017). High WRV can be associated with high swelling and therefore better flexibility (Olejnik et al. 2017). Bleached hardwood and recycled pulp have better flexibility when compared to mechanical pulp as they have relatively high WRV, as shown in Table 4. Enhanced flexibility promotes web deformation or compression of pulps. This is a desirable phenomenon as it is one of the mechanisms by which dewatering occurs (Åslund and Vomhoff 2008). However, extremely flexible fibres are prone to compacting during pulp mat formation which causes a phenomenon known as sheet sealing (Sjöstrand et al. 2019). Sheet sealing occurs when drainage channels in pulp mats are blocked as fibres form compact networks at the point of contact with the forming wire thereby trapping water from the mat (Hubbe et al. 2020). It is an undesirable effect that can be further exaggerated by high fines content due to their high surface area which reduces permeability and slows the dewatering rate. Therefore, pulp flexibility must be promoted in moderation to ensure compressibility while reducing the risk of sheet sealing. Bleached hardwood pulp observed the highest dewatering rates at all vacuum pressures. It had the highest WRV of 1.56 g/g which indicates high compressibility. The low fines content of 25.0 % in the pulp also prevented the sealing phenomenon, consequently resulting in accelerated dewatering when compared to its recycled and mechanical counterparts. Recycled pulp followed with a slightly higher fines content of 46.3% and a relatively lower WRV of 1.39 g/g. Mechanical pulp had the poorest dewatering performance due to its low flexibility which was signified by low water retention value of 1.16 g/g. This was further aggravated by a high fines content of 74.0%, which reduced the permeability of media during dewatering. Figs. 11 and 12 graphically demonstrate the results discussed.
Vacuum pulses were achieved using a robust laboratory suction box design. Appropriate vacuum conditions were maintained due to sufficient vacuum reservation. Formation of hand sheets was achieved through a hand sheet former thus avoiding the need for a large capacity of removed filtrate storage which could have led to excessive air leakage as was the case in the study done by Räisänen et al. (1995). The newly developed laboratory suction box adequately replicated high vacuum dewatering by suction boxes in the forming section of paper machines. This was further proven by the pressure profiles obtained during the dewatering of hand sheets. Multiple suction pulses were successfully generated, and the results observed followed established trends reported previously . Pulp mass concentration and fibre characteristics show that refining negatively affects the vacuum dewatering behaviour of pulps. This is due to its ability to induce fines generation in pulps through external fibrillation hence resulting in low dewatering rates and therefore low pulp mass concentration for all pulps. Lower vacuum pressures result in better dewatering of pulps which eventually plateau after a certain dwell time as shown by the maximum pulp mass concentration constant which is essentially a representation of the point at which pulps stop releasing filtrate regardless of the vacuum pressure and exposure time. The dewatering time constant shows the rate at which filtrate is removed. It was found to be much quicker for bleached hardwood and recycled pulp. Mechanical pulp had the slowest rate of dewatering as it was extremely externally fibrillated. Water retention was found to be a useful measure of pulp compressibility. Freeness or drainability successfully corroborated high vacuum dewatering trends of pulps where pulps with the same rate of drainability achieved similar dryness levels or pulp mass concentration. Therefore, fibre characteristics may be used as a preliminary measure of the expected vacuum dewatering behaviour of pulps. All pulps were dewatered at the same vacuum pressures; however, it would be beneficial to explore specific vacuum regions for each pulp type to determine optimum energy-saving operating conditions in the high vacuum zone of paper machines.
The authors confirm the integrity of the research and its presentation, including the following the rules of good scientific practice:
- The manuscript has not been submitted to more than one journal for simultaneous consideration.
- The submitted work is original and has not been published elsewhere in any form or language (partially or in full), except for the Masters thesis of Mahlohonono Mafela.
- This is not a single study split up into several parts to increase the quantity of submissions and submitted to various journals or to one journal over time (i.e. ‘salami-slicing/publishing’).
- Results are presented clearly, honestly, and without fabrication, falsification or inappropriate data manipulation (including image based manipulation).
- No data, text, or theories by others are presented as if they were the author’s own (‘plagiarism’). Proper acknowledgements to other works are given (this includes material that is closely copied (near verbatim), summarized and/or paraphrased).
- Appropriate and relevant literature has been cited in support of the claims made. Excessive and inappropriate self-citation or coordinated efforts among several authors to collectively self-cite was not attempted
Competing Interests
The authors have no relevant financial or non-financial interests to disclose.
Funding
This work was financially supported by the Paper Manufacturers Association of South Africa (PAMSA), grant number PRU/126.
Author contribution
M.M.: Conceptualisation, Methodology, Formal analysis, Investigation, Data curation, Writing – Original draft preparation and Editing, Visualisation. J.G.: Funding acquisition – Experimental setup, Writing – Reviewing, Supervision, Project administration. T.L.: Formal analysis, Writing – Reviewing, Supervision. L.T.: Writing – Reviewing, Supervision. G.S.: Methodology – Conceptual and detailed design of experimental setup, Writing – Reviewing. M.G.d.G.: Methodology – Conceptual and detailed design of experimental setup, Writing – Reviewing. J.P.K.: Methodology – Methodology – Conceptual and detailed design of experimental setup, Writing – Reviewing. W.v.d.W.: Methodology – Conceptual and detailed design of experimental setup, Writing – Reviewing.
The authors have all consented to publication of the manuscript in its present form.
- ABB (2020) L&W Fiber tester plus: Testing and industry-specific instruments. https://new.abb.com/pulp-paper/abb-in-pulp-and-paper/products/lorentzen-wettre-products/pulp-measurements/l-w-fiber-tester-plus. Accessed 14 Apr 2024
- Abitz P, Luner P (1989) The effect of refining on wet fiber flexibility and its relationship to sheet properties. In: Baker CF, Punton V (eds) Fundamentals of Papermaking, Trans. of the IXth Fund. Res. Symp. Cambridge. FRC, Manchaster, pp 67–86
- Annergren G, Hagen N (2009) Industrial Beating/Refining. In: Ek M, Gellerstedt G, Henriksson G (eds) Pulp and Paper Chemistry and Technology- Paper Chemistry and Technology. De Gruyter, pp 121–135
- Åslund P (2008) On suction box dewatering mechanisms. Dissertation, Royal Institute of Technology
- Åslund P, Vomhoff H (2008) Dewatering mechanisms and their influence on suction box dewatering processes – A literature review. Nord Pulp Paper Res J 23:389–397. https://doi.org/10.3183/npprj-2008-23-04-p389-397
- Bajpai P (2018) Biermann’s Handbook of Pulp and Paper. Elsevier
- Belle J, Odermatt J (2016) Initial wet web strength of paper. Cellulose 23:2249–2272. https://doi.org/10.1007/s10570-016-0961-7
- Brancato AA (2008) Effect of progressive recycling on cellulose fiber surface properties. Dissertation, Institute of Paper Science and Technology
- Gu F, Wang W, Cai Z, et al (2018) Water retention value for characterizing fibrillation degree of cellulosic fibers at micro and nanometer scales. Cellulose 25:2861–2871. https://doi.org/10.1007/s10570-018-1765-8
- Hubbe MA, Sjöstrand B, Nilsson L, et al (2020) Rate-limiting Mechanisms of Water Removal during the Formation, Vacuum Dewatering, and Wet-pressing of Paper Webs: A Review. Bioresources 15:9672–9755. https://doi.org/10.15376/biores.15.4.hubbe
- Jönsson KA, Jönsson BTL (1992) Fluid flow in compressible porous media: II: Dynamic Behavior. AIChE Journal 38:1349–1356
- Kerekes RJ, McDonald JD, Meltzer FP (2023) External fibrillation of wood pulp. Tappi J 22:363–371. https://doi.org/10.32964/TJ22.6.363
- Małachowska E, Dubowik M, Lipkiewicz A, et al (2020) Analysis of cellulose pulp characteristics and processing parameters for efficient paper production. Sustainability 12:1–12. https://doi.org/10.3390/su12177219
- Mitchell C, Parker I, Johnson RE (2002) Development of the moving belt sheet former/drainage tester. Appita Journal 287–292
- Montgomery J (2008) The Role of Suction Boxes on Forming Section Retention and Filler Migration. Dissertation, The University of British Columbia
- Motamedian HR, Halilovic AE, Kulachenko A (2019) Mechanisms of strength and stiffness improvement of paper after PFI refining with a focus on the effect of fines. Cellulose 26:4099–4124. https://doi.org/10.1007/s10570-019-02349-5
- Nilsson L (2014) Air Flow and Compression Work in Vacuum Dewatering of Paper. Drying Technology 32:39–46. https://doi.org/10.1080/07373937.2013.809732
- Olejnik K, Skalski B, Stanislawska A, Wysocka-Robak A (2017) Swelling properties and generation of cellulose fines originating from bleached kraft pulp refined under different operating conditions. Cellulose 24:3955–3967. https://doi.org/10.1007/s10570-017-1404-9
- Pujara J, Siddiqui MA, Liu Z, et al (2008a) Method to characterize the air flow and water removal characteristics during vacuum dewatering. Part I - Experimental method. Drying Technology 26:334–340. https://doi.org/10.1080/07373930801898091
- Pujara J, Siddiqui MA, Liu Z, et al (2008b) Method to characterize the air flow and water removal characteristics during vacuum dewatering. Part II - Analysis and characterization. Drying Technology 26:341–348. https://doi.org/10.1080/07373930801898125
- Rahman H, Engstrand P, Sandström P, Sjöstrand B (2018) Dewatering properties of low grammage handsheets of softwood kraft pulps modified to minimize the need for refining. Nord Pulp Paper Res J 33:397–403. https://doi.org/10.1515/npprj-2018-3037
- Räisänen KO, Paulapuro H, Karrila SJ (1995) Effects of retention aids, drainage conditions, and pretreatment of slurry on high-vacuum dewatering: a laboratory study. Tappi J 78:140–147
- Ramaswamy S (2003) Vacuum Dewatering During Paper Manufacturing. Drying Technology 21:685–717. https://doi.org/10.1081/drt-120019058
- Retulainen E, Moss P, Nieminen K (1993) Effect of fines on the properties of fibre networks. Products of papermaking 10th Fundamental Research Symposium Oxford 727–769. https://doi.org/10.15376/frc.1993.2.727
- Rezk K, Nilsson L, Forsberg J, Berghel J (2013) Modelling of water removal during a paper vacuum dewatering process using a Level-Set method. Chem Eng Sci 101:543–553. https://doi.org/10.1016/j.ces.2013.07.005
- Singh CK (1996) The Effects of Fines, Drying Intensity and Recycling on Pulp and Paper Properties. Dissertation, Western Michigan University
- Sjöstrand B (2017) Dewatering aspects at the forming section of the paper machine: Rewetting and forming fabric structure. Dissertation, Karlstad University
- Sjöstrand B, Barbier C, Ulkten H, Nilsson L (2019) Dewatering of softwood kraft pulp with additives of microfibrillated cellulose and dialcohol cellulose. Bioresources 14:6370–6383. https://doi.org/10.15376/biores.14.3.6370-6383
- Sjöstrand B, Nilsson L, Ullsten H, Barbier C (2020) Numerical model of water removal and air penetration during vacuum dewatering. Drying Technology 0:1–10. https://doi.org/10.1080/07373937.2020.1745825
- Sundblad S (2015) Predictions of pulp and paper properties based on fiber morphology. KTH Royal Institute of Technology
No competing interests reported.