1
KTH ROYAL INSTITUTE OF TECHNOLOGY
Doctoral Thesis in Fibre and Polymer Science
The effect of environment on refining efficiency of kraft pulps
MARIE BÄCKSTRÖM
Stockholm, Sweden 2020
2
The effect of environment on refining efficiency of kraft pulps
MARIE BÄCKSTRÖM
Academic Dissertation which, with due permission of the KTH Royal Institute of Technology, is submitted for public defence for the Degree of Doctor of Philosophy on Friday the 11th September 2020, at 10:00 a.m. in Kollegiesalen, Brinellvägen 8, Stockholm
Doctoral Thesis in Fibre and Polymer Science KTH Royal Institute of Technology
Stockholm, Sweden 2020
3
© Marie Bäckström, 2020 All rights reserved
© De Gruyter Paper 1, 2, 3, 6, 7
© Assoc Tecnica Brasileira de Celulose e Papel Paper 4
© Marie Bäckström Paper 5, 8
ISBN: 978-91-7873-615-7 TRITA-CBH-FOU-2020:39
1 ABSTRACT
Although the pulp and paper mill processes have been operational for long time there is still a need to understand the unit operations in paper making and how they interact with the ingoing pulp material. This is crucial in order to fully utilize the potential of the wood as well as of the unit operations. In order to do that it is vital to have an understanding about the produced pulp fibres, how they are constructed and how they respond to different conditions of their environment.
The aim of this work has been to clarify how the environment influences the refining efficiency of kraft pulps in terms of energy requirement and paper property development. The main hypothesis has been that the swelling of the fibres, due to their inherited polyelectrolytic gel nature, will not only affect the fibre as such but also the strength and properties of the fibre flocs that are mechanically treated between bars in the refiner and in this way affect the refining efficiency.
The main focus has been to study how the chemical environment, in particular the initial fibre swelling, affects the refining efficiency. Therefore, the influence of counter-ions to the charged groups, the number of charged groups, electrolyte concentration, pH and rheological behaviour was studied. Additionally, the importance of fibre flocculation for the refining efficiency was investigated by chemical means, i.e. to chemically flocculate and deflocculate a fibre suspension just before entering the refiner and evaluate the refining efficiency. An investigation to clarify the importance of refining homogeneity was also performed.
The work was performed both on a laboratory scale and in pilot scale using industrial refiners to ensure the validity of the results.
The importance of the counter-ions to the charged groups on refining was demonstrated. When the counter-ions was sodium the refinability, defined as the required energy input to reach a certain WRV or tensile index, was reduced by up to 50%. The more energy-efficient refining of pulps in the Na+-form may be explained as a co-operation between a higher osmotic pressure in the fibre wall and the mechanical stress applied during refining, so called “electrostatic repulsion-assisted refining”. When mechanical forces are applied on the fibre and the fibre wall, the electrostatic repulsion forces due to the ionization act as an additional aid to increase the swelling, and this in turn helps to delaminate the fibre wall. This
“electrostatic repulsion-assisted refining” also resulted in another type of external fibrillation of the fibres than that produced for the reference pulp in calcium form. The fibrils were very short and tiny. The improved refining efficiency could not be attributed to any rheological effect such as floc strength or floc size.
The number of charges correlated to refinabiliy of the pulp material, but there has to be a balance between the number of charges and ultrastructure of the pulp fibre. If too many charges were introduced, the internal ultrastructure was eventually damaged, and no property development was obtained in the refining.
The importance of refining heterogeneity on paper properties was investigated by mixing less refined or unrefined pulps and highly refined pulps in different proportions giving a wide distribution on energy input to individual fibres, as well giving swelling distribution curves. The mechanical properties of the produced paper were surprisingly alike, and the influence of the extreme inhomogeneity was rather small.
In a pilot paper machine trial, clear effects due to heterogeneous refining was only observed when 50% of the stock fed to the paper machine was unrefined. This implies that the mills can have large freedom in refining strategies without any significant negative impact on the tensile strength properties at a given density.
2 SAMMANFATTNING
Industriell tillverkning av massa och papper har skett under en lång tid. Trots detta så finns det fort- farande behov av att bättre förstå enhetsoperationerna vid tillverkningen. Detta är viktigt för att kunna utnyttja hela potentialen av såväl startmaterialet som för de enskilda enhetsprocesserna. För att göra detta är förståelsen för egenskaperna hos de producerade massafibrerna, hur de är uppbyggda och hur de svarar på olika förhållanden vid malningen viktiga.
Målet med detta arbete har varit att klargöra hur malningseffektiviteten, dvs energibehov och utveckling av papperets mekaniska egenskaper påverkas av den omgivande miljön vid malningen av kemisk massa.
Huvudhypotesen har varit att fibrerna ur svällningssynpunkt kan betraktas som polyelektrolytiska geler och att detta kommer att påverka fibern och även ha betydelse för styrkan hos fiberflockarna som utsätts för den mekaniska bearbetningen mellan bommarna i kvarnen och på så sätt påverka
malningseffektiviteten.
Huvudfokus har varit att studera hur den kemiska miljön, i synnerhet hur den initiala fibersvällningen, påverkar malningseffektiviteten. Därför har motjonen till de laddade grupperna, antalet laddade grupper, elektrolytkoncentration, pH och reologiska egenskaper studerats. Dessutom undersöktes betydelsen av fiberflockning för malningseffektiviteten genom polymertillsatser, d.v.s. tillsats av flockulerings- och deflockuleringsmedel till en fibersuspension strax före kvarnen. Påverkan på malningen utvärderades. En studie för att klargöra betydelsen av malningshomogenitet genomfördes också. Arbetet har utförts både i laboratorie- och pilotskala.
I detta arbete har betydelsen av motjonen till de laddade grupperna för malningsresultatet demonstrerats.
När motjonen var natrium så förbättrades malningseffektiviteten, uttryckt som den malenergi som krävs för att uppnå ett givet WRV eller dragindex, med upp till 50%. Den mer energieffektiva malningen av massor i Na+-form kan förklaras med ett samarbete mellan det högre osmotiska trycket i fiberväggen och de mekaniska applicerade krafterna i kvarnen som ger en så kallad "elektrostatisk repulsions-assisterad malning". När mekaniska krafter appliceras på fibern och fiberväggen, och de laddade grupperna är joniserade, bidrar den elektrostatiska repulsionen till att öka svällningen. Detta leder i sin tur till en delaminering av fiberväggen. Malning av fibrer i Na+-form resulterade även i en annan typ av extern fibrillering än den som sker vid de malning av referensmassa i kalciumform. Den förbättrade
malningseffektiviteten kunde inte tillskrivas någon reologisk effekt vad gäller vare sig flockstyrka eller flockstorlek.
Antalet laddade grupper korrelerade med malningseffektiviteten hos massan, men antalet laddningar måste balanseras mot ultrastrukturen hos fibermaterialet. Om alltför många laddningar introducerades så
förstördes till slut fibrernas ultrastruktur och ingen utveckling av fibrernas mekaniska egenskaper skedde vid malningen.
Betydelsen av malningsheterogenitet på pappersegenskaper undersöktes genom att blanda massor med olika bearbetningsgrad i olika proportioner för på så att få en bred fördelning av hur mycket energi de enskilda fibrerna fått vid den mekaniska bearbetningen. Påverkan av den extrema heterogeniteten var liten då de mekaniska egenskaperna hos det producerade papperet var överraskande lika. I ett
pilotpappersmaskinförsök observerades endast en effekt av malningsheterogenitet när 50% av massan var omald. Detta innebär att pappersbruken har stor frihet när man väljer malningsstrategi utan att få någon betydande negativ inverkan på dragstyrkan vid en given densitet.
3 LIST OF PUBLICATIONS
This thesis is a summary of the following papers.
I. Hammar L-Å, Bäckström, M Htun M (2000)
Effect of the counterion on the beatability of unbleached kraft pulps Nord. Pulp Pap. Res. J. vol. 15, no. 3, pp 189-193
II. Hammar L-Å, Bäckström M, Htun M (2000)
Effect of electrolyte concentration and pH on the beatability of unbleached kraft pulps Nord. Pulp Pap. Res. J vol. 15, no. 3, pp 194-199
III. Bäckström M, Hammar L-Å, Htun M (2009)
Beatability and runnability studies of ion-exchanged unbleached kraft pulps on a pilot scale Nord. Pulp Pap. Res. J vol. 24, no. 1, pp 94-100
IV. Bäckström M, Bolivar S, Paltakari J (2012)
Effect of ionic form of fibrillation and the development of the fibre network strength during the refining of the kraft pulps
O´Papel vol. 73, no. 7, pp 57-65 V. Bäckström M, Hammar L-Å (2010)
The influence of the counter-ions to the charged groups on the refinability of never- dried bleached pulps
BioResources 5(4), 2751-2764
VI. Bäckström M, Melander E, Brännvall E (2013)
Study of the influence of charges on refinability of bleached softwood kraft pulp Nord. Pulp Pap. Res. J. vol. 28 no. 4, pp 588-595
VII. Bäckström M, Tubek-Lindblom A, Wågberg L (2009)
Studies of the influence of deflocculants and flocculants on the refining efficiency on a pilot scale
Nord. Pulp Pap. Res. J vol. 24, no. 3, pp 319-326 VIII. Bäckström M, Mohlin U-B (2019)
The Influence of refining heterogeneity on paper properties BioResources 14(3), 6577-6590
Other related material.
Bäckström M, Kolar M-C, Htun M (2008)
Characterisation of fines from unbleached kraft pulps and their impact on sheet properties Holzforschung, vol. 62, no. 6, 546-552
Bäckström M (2011)
Fractionation and refining of pulp fractions- a potential for paper property tailoring Biorefining& Fibre Engineering, PIRA conference 2011 25-26 May, Barcelona, Spain
4 CONTRIBUTIONS TO THE PUBLICATIONS This thesis is a summary of the following papers.
I. Principal author, experimental layout, evaluation and interpretation. Participation in the practical work.
II. Principal author, experimental layout, evaluation and interpretation. Participation in the practical work.
III. Principal author, experimental layout, evaluation and interpretation. Participation in the practical work.
IV. Principal author, experimental layout, evaluation and interpretation of the data.
V. Principal author, experimental layout, evaluation and participation in the practical work.
VI. Principal author, experimental layout, evaluation of the data.
VII. Principal author, experimental layout, evaluation and participation in the practical work.
VIII. Principal author, experimental layout, evaluation and participation in the practical work.
5 LIST OF ABBREVIATIONS
APAM = anionic polyacrylamide CD= cross direction
CMC= carboxymethyl cellulose CPAM= cationic polyacrylamide DD = double disc
L= cutting edge length per time, km/s MD= machine direction
P= total refining power, kW
P0= idling power (no-load power), kW SEL = specific edge load (J/m) SEM= scanning electron microscopy SRE = specific refining energy input, kWh/t WRV= water retention value
6
TABLE OF CONTENTS Page
1. AIM OF THE WORK 7
2. INTRODUCTION 8
2.1 Structure and composition of wood fibres 8
2.2 Fibre swelling 10
2.2.1 Fibre charge 11
2.2.2 Measuring fibre swelling 14
2.3 Influence of the polyelectrolytic gel nature of fibres on papermaking properties 14
2.3.1 Stock preparation 14
2.3.2 Consolidation 17
2.4 Refining 18
2.4.1 Refining effects on fibre characteristics 19
2.4.2 Refining mechanism 20
2.4.3 Property development during refining 23
3. EXPERIMENTAL 24
3.1 Materials 24
3.2 Equipment and experimental lay-out 24
3.2.1 Refiners 25
3.2.2 Gap Clearance 26
3.2.3 Rheometer 26
3.3 Methods 26
3.3.1 Measurement of total charge 26
3.3.2 Measurement of surface charge 26
3.3.1 Ion-exchange 26
3.4 Analyses 27
4. RESULTS AND DISCUSSION 28
4.1. Influence of charged groups and counter-ions on the refining efficiency 28
4. 1.1 Type of counter-ion 28
4.1.2 Influence of number of charges on refinabaility 32
4.1.3 Influence of electrolyte concentration and pH 35
4.1.4 Fibrillation 37
4.1.5 Electrostatic repulsion-assisted refining 40
4.1.6 Floc strength 41
4.2 Flocculation and refinability 42
4.3 Refining heterogeneity 43
5. CONCLUSIONS 46
6. CONCLUDING REMARK AND FUTURE STUDIES 47
7. REFERENCES 49
8. ACKNOWLEDGEMENTS 53
7 1. AIM OF THE WORK
The aim of this work has been to clarify how the environment influences the refining efficiency of kraft pulps in terms of energy requirement and paper property development.
The hypothesis has been that the swelling of the fibres, due to their inherent polyelectrolytic gel nature, will not only affect the fibre as such, but also the strength and properties of the fibre flocs that are mechanically treated between bars in the refiner and in this way affect the refining efficiency.
The focus has been to study how the chemical environment, in particular how the initial fibre swelling, affects the refining efficiency. Therefore, the influence of counter-ions to the charged groups, the number of charged groups, electrolyte concentration, pH and the rheological behaviour was studied.
Additionally, the importance of fibre flocculation for the refining efficiency was investigated by chemical means, i.e to chemically flocculate and deflocculate a fibre suspension just before entering the refiner and evaluate the refining efficiency. An investigation to clarify the importance of refining homogeneity was also performed.
The work was performed both on a laboratory scale and in a pilot scale using industrial refiners to ensure the validity of the results.
8 2. INTRODUCTION
The production of paper from different biomass resources has a long history; the Egyptians started to write on papyrus. The word paper is etymologically derived from papyrus. The paper making process evolved in China, Asia and Africa and reached Europe 1085. Since then it has been a tremendous development in the field of pulp and paper making. There has been a huge increase in the production volume which has increased the application areas for papers. Nowadays the main raw material for paper is wood and the Nordic countries are at the fore front in both production and technical development but the competition from fast-grown forest plantation in South America is tough.
Although pulp and paper mills have been in operation for a long time there is still a need to understand the unit operations in the paper making processes and how they interact with the ingoing material. This is crucial in order to fully utilize the potential of the wood as well as the potential of the unit operations. In order to do so it is vital to have an understanding about the produced pulp fibres, how they are
constructed and how they correspond to different stimuli.
To make a very generalized overview, the fibres that are fixed in the wood are liberated in the pulp mill.
The pulp consists of fibres, in our case fibres acquired from wood. Liberation of fibres can be done either chemically or mechanically. In chemical pulping approximately half of the amount of wood become pulp fibres and the other part is dissolved in the cooking liquor. The two major chemical pulping processes are kraft pulping and sulphite pulping. In the kraft pulping, which is the dominant chemical pulping process worldwide, sodium hydroxide ions and hydrogen sulphide ions are the active cooking chemicals. The sulphite process uses various salts of sulphurous acid, either sulphites (SO3 2−), or bisulphites (HSO3−), depending on the pH. The counter ions can be sodium (Na+), calcium (Ca2+), potassium (K+), magnesium (Mg2+) or ammonium (NH4+). Some products need further treatment in a bleaching plant in order to remove remaining lignin to achieve a fully white product. The characteristics of a chemical pulp fibre is that it has a porous fibre wall, is flexible and can produce a paper sheet with high mechanical strength.
Mechanical pulping is produced by mechanical means through grinding or refining and the pulp yield is between 80-100%. The dominant mechanical pulping techniques are TMP (Thermo Mechanical Pulp) and CTMP (Chemo Thermo Mechanical Pulp). The characteristics of TMP is that it has very good light scattering properties whereas the CTMP pulp fibres is providing stiffness. TMP is usually a furnish for newsprint production while CTMP is a very effective furnish as middle layer in a paper board.
After the pulping process the liberated pulp fibres are consolidated to form a paper web. To prepare the fibres for this and to meet the specified requirements of the paper, the in-going fibres need to be refined to improve their function in the paper. In addition to pulp fibres, paper products may also constitute of fillers, dyes, strength additives and other chemicals to retain the components on the fibres and in the produced paper product. The paper products are manufactured by paper machines specially design for the intended products, e.g. paper machines for manufacturing board, tissue, sack, liner board, fine paper etc.
2.1 Structure and composition of fibres
Softwood fibres have generally a length of 2.5 -3.5 mm and a fibre width of 25-50 μm while hardwood fibres only are about 0.5-1.5 mm in length with a thickness of 20-25 μm.
The cell wall of a softwood fibre is considered as a hierarchal structure composed of cellulose fibrils, hemicellulose and lignin, Figure 2.1. These three wood polymer components build up several concentric lamellae, layers formed during the cell growth and surrounding an open space, the lumen. The layers are mainly distinguished by the different orientation of the fibrils.
9
Figure 2.1. Schematic illustration of a typical softwood fibre. M = middle lamella, P = primary wall, S =secondary wall and L = lumen (Salmén 2009). The striations indicate the direction of the cellulose microfibrils.
Due to the yearly seasons the tree will consist of early- and latewood fibres. Earlywood fibres are grown during spring and latewood fibres later in the season of growth. Earlywood fibres have thin walls, large lumen and their main function is to transport water and nutrients while latewood fibres have thicker cell walls and smaller lumen with the main function to give mechanical support to the tree. The differences in morphology means that their requirement in refining is different. However, in a commercial pulp these fractions are not generally separated.
In Scandinavia, the major wood resource is softwood, pine and spruce, but pulps from hardwood trees are also used to some extent in the pulp and paper industry. In Scandinavia birch is the dominant hardwood resource. One main difference between hardwood and softwood is how the water transportation is occurring in the tree. In softwood, the water transportation is mainly occurring in the earlywood fibres via the bordered pits while for hardwood the water is transported through thin-walled vessels. The fibres in hardwood are mainly contributing to the strength and have a similar structure as softwood fibres but are smaller in dimensions.
The main wood component polymers; cellulose, hemicellulose and lignin are distributed differently in the cell wall and play different roles in the structure of native fibres. Cellulose molecules are ordered into fibrils and fibril aggregates, whose different orientation gives the fibre its tensile stiffness, strength and flexibility. The role of lignin is to support the slender cellulose fibrils and to prevent them from buckling, thereby giving the fibre high compressive stiffness and strength. Hemicellulose has been proposed to act as a coupling agent or as an intermediate between the cellulose and the lignin. The content of cellulose is usually 40-45% of most wood species and it consists of glucose units linked together by glycosidic bonds, Figure 2.2. Native cellulose consists of about 10000 glucose units (Sjöström 1981). The other
carbohydrates in the wood are called hemicelluloses and have a lower degree of polymerization, around 100-150 sugar units per molecule. The type of hemicelluloses and amounts depend on the wood species.
For softwood the content of galactoglucomannan is approximately 20% and the content of
arabinoglucuronoxylan is approximately 5-10%. Birch wood has a glucuronoxylan content between 15- 30% and a glucomannan content between 2-5%. Lignin, consisting of phenyl propane units are linked together by various bonds in a three-dimensional structure. The lignin constitutes 25-30% of the wood material and the structure differs between softwood and hardwood. These differences in chemical composition have an impact on the process conditions in the pulp fibre line but also during the papermaking process.
10
a) d)
b)
c)
Figure 2.2 a) cellobiose units builds the cellulose chain b) arabinoglucuronxylan c) galactoglucomannan d) lignin macro molecule (Sjöström 1981).
During chemical pulping the main part of the lignin is dissolved in the cooking liquor. As the
delignification proceeds, the dissolution of lignin and hemicelluloses results in a porous cell wall structure.
The creation of pores makes the fibre flexible and compressible which is a prerequisite for good fibre bonding. The flexibilization is then further increased during the refining process. As the fibre gets more flexible during the refining event, the outer layers of the fibres are disrupted and loosened and a “hairy”
appearance, due to liberated microfibrils, is created.
Flexibilization of the fibre is required to achieve good bonding between the fibres. As the fibres get flexible, they swell. The swelling is not desirable as it impacts the dewatering in the paper machine but is a consequence of the flexibilization process. In the paper making process the flexible fibres together with,
“glue material”, create a paper sheet with good mechanical strength (e.g. tensile strength). The “glue material” can either be a hairy fibre surface with material partly attached to the fibre or fine particles or fine materials that have been released from the fibre. Understanding of the swelling behaviour of the fibre material in the paper making process is essential since it creates the major part of the mechanical strength of the produced paper.
2.2 Fibre swelling
Swelling is generally considered as a process in which the dimensions of a material is enlarged as the cohesion of the material is reduced followed by a softening and increased flexibility of the material. In pulp and paper making the swelling has often been defined as the water holding capacity of the material, i.e. grams of water per grams of solid contents, either as g/g or as %.
In a simplified way, the swelling of the fibre wall is a result of three parameters; a non-ionic contribution, the osmotic pressure created by the charged groups in the fibre wall, and the restraining forces caused by the network structure of the fibril aggregates in the different cell wall layers (Sjöstedt 2014). The non-ionic contribution to the swelling of the fibre wall is connected with the association of water molecules to the hydroxyl groups of the cellulose and hemicellulose within the fibre wall.
11
A schematic model on how fibrillation of the fibre wall takes place when the cell wall swells has been presented by Scallan (1974), Figure 2.3. Internal fibrillation leads to separation of lamellae in the fibre wall (Page, DeGrace 1967; McIntosh 1967; Page 1989). Water can enter into the spaces created between the lamellae thus swelling the fibre wall, which can be measured as an increase in WRV (Water Retention Value). A higher WRV is often associated with greater fibre flexibility and more conformable fibres, which leads to higher tensile strength. Increased swelling can be achieved either by refining of fibres or by chemical modification of the fibre wall material.
Figure 2.3. Schematic view of how the fibrillation of the fibre wall takes place when the swelling is increased (Scallan 1974).
2.2.1 Fibre charge
Swelling of fibres is depended upon the ionizable functional groups of the wood polymers. Carboxylic acid groups are the primary functional groups that are ionized during normal papermaking conditions.
The majority of the carboxylic groups are uronic acid groups mainly attached to the xylan. During kraft pulping the content of the carboxylic groups decreases due to dissolution of the hemicelluloses (Sjöström 1989). Hardwood have a higher charge content due to a higher xylan and uronic acid content. During kraft pulping phenolic carboxylic acid groups can also be formed (Sarkanen, Ludwig 1971) which can at high pH-levels contribute to the swelling properties of the fibres. The carboxylic groups give the pulp fibres an anionic charge.
Cellulosic fibres can be said to behave like polyelectrolytic gels due to their anionic charges. The charges within the gel cause electrostatic repulsion that cause swelling of the fibres. The fibre wall swells in different ways depending on the conditions of the surrounding environment.
There are three different locations where water can be retained by a fibre, Figure 2.4 (Salmén, Berthold 1997). Water can be found in pores in the fibre wall; pores that are present in the wood fibre or have been created when lignin and hemicelluloses were removed during pulping or created during refining. Water can also be found as bound water that is associated with the hydroxyl groups and the charged groups of the wood polymers. The counter-ions also affect how much water is associated with each charged group.
Water can also be found on the surface of the fibre and attached to the fibrils on the surface.
12
Figure 2.4. Schematic illustrating the water retained by the fibres (Salmén, Berthold 1997).
Charged groups, mainly carboxyl groups present on the hemicellulose or in the oxidized lignin are located both inside the fibre wall and on the fibre surface. The amount of charged groups available on the fibre surface is commonly less than 10% of the total charge of the fibres and is referred to as surface charge and the charges within the fibre wall as bulk charges (Fors 2000; Pettersson et al. 2006). The introduction of charges changes the swelling behaviour of the fibre as well as the pore size distribution within the fibre wall (Salmén, Berthold 1997).
The fibre wall can thus, as said earlier, be seen as a polyelectrolytic gel and its swelling behavior may be explained by the Donnan theory. This theory applies to semi-permeable membranes and the fibre wall can be seen as such, since the negatively charged groups attached to the fibre cannot move freely, whereas the counter-ions can move in and out of the fibre wall. If the water contains cations, these will be attracted to the charged groups and there will be an accumulation of cations within the fibre wall (Towers, Scallan 1996). Since the system will try to overcome the difference in concentration between cations in the fibre wall and in the bulk liquid outside the fibre wall, water molecules will migrate into the fibre wall, thus swelling it. The swelling behavior can also be explained by electrostatic interactions and at higher pH ionisable groups will be deprotonated. The negative charges then repel each other leading to a swelling of the fibre wall (Lindström 1992). However, the swelling do not lead to dissolution of the fibre due to restraining forces in the fibre wall from the fibril aggregates arranged in a network structure.
As shown in Figure 2.5, the amount of charged groups influenced the degree of swelling as do the restraining forces which decreases when pores are formed due to lignin removal. The impact of pH and electrolyte concentration (in this case NaCl concentration) is also shown. Electrolyte concentration affect the swelling since the osmotic pressure depends on the concentration gradient of mobile ions; an increase in the electrolyte concentration reduces this concentration gradient and thereby the osmotic pressure (Lindström, Carlsson 1982).
a) b) c)
Figure 2.5. The effect of pH and NaCl concentration on WRV for a) unbleached sulphate pulp with a kappa number of 145 b) unbleached sulphate pulp with a kappa number of 38 c) bleached sulphate pulp. Softwood kraft pulps (Lindström, Carlsson 1982).
13
The type of counter-ion to the charged groups is also important for the swelling effect of fibres. Scallan and Grignon (1979) found that the swelling ability increases in the following order:
Al3+ < H+ < Mg2+ < Ca2+ < Li+ < Na+.
An increase in valency of the counter-ion reduces the charge interactions; simplified one could say that multivalent cations interact with more than one charged group and draw them closer which results in reduced swelling, Figure 2.6. Thorium with the highest valency showed the steepest decrease in WRV as a result of salt concentration, lanthanum the second steepest and sodium as monovalent ion was least sensitive towards salt concentration (Lindström, Carlsson 1982).
Figure 2.6. WRV for an unbleached sulphate pulp (yield 52.8%) versus molar concentration of salt with different cation valencies (anionic NO3-) (Lindström, Carlsson 1982).
The type of counter-ion also affects the swelling between the wood polymers themselves. Type of counter-ion to the charged groups have an influence on pore size distribution (Salmén, Berthold 1997). If an unrefined pulp is ion-exchanged from the dissociated Na+- form to H+-form the amounts of small pores is reduced, Figure 2.7. This indicates that it is mainly the small pores where the electrostatic repulsive forces are active.
Figure 2.7. Apparent pore volume in different pore size intervals for an unrefined kraft pulp with kappa number 85 in Na+-form. The darker parts of the bar represent the net effect on the distributions of ion-exchange to H+-form. The difference in pore volume due to the ion-exchange is given for each pore size interval (Salmén, Berthold 1997).
14 2.2.2 Measuring fibre swelling
Swelling is generally defined as how much and where water is retained in the fibre material. The most common way to measure the water holding capacity of the material is the WRV, Water Retention Value.
WRV is a method which measure the amount of water remaining in the fibres during controlled and standardized temperature, time, centrifugal force. The increase in WRV as a result of refining is mainly an effect of that new fibre material surfaces are created during mechanical treatments. The conditions of the WRV measurements have been adjusted to match the Fibre Saturation Point (FSP) whose fundament is more based on physics. It is also possibly to quantify surface swelling of a fibre material and bulk swelling by calculating the difference in WRV for a pulp in its sodium form and the WRV for the pulp when the surface charges have been blocked with high molecular weight poly-DADMAC (Swerin et al. 1990; Fors 2000).
The most common method to measure swelling in conjunction with refining is the SR-number, which is a drainage test. The SR-number increases with refining and is mainly related to the fibre surface area, but also sensitive towards the fines content. Therefore, if fibre treatment is in focus, WRV is to be preferred as a swelling measure (Bokström et al. 1981).
In addition to SR and WRV, the Fibre Saturation Point (FSP) is also widely used. In FSP measurement, wet pulp of known moisture content is immersed in a dilute aqueous solution of a high molecular weight dextran or similar non-interacting polymers. The polymer molecules are larger in size than the pores and cannot enter the cell wall. Thus, water contained in the pores will lead to a change in the polymer concentration. The FSP, which can be seen as the amount of water in the cell wall, is then determined using polarimetry. Salmén, Berthold (1997) has also used inverse size exclusion chromatography (ISEC) to determine the swelling. During the inverse size exclusion chromatography, the pulp is packed in column and the retention of dextranes is evaluated.
2.3 Influence of the polyelectrolytic gel nature of fibres on papermaking properties
The importance of the polyelectrolytic gel behaviour can be seen in several of the unit operations in papermaking, all from stock preparation, wet end chemistry, forming, pressing and drying. In principle everything that is related to how the fibre respond to fibre-water interaction and to the fibre-fibre interaction can be related to its polyelectrolytic gel nature.
2.3.1 Stock preparation
In the stock preparation, i.e. the preparation of the pulp suspension to be delivered to the paper machine, the refining process is the main unit operation and it was earlier found that stronger sheets were formed from alkaline stocks and that the refining rate increased more at alkaline conditions than during acidic conditions (Jayme, Büttel 1964; Jayme, Büttel 1966; Laivins, Scallan 2000).
The counter-ions to the charged groups have a profound influence on the strength properties as illustrated in Figure 2.8 where the effect of different cations on the swelling of unrefined fibres and on the strength properties made of paper from them is seen (Scallan, Grignon 1979).
15
Figure 2.8. The effect of cation exchange on the swelling of unrefined pulp and paper made from them (Scallan, Grignon 1979).
The effect of electrolyte concentration and pH during refining is also known to have an influence on the sheet properties. An increased pH during refining increases the refining efficiency, which results in stronger sheets (Jayme, Buttel 1964). It is also known that an increased concentration of electrolytes decreases the refinability, improves the drainage and decreases the strength properties of the paper (Cohen et al. 1949; Cohen et al. 1950; Scallan, Grignon 1979; Grignon, Scallan 1980; Lindström, Kolman 1982;
Lindström 1986; Lindström 1992).
The influence of the fibre's polyelectrolytic nature during refining at different pH and different electrolyte concentration was elegantly demonstrated by Lindström and Kolman (1982). They showed the
importance of keeping control of the chemical environment to understand and to interpret the refining results, Figure 2.9. The effect of pH and electrolyte concentration during refining can be explained, to a large extent, if the cell wall is considered to be an ionic gel with fixed charges (acidic anions). Swelling is controlled by the amount of these acidic groups and by the balance of the elastic forces of the gel and the osmotic pressure difference between the gel phase and the surrounding medium. The swelling maximum can be determined by the type of acid groups present in the cell wall. Charged groups with higher pKa than the carboxyl groups shift the swelling maximum towards higher pH values. It is therefore important to separate the effect of electrolyte concentration and the effect of pH. The maximum swelling occurs at a pH-interval of 8-10 for the unbleached pulp at low electrolyte concentration while the maximum swelling for the bleached pulp was not so pronounced due to its low content of charged groups.
The change in pH is usually done by addition of an electrolyte, such as NaOH, Na2CO3, H2SO4. Initially, the increased pH usually leads to an increased swelling because of the dissociation of the acidic groups in the fibres. As the pH increases the electrolyte concentration increases and may cause a reduction of swelling due to the shielding of the charged groups according to the Donnan theory (Lindström 1992).
16
a) b)
Figure 2.9. The effect of pH and electrolyte concentration during refining of a) unbleached kraft pulp, yield 50.9% and b) bleached kraft pulp on the WRV measured under different chemical conditions. Refining was performed in deionized water at different pH values and the WRV was measured under the same chemical conditions. within circle: The same as before, but WRV was measured after washing with deionized water, pH-4 buffer and deionized water (H+-form). Refining was performed in 0.1 M NaCl at different pH-values and the WRV measured under the same chemical conditions. within circle: The same as before, but the WRV was measured after washing with deionized water, pH-4 buffer and deionized water (H+-form) (Lindström, Kolman 1982).
The polyelectrolytic gel nature can also affect the rheology of the fibre suspension and in particular the strength and the size of flocs which have an impact on refining as discussed below. Fibre length and fibre consistency are the most important parameters for floc strength, measured as the apparent yield stress (Kerekes, Schell 1995; Dalpke, Kerekes 2005).
An increase in the number of charged groups decreases the floc size and was interpreted as an effect of electrostatic repulsion between the charged fibres that decreases the coefficient of friction between the fibres and reduced the network strength (Beghello 1998; Yan 2004; Horvath 2006). Neither pH nor electrolyte concentration have any great impact on the flocculation and fibre network strength (Beghello, Eklund 1999; Horvath 2006). According to Bjellfors et al. (1965), the shear strength and shear modulus of a fibre network are influenced by the counter-ion to the charged groups, where fibres in the Na+-form have a slightly lower shear modulus than fibres in H+-, Ca2+-, Al3+- or Th4+-forms. A tentative explanation is that the cations affect the interaction between fibres and the flocculation process, and thus influence the shear modulus of the network.
Industrially the electrolyte concentration during refining is affected by the degree of system closure of the paper mill. According to measurements in two Swedish paper mills producing unbleached kraft papers the electrolyte concentration in the white water before and after refining was rather low, around 0.005 mol Na+ /l. The content of Ca2+ was even lower but due to its valence the contribution from calcium to ionic strength would approximately correspond to the contribution from sodium. A metal analysis was also
17
performed on the pulp fibres. The result confirmed that industrial pulp fibres are in Ca2+-form (Hammar et al. 1997).
2.3.2 Consolidation
The polyelectrolytic gel nature of the pulp fibres has also been observed during the paper consolidation process. Torgnysdotter (2006) characterized when two fibres physically cross each other and a fibre/fibre joint is created and how paper strength was influenced. The formation of fibre/fibre joints was affected by the type of counter-ions to the charged groups. The change of counter-ion from Na+ to Ca2+ or H+ drastically reduced (by more than 50%) the joint strength and this was ascribed to a loss in surface swelling with decreasing degree of dissociation of the carboxyl groups in the fibres, while the observed smaller influence on the tensile index (between 15-20%) was suggested to be due to more efficient fibre/fibre contacts being formed due to a decreased electrostatic interaction between the fibres during the consolidation of the sheet (Forsström et al. 2005). It is worth noticing that in these studies, unrefined fines free pulps were used.
The polyelectrolytic gel nature of the pulp fibre can also be found when pressing the wet paper web.
Addition of cationic polymers to the wet web prior to pressing reduced the fibres swelling and the WRV value which enhance the dewatering and increased the out-going solids content after the pressing stage (Swerin et al. 1990), Figure 2.10.
Figure 2.10 Relation between maximum solids content (%) during wet pressing and WRV (%). Different degrees of deswelling reached by cationic polyelectrolyte (polybrene and poly-DMDAAC) on different pulp fibres are shown (Swerin et al. 1990).
Hornification is a structural change that is usually observed as a loss in swelling and plasticity of the fibres when they are rewetted after drying. The amount of charges and ionic forms of the pulp fibres affects the degree of hornification that the fibres retrieve during the drying process. Lindström and Carlsson (1982) showed that hornification could be prevented by introducing carboxylic acid groups within the fibre and ensuring that the counter- ions was sodium, Figure 2.11. They suggested that the introduction of ionized carboxyl groups interrupted the cooperative hydrogen bonding between cellulose lamellas in the fibre wall.
If the carboxyl groups were protonated they participated in the hydrogen bonding between the lamellas and thus caused hornification.
18
Figure 2.11. Swelling, measured as WRV, for a bleached chemical softwood pulp carboxymethylated to different degrees. A comparison is made between never dried pulps and pulps dried in different ionic forms (Lindström, Carlsson 1982). • Never dried fibre, o After drying of the fibres in their H+-form, ∆ After drying of the fibres in their ionized form.
2.4 Refining
Pulp refining is one of the most important unit operations in papermaking. Refining of chemical pulps is a mechanical treatment with the aim to modify the fibres so that they can be formed into paper, board or tissue grade with desired properties. The most common way is to treat fibres in a water suspension at a consistency of 2-5%, normally called low-consistency refining (LC). For certain grades other consistencies may be used, e.g. sack paper which normally includes a high consistence refining stage (HC). In the refiner the fibres are treated between two segments, one stationary called stator and one that rotates, called rotor.
The equipment for refining can be divided into two main categorizes; laboratory and industrial refiners.
The most common laboratory refiner is the PFI-mill, Figure 2.12. It is a high energy and low intensity refiner (Kerekes et al. 2005). The pulp is refined at a pulp consistency of 10%. 30 gr o.d. pulp is refined per run. PFI-refining is considered to give a very gentle treatment of the fibres with little fibre cutting and high degree of internal fibrillation.
Figure 2.12. PFI refiner at RISE Innventia AB.
Another laboratory refiner that more resembles industrial refiners is the Escher-Wyss conical refiner where a minimum of 600 g o.d. pulp at a consistency of 2-4 % can be refined by multiple passages through the refiner. Escher-Wyss is now supported by Voith Paper who supplies laboratory refiners that can both be used with conical as well as disc-fillings.
Two large categorises dominate the industrial market, disc and conical refiners, Figure 2.13. The disc refiners can either be single disc or multidisc refiners. Today single disc refiners are rare and are almost
∆ ∆
∆
∆
o o o o
19
exclusively used for high-consistency refining. The disc diameter is normally between 18 inch (467 mm) to 58 inch (1473 mm) and the power ranges from 200 kW to 3000 kW. In a conical refiner the conical angle is between 10-30° and the applied forces can be considered to be slightly different from disc-refining due to that forces acts slightly differently because of the conical shape.
a) b)
Figure 2.13 a) conical refiner and b) disc refiner.
The design of the fillings is very different for different furnishes, i.e. there are fillings suitable for
softwood and fillings suitable for hardwood. The refining fillings are constantly worked on by the supplier companies to reduce the energy input, improve property development and improve production
robustness. In Figure 2.14 some examples are given to show how different the refining fillings can be.
Figure 2.14 a) Refiner segment, The Wall Disperger filling (Voith) b) Pluralis Line (Voith) c) Conical refiner fillings from AFT.
2.4.1 Refining effects on fibre characteristics
The purpose with refining is to modify the fibres to obtain desired paper properties, for instance, tensile strength or surface smoothness. The mechanical treatment in a refiner makes the fibres more flexible by creating internal fibrillation as the fibril layers in the fibre wall separate and additional pores are created.
This leads to increased water holding capacity of the fibre wall. In addition, fibrils on the fibre surface are loosened and may extend from the surface, producing external fibrillation. Some of the fibrils are entirely separated from the fibre, either as single fibrils or as flakes, giving rise to fines material. Refining also affects the fibre shape, either by straightening them or by instead make them curlier. The mechanical treatment can also lead to fibre shortening as fibres may be cut by the refining bars. All these effects may occur during the refining event. For different paper grades different refining effects are desirable, in some cases fines production is undesired while in some cases it is preferred. The refining effects on the fibres have been nicely summarized by Page (1989), Figure 2.15.
20
Figure 2.15 The effects of refining on the fibres based on Page (1989) and Nugroho (2012).
The main effects on the fibres during refining is fibre shortening, fibre straightening, internal, and external fibrillation and fines generation. Fibre shortening is something that one normally wants to avoid and as a rule of thumb, fibre cutting occurs when the fibre length has been reduced by more than 10%. Fibre straightening in refining of chemical pulps was first noticed for laboratory refining using the PFI-mill (Page 1989). Mohlin (1991) showed that during refining in a PFI-mill fibre straightening always took place if the fibres initially was curly. In industrial refining, both fibre curling and fibre straightening may take place. If fibres remain curly or kinked in a paper network, they must be stretched before they can carry load, meaning that fewer fibres carry load during the initial stretching of paper, and therefore the paper is weaker. Internal fibrillation results in an increased tensile strength due to an increase in bonded area in the paper sheet as the fibres become more flexible and thus more conformable to other fibres (Scallan, Tigerström 1992). External fibrillation can be defined as a peeling off of fibrils from the fibre surface, while leaving them attached to the fibre surface. This material is important in the consolidation process to enhance the contact between fibres and strengthen the fibre joint. It is also of importance for the press event in the paper machine where one could interpret that fibres that are mainly externally fibrillated ought to be easier to dewater than fibres that are mainly internally fibrillated. External fibrillation increases the tensile strength by increasing the surface area available for fibre-fibre contact (Clark 1969; Kang 2007).
Fines are generated by the abrasion of fibres either with other fibres or with the refiner bars. Fines from chemical pulps usually consists of fragments of the P and S1 layers of the fibre wall.
2.4.2 Refining mechanism
Historically, the specific edge load (SEL) theory has been very common and normally used as a base in scientific or development work related to refining and to evaluate the refining response of different fibre materials. The SEL can be regarded as the amount of effective energy spent per unit length of bar crossing (Brecht, Siewert 1966). SEL is defined:
SEL= (P- P0)/ L where:
SEL = specific edge load (J/m) P= total refining power, kW
P0= idling power (no-load power), kW L= cutting edge length per time, km/s
There are some limitations with the SEL theory since it does not consider pulp consistency, bar and groove dimensions and gap clearance. However, it has despite that been proven to be a useful method to compare pulps. Specific refining energy (SRE) is used to describe how energy is given from the refiner to the fibres and is calculated by dividing the net power (P- P0) with the fibre mass flow rate.
21
The common view is that the refiner segments are designed so that the bars treat the fibres and the space between the bars, i.e. grooves, are for fibre transportation. Lumiainen (2000) presented in a schematic view how the different refining stages could be viewed, Figure 2.16. At first fibre flocs are collected at the leading edge of the bars. When the leading edge of the rotor bar approach the leading edge of the stator bar, the floc is compressed and receives a strong hit which compress most of the water out from the floc and the fibres within the floc receive a refining effect. Some fibres escape with the water and receive no or little mechanical treatment. The refining includes tensile, shear and compression forces and the general view is that the fibres are treated in flocs as suggested by e.g. Ebeling (1980). Fibres are not free to move independently but are transported in flocs that are broken up and formed continuously due to the shear forces in the refining zone. The thickness of individual flocs is much larger than the gap clearance. This reduces the probability for the flocs to be driven into the gap.
Fibre Pick up Edge to edge fibre treatment
Edge to surface fibre treatment
Surface to surface fibre treatment
Surface to surface fibre treatment
End of refining
Figure 2.16. Refining mechanism (Lumiainen 2000).
The refining fillings are important for the fibre property development and have a profound effect on the probability of fibre treatment. The design of the fillings will determine the available refining sites and it is not clear if it is the bar length or the number of bar crossing that is important. Roux has suggested that refining occurs at bar crossings (Roux et al. 2009). The direction of bars and grooves deviates from the radial direction and the bar angle has a large effect on the number of bar crossings, (Mohlin 2011b), Figure 2.17. The picture is a computer-created image of the cross-over pattern of two identical fillings;
white represents bars in one filling and red represents the areas where bars from the other filling cross over. An increase in bar angle will greatly increase the number of bar crossings and thus the refining sites, resulting in a larger gap clearance at a given power consumption (Mohlin 2011b).
22
a) b) c)
Figure 2.17. a) Schematic drawing of a refiner plate segment where the bar angel is shown b-c) The effect of bar angle on the number of bar crossings. The pictures show a computer-created image of the cross-over pattern of two identical fillings;
white represents bars in one filling and red represents the areas where bars from the other filling cross over. Picture to the left shows a filling with a small bar angle and to the right a filling with large bar angle. Bar and groove width are the same (Mohlin 2011b).
The refining mechanism has also been studied by Mohlin (2011b) by high speed imaging proposing a fibre trapping mechanism, Figure 2.18. The fibre trapping on the stator bar edges is initiated by the rotation of the suspension in the groove throwing fibres towards the bar edge. The fibre floc become locked into position on the bar edge by the shear field from the passing bar acting along the bar surface and also from the shear field from the rotating body acting downwards along the groove side. The fibre flocs are on the bar edge for a few bar passages. The fibre floc is not completely fixed but move outwards along the bar edge during the event. Eventually, the shear forces pull the fibre floc out of the groove and push it over the bar down into the next groove. The fibre flocs will thus be under tensile strain as long as they are positioned at the bar edge.
Figure 2.18. Fibre trapping mechanism on stator bars as deduced from high-speed imaging (Mohlin 2011b).
According to Mohlin and Roos (2007) the gap clearance is the main variable to control the power to be applied in the refining and thus it controls what type of treatment the fibres will receive. An
experimentally determined power – gap clearance relationship will give information about the interaction between the refiner and pulp suspension (Mohlin 2011b). In the pulp refining a typical refiner gap is 0.1- 0.2 mm which is substantially wider than any fibre width. This may be interpreted in the way that the fibres trapped between the bars are not individual fibres but fibre flocs.
Goosen et al. (2007) has suggested that the role of multiple loading cycles that occur during refining is to redistribute the fibre network between the loading cycles and proposed a cumulative probability model for
Fibre trapping mechanism
1. Fibre floc is rotating in groove
2. Anchored on the bar edge by the shear field
3. Stays on the bar edge for a number of bar passages – tensile strain
4. Shear field pulls fibre out of groove and across bar.
5. Fibre is moved over to next groove
1 2
5 4
3
Stator
Centre of refiner
23
tensile strength increase. This suggests that the role of cyclical loading is largely one of exposing new fibres to loading cycles rather than producing fatigue weakening (Goosen et al. 2007; Wang 2006; Heymer et al. 2011).
However, as the pulp passes through the refiner, not all fibres experience the same mechanical treatments.
Heymer (2009) suggested that heterogeneity in refining is governed by the probability of the fibre to be captured and transported into gaps and the probability of suitable forces applied to fibres within the gaps.
Several investigations indicate that only a small part of the fibres is treated during the refining (Halme 1962; Ryti, Arjas 1969; Steenberg 1963; Lidbrandt, Mohlin 1980; Page 1989; Olson et al. 2003; Decker 2005; Batchelor et al. 2006; Goosen et al. 2007; Heymer 2009). Decker (2005) reported that only 7 to 20%
of the fibres are treated in an industrial disc refiner.
On a laboratory scale, the pulp is usually treated by multiple passages. Particularly PFI-refining is believed to treat the pulps homogenously (Dillén 1980; Lidbrandt and Mohlin 1980). Lidbrandt and Mohlin (1980) showed from scanning electron microscopy that most fibres in the PFI-mill were treated, whereas most of the fibres in an industrial refiner were not.
2.4.3 Property development during refining
The overall goal of the refining is to give the fibres desired characteristic so that papers with targeted mechanical properties can be produce in an energy efficient way. Mohlin (2011a) made a list of how the swelling variables SR-number and WRV relates to the mechanical properties, Table 2.1.
From the table one can deduce that refining cause external fibrillation which increase both WRV and the SR number, and are very essential for the mechanical property development of the paper produced.
Internal fibrillation on the other hand, increase WRV, but have not a large impact on the SR-number, but is important for the mechanical property development. Mechanical properties of interest are usually tensile strength, elongation and stiffness in relation to the density of the produced paper.
From a pulp characteristics point of view, the pulp producer gives technical specification for their pulp which usually involves a PFI-refining curve, including drainability (SR-number), tensile index, tear index, light scattering coefficient, air permeance and density. This could differ a bit depending on what market segment the pulp is aimed for.
Table 2.1. Effect of fibre characteristics on WRV and SR-number based on Mohlin (2011a).
WRV SR number Mechanical
properties
Internal fibrillation + 0 ++
External fibrillation + + ++
Fines particles (not fibrils) + ++* +
Fibre kink 0 0/- -
Fibre charge + 0 +
Fibre collapse 0 + +/0
Fibre coarseness 0 + 0
Fibre wall pore collapse - 0 0
*effect is larger at high level of fibre treatment than at low level of fibre treatment.
