Paper Making a Glance

7/27/2019 Paper Making a Glance

http://slidepdf.com/reader/full/paper-making-a-glance 1/12

9

2 Te paper production process at a glance

Although the ocus o this thesis is on wet-pressing, the dierent parts o the productionprocess are related in such a way that some background inormation on the whole

process is necessary to understand the experimental approach. In the ollowing an

attempt is made to provide this background inormation. Tis inormation is by no

means complete and interested readers are reerred to the Fapet Oy series, Paper-

making Science and echnology, or other handbooks on paper-making technology.

In this chapter the dierent stages o the paper-making process are described, rom the

bre source, the pulping method and the paper machine to the paper characteristics

o the nal sheet. Te characterisation o the paper-making process on the papermachine occurs at dierent size scales:

• Paper machine 0. - 00 meter scale

• Nip dimensions 0.05 mm - meter scale

• Pores 0.0 - 50 micron scale

In the next chapter we will ocus on how these dierent scales are related.

2.1 Fibre sources and pulping methods

Paper and board is made o bres. Fibres are made o dierent types o polymers:

cellulose, hemicellulose, lignin and natural resins. Te cellulose and hemicellulose

orm twines. Te twines orm a layer. Te layers are covered with lignin and resins,

giving strength and stiness. Several layers together orm the bre, c. gure -.

When twines o (hemi)cellulose are partially reed rom the bre structure they are

called brils. Fibrils are very important to bre bonding.

Te characteristics o these bres have a signicant inuence on the properties o the nal product. Te properties o the bres are determined by the bre source and

pulping method.

Te most used bre sources are wood and recycled paper. Te rst source yields

so-called virgin bres and the second source recycle bres. Te process o releasing

bres rom the wood or paper is called pulping. Te pulping method depends on the

bre source.



0

When recycling paper, bres are released by dispersing the paper in water. Te

pulping method consists o dispersing the collected paper afer which the dispersed

bres are puried rom ink and dirt using cyclones and screens, and rened to get

sucient active bre surace.

Contrary to bres in paper, bres in wood are held together by chemical bonds in

a matrix structure. Lignin molecules, i.e. a type o three dimensional polymer and

resins, act as glue. Tereore, virgin bres are not so easily released as secondary

bres. Basically, two approaches exist to pulp wood: mechanical and chemical. Tese

methods dier mainly in the extend to which the lignin and resins are removed and

the degree to which the other polymers are aected.

In mechanical pulping the bres are torn apart orm the wood by pressing the woodagainst a ast rotating grinder. Te yield o this pulping process is high (95-99%). Te

pulp obtained by this method is characterised by:

• Damaged bres

• High nes (small bre parts ) content

• High degree o brillation (strings o cellulose molecules partly reed rom the bre wall)

• High percentage o bres with a thick and sti bre wall.

• High lignin content

Tereore, paper made using mechanical pulps has a low density combined with a

high stiness, but at a low strength. Additionally, these papers yellow rapidly.

In chemical pulping the wood is cooked under high pressure in a very acid or a very

alkaline solution. Te yield o this method is about 50 to 60%. Te pulp obtained by

this method is characterised by:

• No bre damage afer the alkaline sulphate process. Afer the acid sulphite process

the cellulose chains may be damaged

• Long exible bres with a relatively thin bre wall.• Low lignin content and reduced hemicellulose content.

• Low degree o brillation. o obtain sucient brils these pulps are ofen rened

afer pulping.

Tereore, papers made o chemical pulps have a high surace smoothness, a high

strength, and a low bulk. Afer bleaching they posses a high level o whiteness and

a low tendency or yellowing. Paper made o sulphate pulps oers particularly high

strength, while sulphite pulps may yield the highest whiteness.

| Yield - Production o pulp bres expressed as a mass percentage o the wood supplied to the process.



Te sulphate process is the most widely applied chemical pulping process (Korpeinen

and Ainamo 00). Tis is an alkaline process, during which a pH o can be

reached at temperatures o 60-80 oC and pressures o -6 MPa. Pulp resulting rom

this process is called kraf pulp

. Due to the chemical reactions occurring duringkraf pulping, the pulp may darken, so bleaching is necessary or all printing grades.

Termo-mechanical pulp (MP) currently is the most widely used mechanical pulp.

It is called thermo-mechanical pulp since the pulp is heated with steam beore being

torn apart in a rener. As a result o the applied heat the bres are more easily reed

rom the wood matrix, reducing bre damage and coarse bre content.

o improve whiteness, mechanical pulp may also be bleached.

Apart rom the pulping method, the bre source also has a signicant eect on bre

properties. Wood rom deciduous trees, also known as hard wood, generally yields

coarser and shorter bres than those obtained orm coniers (sof wood). Additionally,

large dierences exist between dierent types o hardwood and sofwood, or even

between samples o a single type o timber grown under dierent circumstances.

Te properties o recycled bres vary greatly according to the source. Te three main

categories are: Mill Broke, Post Industrial Waste (PIW) and, Post Consumer Waste

(PCW).

• Mill broke is reused paper that was produced o specication. Tis may be paper

rom any stage in the production process, rom the wet web directly o the wire to

nalised paper that has been incorrectly cut.

• PIW consists o paper collected rom oces or printing shops. Tis is well-sorted

paper, normally made o virgin bres. Consequently, bres obtained rom PIW will

be more similar to virgin bres.

• PCW is paper collected rom households. PCW normally consists o a mixture o paper and board, with varying percentages o recycled paper.

Each time bres are reused, the bre wall degrades a little urther, until it is completely

worn down. Degradation o the bre walls o mechanically pulped bres may change

the properties o these bres to more closely resemble chemically pulped bres.

Further recycling will inevitably results in loss o bre quality.

| Kraf is German or strength. Te German developer Dahl called his invention the kraf process inreerence to the high bre strength.



2.2 Paper machine

Afer bres have been pulped the pulp is diluted to a highly aqueous mixture o

-% bres by mass. Additives are added, and the pulp is sent to the paper machine.Te unction o the paper machine is to separate bres and water in such a way that

a sheet with the required properties is ormed. Tis is achieved in three steps. First

the orming section, ollowed by the press section and the dryer section, c. gure -.

More modern versions o the orming section and the press section are shown in

gure - and gure - respectively.

Depending on the type o paper machine, machine speed varies between 5 and 000

m/min, i.e. 0. and m/s, while paper machine widths vary rom to meters.

2.2.1 Formation:

Te aqueous solution is spread over a wire, and drained by gravity and suction. Te

aim o the ormation process is to orm a web with an even bre distribution and to

remove 90-95% o the water. Afer ormation the web’s dry content is 0-0%. Tis is

enough to give the wet web sucient strength to support it’s own weight. Te oldest

ormation device is the Fourdrinier table shown in gure -. Te draw-back o this

method is that a density gradient may occur over the thickness o the sheet. Modern

machines dewater almost instantly between two wires, allowing or aster dewatering

and a more symmetrical sheet structure, c. gure -.

2.2.2 Wet-pressing

During ormation the pulp orms a wet capable o supporting it’s own weight. However,

since the wet web is very sensitive to stretching in the machine direction, the web is

supported as much as possible while moving through the wet-pressing section.

Conventionally, the wet web is orced to dewater by leading it through several nips.A nip is the contact area between two rolls pressed together, c. gure -5. In this

conguration the aperture o the nip is variable and the load is a setting parameter.

Te load is applied to the shafs o the rolls, as a results o which the rolls tend to bend

slightly in the middle. Te crowning o the rolls is adapted to keep the applied pressure

prole even over the cross-direction (direction parallel to the machine width).

ogether with the web a elt moves through the nip. Tis elt serves a double unction:

it supports the web between two consecutive nips, and inside the nip it provides an



escape route to the water orced rom the web. o prevent slipping between the web

and a elt, elts move through the nip(s) at the same speed as the web.

Te elt’s name derives rom the act that it used to be made o elt

. Nowadays itis usually made o polymers, mainly nylon, and consists o a combination o non-

woven abrics needled on top o a woven substrate.

In addition to the roll nip described above, extended nip presses (ENP) or shoe nip

presses are in use. Te advantage o these nips is that they have an extended nip

residence time, resulting in improved dewatering without increasing the pressure.

Tis extended nip residence time is realised by replacing one o the rolls by a load

shoe, providing a comb-shaped contact surace to the mating roll. Since the load

shoe is static, a belt moves over the load shoe at the same speed as the elt to decreaseelt attrition c. gure -6. Figure - shows an example o a modern press section

consisting o a combination o ENPs.

Pressures applied in the nip vary between .0 and MPa. Te pressure increases

with increasing dry content o the paper. Te temperature varies between 5 and 75 oC

(Dahl 989). Te pressure applied depends on the type o paper being produced, the

nip length, and the dry content o the web when entering the nip. Te nip length

varies between 0 and 5 mm or a roll nip press, and between 00 and 50 mm or

an ENP or a shoe nip press. Te press roll diameters normally vary between 00 and

900 mm.

2.2.3 Drying

Afer wet-pressing the dry content o the wet web varies between 5 and 55%,

depending on the type o paper/board being produced. Te remaining water has to

be removed by means o evaporation. o supply the required heat to the wet web, it is

leaded over a sequence o steam-heated drums, commonly reerred to as drying cans.o ensure good contact between the cans and the steaming hot web, drying wires

tighten the web to the cans. Drying by evaporation is an energy-intensive process.

2.2.4 Finishing

Te procedures described above orm the basis o paper-making. o optimise the

web’s surace properties or printing purposes, some additional processes may be

applied such as sizing, coating, and calendering. Tese process steps are important to

| Felt is a material made by pressing together wet layers o wool.



the nal appearance and properties o the paper. Tey improve the surace structure

and variations in the base paper may be corrected to some extend. However, the

success o these nishing steps depends heavily on the quality o the base paper.

Proper process control on ormation and wet-pressing thereore remain key toproducing top quality paper and board. Tereore, the scope o this paper is limited

to the wet-pressing section and it’s eect on paper properties.

2.3 Nip dimensions

In this thesis the x-direction is taken as parallel to the machine direction (MD), the

y-direction is taken as parallel to the machine width or cross-machine direction

(CD), and the z-direction is taken as parallel to the nip height or the paper thickness.Te applied pressure in the press nip works in the z-direction.

A nip is ormed by the contact area between two superimposed rolls and nip has the

ollowing dimensions:

• Nip height: the distance between the rolls parallel to the z-direction.

• Nip length: the contact length with the paper in the machine direction.

• Machine width.

Unlike nips in polymer converting, in wet pressing the nip height is not xed. Te rolls

are pressed together by orces applied to the roll shafs. Te applied load is xed, so the

nip length and height result rom the compressibility o the material between the rolls.

Te nip geometry and the compressibility o the rolls, the elt, and the web, determine

the width o the pressure prole. Te applied load determines the maximum pressure

that can be reached in the nip. Since rolls, elt(s), belt and paper web all move at the

same speed through the nip the applied load is essentially unidirectional.

I the nip is made up using two non-compressible rolls, the nip is the contact linebetween these two rolls, while the applied load equals the load applied to the paper

on that line. In practice the rolls deorm signicantly, or they are tted with a rubber

cover that deorms signicantly, resulting in a nip length o several millimetres even

without the presence o elt and paper. Consequently, the load applied to the paper is

applied over a larger area resulting in a nite pressure.

Veenstra has shown that during calendering the actual pressure prole to which the

paper is subjected is a result o both the geometry o the press and the mechanical



5

behaviour o the paper (Veenstra 00). Te pressure in the calendar nip is described

by the Hertz laws (Szabo et al. 97), the hardness o the cover, and the mechanical

behaviour o the web. During calendering the web properties may dier signicantly

beore and afer compression. In these cases the pressure prole is not likely to besymmetrical.

Veenstra’s work applied to calendering. Te situation during wet-pressing is

signicantly dierent, since a elt moves with the web through the nip. Tereore,

the actual pressure prole is the result o the same actors that aect the calendering

process and the mechanical behaviour o the elt. Rolls may be assumed to compact

and deorm elastically. Felts and web deorm visco-elastically (El-Hosseiny 990; El-

Hosseiny 99). Veenstra has shown that the visco-elastic behaviour o paper alone

can signicantly aect the applied pressure prole during calendering (Veenstra00). During wet-pressing the deormation o the paper is ar more signicant,

not to mention the hysteresis occurring in elts during compression and expansion.

Tereore, the pressure prole can never be symmetrical.

Te aim o this work is to provide a good description o the mechanical behaviour o

the wet sheet. Ultimately this will enable the applied pressure curve to be calculated

on the basis o a description o the combined mechanical behaviour o the rolls, the

cover(s), the elt(s) and the paper.

A graphical representation o the rst attempt to describe the mechanical behaviour

o the wet sheet is given in gure -7. Tis gure shows the graphical representation

o the wet-pressing theory presented by Nilsson and Larsson (Nilsson and Larsson

968). We will use this theory as a starting point or our own work. A detailed

description o gure -7 is given in the ollowing sections.

Te top part o this gure shows a schematic diagram o a nip ormed by two rolls.

In the nip we can see the paper web moving rom lef to right. Te web enters the nipsupported by the elt.

Te press nip is the part where the paper, the elt and the rolls are in contact. As

mentioned beore, the rolls are pressed together. Tis means that the applied load is

at its maximum at mid nip, i.e. the point at which the rolls are in closest contact. Te

applied load increases strongly as the web moves towards mid-nip, and decreases as

it moves past mid-nip. Tis is indicated by the upper pressure prole plotted below

the press nip. Te pressure prole is divided into zones.



6

For each zone the erzaghi principle is assumed to apply. Tis means that the applied

load equals the sum o the hydraulic pressure and the structural pressure, i.e. the

deormation stress in the web.

Te our zones are:

. Saturation o the wet sheet. In this zone the hydraulic pressure is insignicant. Te

ull load is counteracted by the structural pressure, causing compaction o the web.

. Compaction o the saturated sheet. Te hydraulic pressure reaches it’s maximum

in this zone. Depending on the ow resistance met by the water, the hydraulic

pressure in this zone can be higher or lower than the structural pressure. I the

hydraulic pressure is higher than the structural pressure, the compaction is ow-

controlled. Tis occurs or example at the entrance o the rst nip when a lot o

water is being removed. I the structural pressure is highest the ow regime iscalled compression-controlled. Te ow in the last nip normally is compression-

controlled.

. Compaction o the wet web while pressure is being relieved. Te rate o dewatering

decreases. However, water is still being removed rom the web by the applied load,

because the time to mid nip is too short to reach the equilibrium moisture content.

. Expansion o the wet web. Te deormation stress o the wet web is assumed to

decrease slower than the applied load, causing an under pressure in the web. Tis

is the historical explanation o rewet and is called in-nip rewet. Currently, the

signicance o rewet in the nip is being denied (MacGregor 989).

As mentioned above, the compaction during wet-pressing is proportional to the part

o the load, deorming the web. Te question is whether the compaction during wet-

pressing also determines the eect wet-pressing has on the density o the web afer

wet-pressing. Tis aspect is addressed in chapter .

Based on their calculations, Nilsson and Larsson warned or the occurrence o

compaction gradients, causing two-sidedness (Nilsson and Larsson 968). Tey

reasoned that i the erzaghi principle applies, the sum o the hydraulic and structuralpressure has to equal the applied load at any height in the paper. In event o a high

gradient occurring in the hydraulic pressure due to low ow resistance at the elt side

and high ow resistance at the roll side, the erzaghi principle requires the existence

o an equally high gradient in the structural pressure in a direction opposing the

gradient in the hydraulic pressure. Tis should cause a deormation gradient over

the z-direction o the paper. In 98 Wicks and Szikla measured the occurrence o a

density gradient under controlled conditions (Szikla 99; Wicks 985). MacGregor

explained press elt marking and two-sidedness by this phenomenon combined with

| Te erzaghi principle and it’s implications or wet-pressing are described in detail in chapter ,

paragraph .



7

the relocation o llers and nes by the water moving out o the sheet (MacGregor

98). Szikla determined the exact conditions under which relocation o llers and

nes may occur (Szikla and Paulapuro 987).

2.4 Pore dimensions

Te dewatering theory o Nilsson and Larsson as described in the previous paragraph

implicitly assumes that all water is located between the bres, or that there is no

dierence between the water inside the bres and the water between the bres. In

the ollowing we will ocus on the dierences between the void raction inside and

between the bres. Te void space ormed by the pores between the bres is called

the inter-bre pore space and these pores are called the inter-bre pores. Te voidspace ormed by the pores within the bre walls is called the intra-bre pore space

and these pores are called the intra-bre pores.

Maloney and co-workers reported pore size distribution o both the inter-bre

and intra-bre pores (Maloney et al. 997). Teir measurements were an estimate

based on volumes o water that could be removed by centriuging. According to

these measurements the pore size o the inter-bre pores o bleached kraf sofwood

(BKSW) bres ranges between twenty and several hundreds o micrometers.

Kettle and co-workers studied the eect o dierent calendering methods on the

pore size distribution o a typical super calendered (SC) grade (0% kaolin ller and

pulp consisting o 8% by weight chemical pulp and 8% by weight mechanical pulp,

5 g/m). Tey ound that the combined results o air absorption and mercury

porosimetry data provided them with the most representative pore size distribution.

According to their measurements pore sizes between 0 and 00 micron are mainly a

measure or surace roughness (Kettle et al. 99). Tey ound that beore calendering

about hal o the pore volume was divided into three pore size distributions the porediameters o which ranged rom 0 to 00 micron, rom 5 to 0 micron, and rom

0. to 5 micron, respectively. Afer calendering, only one pore size distribution

remained, ranging rom 0. to 0 micron. Unortunately, they did not study the eect

o wet-pressing on the pore size distribution. Te eect o wet-pressing on the inter-

bre pore space is investigated in chapter 7 using mercury porosimetry.

Maloney and co-workers also studied the intra-bre pore size distribution. Tey used

the ollowing indirect measuring techniques:



8

• Nuclear Magnetic Resonance (NMR), and

• Solute Exclusion (SE) techniques.

According to these measurements the intra-bre pores range rom 5 to 0 nanometer(0-9 meter), depending on pulp type and pressing history (Maloney et al. 997).

Tis contradicts the current understanding o ow rom the intra-bre pores, or the

ollowing reasons:

First o all, the general opinion is that dewatering o inter-bre pores can be described

in terms o convective ow (Darcy equation) because it is ast enough to occur in

the short nip residence time o commercial machines. Te size o the intra-bre

pores as reported by Maloney is in the range were a uid cannot be regarded as a

continuum anymore. But even i one describes ow in terms o a continuum withsuch small pores the ow resistance becomes too high to allow or signicant ow

to occur during the nip residence time. ransport o material in a non-continuum is

generally assumed to be slower than ow by convection in a continuum. Tereore,

non-continuum transport rom the intra-bre pores becomes rather unlikely given

the short nip residence times o commercial machines.

Furthermore, the reported intra-bre pore-size distribution ignores the act that in

sofwood pores are ofen visible using a normal light microscope, which means that

they are tens o microns wide. We thereore assume that the intra-bre pores range

between the nanometer range reported by Maloney and co-workers and the microns

range visible in sofwood bres under a light microscope.

Currently, a lot o eort is put into studying the exact geometry o a small piece

o paper using a combination o microtomography and image analysis techniques

(Auran et al. 999; Gregersen and Niskanen 999). Te exact pore geometry at each

moment during wet pressing remains impossible to measure.

An exact geometry would allow us to determine the actual ow velocity within the

sheet as well as the actual stresses within the web. As mentioned beore, the load

applied to the web is unidirectional. However, in the contact point o two particles a

normal orce and a shear orce can be transmitted. I unidirectional orce is transerred

rom one particle to another, and i these particles do not have perectly at parallel

suraces, part o the applied orce may be diverted to shear orces. I these shear orces

exceed a certain value, i.e. the yield stress, the particles will start to slide over each

other. Tis will lead to relatively large deormations.



9

2.5 References

Auran, P., Gregersen, Ø, and Brevik, M. “ Dimage analysis o paper samples based

on phase contrast X-ray microtomography.” Characterization methods or bre and paper Joint COST E11/PTS Workshop, Munich.

Dahl, H. (1989). “Die Bedeutung des Nasspressens ür Produkt und Produktität

- Anwendung der Teorie des Nasspressens ür heutige Presse.” Wochenblatt ür

Papierabrikation, , 9-00.

El-Hosseiny, F. (1990). “compression behavior o press elts and wet webs.” Nordic

pulp and paper research journal, 5 No , 8-.

El-Hosseiny, F. (1991). “An equation or press elt compression.” Tappi Journal,

August 99, 9-96.

Gregersen, Ø, and Niskanen, K. “Measurement and simulation o paper D

structure.” Characterization methods or bre and paper Joint COST E11/PTS

Workshop, Munich.

Kettle, J., Matthews, P., Ridgway, C., and Wågberg, L. “Investigation o the pore

structure o paper by novel porosimetric techniques: application to super and sof-

nip nishing.” 10th Fundamental Research Conerence, Oxord, 55-9.

Korpeinen, ., and Ainamo, A. (00). “Look over your horizon - intelligent new

solutions or the paper network.” Know-how wire, Jaakko Pöyry Magazine.

Lange, D. “Shoe pressing o paper grades.” Tappi 1996 Papermakers Conerence,

Philadelphia, 5-8.

MacGregor, M. A. “description o sheet stratication caused by wet pressing.” 1993

practical aspects o pressing and drying seminar, 55-59.

MacGregor, M. A. “Wet Pressing in 989: An historical perspective, analysis and

commentary.” 9th Fundamental Research Conerence, Cambridge, 5-585.

Maloney, . C., Li, .-Q., Weise, U., and Paulapuro, H. (1997). “Intra- and Inter-

bre pore closure in wet pressing.” Appita, 50 No., 0-06.



0

Nilsson, P., and Larsson, K. (1968). “Paper web perormance in a press nip.” Pulp

and paper magazine o Canada, December 0, 968, 8-5.

Szabo, I., Wellnitz, K., and Zander, W. (1974). Mathematik; 2 Auf, Springer Berlin.

Szikla, Z. (1992). “On the basic mechanisms o wet-pressing,” PhD, HU, Helsinki.

Szikla, Z., and Paulapuro, H. “Changes in z-direction density distribution o paper

in we pressing.” Tappi-CPPA 1987 Paper physics conerence, Atlanta, -9.

Veenstra, P. “Te orgotten calander parameters and how to determine them.” COST

E11 WG Paper meeting on calandering, rondheim, Norway, 7.

Wicks, L. “Te inuence o pressing on sheet two sidedness.” 1985 Tappi Pressing and

Drying course.

Paper Making a Glance

Documents

Transcript of Paper Making a Glance