Paper and Water

Chapter 7 Paper Drying

in the Manufacturing Process
D. Steven Keller

Paper Drying in the Manufacturing Process 1

Paper and Water

Although water is central to the

papermaking process, it is only
when it is almost completely re-
moved that the paper takes on
the form we find most useful. The
manner in which the water is re-
moved plays an important role in
the nature of the structure that is
ultimately formed (Fig. 7.1). Fur-
thermore, the manner in which
paper was dried also influences
how it behaves when water is
Fig. 7.1 View of a modern papermaking reintroduced to the dry sheet. Most papers remain vulnerable to the
machine used to make newsprint. reintroduction of moisture, whether it is as liquid water or as water
vapour at elevated relative humidity. Increasing the moisture content
weakens the hydrogen bonds, causes fibres to swell and become
more flexible, and can have a profound effect on the characteristics of
the paper.
This chapter considers the removal of water during the papermak-
ing process. It will concentrate on contemporary papermaking in
which dilute aqueous suspensions of cellulosic or lignocellulosic
fibres and additives are dewatered by filtration, pressed and dried to
form a web structure. Understanding contemporary dewatering will
also enhance our appreciation of the equivalent historical processes
that, although significantly different from modern ones, were gov-
erned by comparable mechanisms. Papermaking has always been a
cost-intensive industry in which, increasingly, a narrow margin of rev-
enue can only be achieved if product quality is reconciled with pro-
cessing efficiency. Today, the dewatering mechanisms are tightly con-
trolled to remain economical while achieving properties needed in
the product. The focus of this chapter is on the changes in the fibres
and web structure that take place during drying. These changes are
influenced by many factors, including the fibre species, origin, fibre
conditioning by pulping, bleaching, beating and/or refining, reuse af-
ter recycling, and the drying process conditions. These factors have a
significant influence on the structure and properties of paper, leaving
it in a state with the potential for either subtle or pronounced re-
sponses to moisture, both during natural ageing or conservation
treatments. The discussion of manufacturing drying processes is par-
ticularly relevant for understanding the response of paper to conser-
vation treatments involving the removal of water. Thus, the goal of this
chapter is to provide the reader with an appreciation of the changes
experienced by the fibrous structure of paper as it is dewatered and

2 Paper Drying in the Manufacturing Process

 Paper and Water

dried, and the origins of dimensional, structural and chemical prop-

erties of paper encountered during conservation.

7.1 Water removal Water serves several essential functions during the papermaking pro-
in paper manufacturing cess. When imbibed into the fibres, it provides the flexibility needed
to maximize the contact area between fibres. It is the medium
through which surface tension forces and hydrogen bonding between
the cellulose surfaces takes place and is important as a suspending
fluid for the fibres. Copious amounts of water are necessary to dis-
perse and separate fibres, thereby ensuring the uniformity of their
distribution within the plane of the sheet, a characteristic known as
formation. Kerekes and Schell (1992) introduced the crowding factor
as a useful parameter that relates the amount of water needed to
suspend fibres with the formation of the paper sheet that is formed.
To determine the crowding factor, one need only know the fibre
length, the fibre coarseness defined as the mass per unit length, and
the mass concentration in the pulp suspension. The crowding factor
can be calculated according to the formula:
2
5Cl f (7.1)
NC =
ωf
where:
NC = crowding factor (dimensionless)
C = fibre mass concentration in the pulp suspension (mass %)
l f = fibre length (mm)
ω f = fibre coarseness (mg/m).

The calculation of the crowding factor allows the determination of the

extent of fibre-fibre interaction leading to entanglement and floccu-
lation. If the crowding factor, NC, exceeds 60, then the fibres will
tend to flocculate in the pulp suspension and the uniformity of the
formed paper, or formation, will be poor. For NC less than 60 the
space separating fibres in the pulp suspension is sufficient to reduce
the extent of fibre entanglement and flocculation. This results in a
uniform formation for the formed sheet. In the following example,
typical values for a Douglas-fir softwood pulp suspension at a con-
centration of 0.5% are used to calculate the crowding factor which
falls close to the threshold for flocculation to occur:
2
5(0.4%)(2.7mm)
NC = = 58 (7.2)
(0.25mg/m)
Most papers will therefore begin their passage through the paper ma-
chine as a dilute suspension of fibres in water at a solid content of

Paper Drying in the Manufacturing Process 3

Paper and Water

about 0.5%. For each kilogram of fibre, nearly 200 kilograms of water
must be removed.
The purpose of the paper machine is to distribute fibres continu-
ously and uniformly within a planar region. All of the water must then
be removed and the web consolidated so that adequate bonding oc-
curs between the fibres. The result is a thin, dry, fibrous, layered mat.
While contemporary papermaking processes reflect centuries of in-
cremental enhancement in the speed and efficiency in achieving
these tasks, the function of each process has remained essentially un-
changed. Paper machines of the past 150 years, including those in use
today, are divided into two primary areas. The wet end is where as
much as 99.5% of the suspension water is removed. In the dry end,
the wet web is about one-half water, and is dried to a solid content
exceeding 90% (Fig. 7.2). In traditional hand papermaking as it was
practised since the beginning of papermaking in Europe, the basic
procedures equivalent to the wet and dry ends of a paper machine
require four separate operations:

• By dipping the mould, the vatman gathers a fibre mat from which
is filtered excess water.
• The coucher transfers each freshly formed fibre mat to a wool felt,
piling up many of them to accumulate a post.
• The post is pressed so that water is expressed from the fibre mats
into the felts, squeezing it out of the post. This causes consolida-
tion of the sheets so that they can be handled without support in
subsequent operations. The sheets are separated from their felts
and piled on top of each other for a second pressing that removes
more water and further densifies the sheet.
• The dry end of a contemporary paper machine has substituted loft
drying, which involved hanging small gatherings of sheets (spur)
over ropes stretched in tiers across the room. Hanging multiple
moist sheets for drying while they were still in close contact with
each other improved their planarity. The rate of drying could only
be adjusted through shutters that moderated the environmental
conditions of the room.

Important surface and strength properties are imparted to paper dur-

ing drying. We will limit this discussion to how the basic components
of the wet end and dry end contribute to the inherent properties of
paper (see Fig. 7.2). The dewatering, interfibre bonding and fibre
shrinkage are modelled in Video 7.1.

4 Paper Drying in the Manufacturing Process

 Paper and Water

Fig. 7.2 Essential parts of a modern

Fourdrinier papermaking machine. The
dilute fibre suspension or stock is deliv-
ered from the head box on to the moving
forming fabric (wire) in the wet end, where
forming press dryer soft winder
forming elements and suction boxes re- section section section calender

move most of the free water. The wet web filtration and
vacuum
mechanical
pressing
wicking and
evaporation evaporation

is then consolidated in the press section solid 0.5% 15–20% 45–50% 92–97%

and dried in the dryer section. The

finished paper is collected on the wind-
up reel.

Water 99.5% 89–85% 50–55% 3–8%

7.1.1 Wet end The wet end includes the headbox, the forming section and the press
section. The headbox is designed to convert pipe flow into slit flow in
order to distribute fibres uniformly across the width of the machine.
This continuously formed fibrous mat that spans the width of the pa-
per machine is known as the web. Modern paper machines are usual-
ly more than 4m wide and may even exceed 10m. The dilute stock,
consisting of the fibre suspension, fillers and chemical additives, is
discharged from the headbox on to the moving forming fabric, or
‘wire’. The suspension is transported on this continuous mesh
through the initial stages of dewatering in the forming section. The
design of the headbox, the composition of the dilute stock and the
process operating conditions all contribute to the flow dynamics that
occur in the forming section as the stock passes through the headbox
and impinges on to the moving wire. The result can be a highly uni-
a b
form sheet, or one that is considered of inferior quality containing
streaks or a floccy structure (Fig. 7.3). For instance, consider a floc
region that has a greater mass of fibres and is thicker than the sur-
rounding regions. As the paper is compressed in the press section
and calender, the floc region will be densified to a greater extent. The
c d reduced internal porosity will affect the rate of water absorption and
Fig. 7.3 Examples of four machine-made structural swelling when water is reintroduced by printing with water-
paper sheets showing different forma- based inks, humidification, or rewetting methods used in paper con-
tion characteristics. a: well formed, ran- servation. This may cause an increase in print mottle or cockling for
dom distribution of fibres in the struc-
ture. b: flocculated paper with poor for-
papers that have poor formation. While we have used the term ‘for-
mation. c: paper showing streak defects in mation’ to describe the uniformity of the fibre distribution in the
commercially made paper. d: highly ori- sheet, the word is also commonly used to describe the transition of
ented fibres and flocs as generated on a the dilute stock into a cohesive wet web that occurs just after the
commercial paper machine (images are
10 cm on a side and were generated by β-
headbox in the wet end, in the forming section.
radiographic imaging). Dewatering of the suspension of fibres and fillers to create a cohe-
sive web structure occurs in the forming section. Dissolved and minor

Paper Drying in the Manufacturing Process 5

Paper and Water

suspended pulp additives do not contribute significantly to the mass

of the cohesive web structure, although mineral filler content can be
as much as 30% in modern office papers. Filtration is the principal
mechanism for water removal in the initial part of the forming section.
Fig. 7.2 illustrates a Fourdrinier former that uses a single wire. Mod-
ern paper machines may also dewater the stock from two sides as the
stock is sprayed between two converging wires moving at essentially
the same speed. The dilute stock is squeezed between the forming
fabrics under tension and water is expressed from the web. Such for-
mers are referred to as twin wire or gap formers. Whether configured
as a single- or twin-wire forming section, the forming fabric is backed
with forming elements used to improve dewatering rates and main-
tain uniform material distribution. In the first part of the forming sec-
tion, foils or table rolls are used to remove water by a filtration pro-
cess that leaves the fibres stratified within the fibrous web. The dilute
stock exits the headbox at high velocity through a narrow slit that
spans the full width of the machine, known as the slice. This jet of di-
lute stock strikes the forming fabric(s) at a low incident angle. The jet
may travel slightly slower or faster relative to the velocity of the mov-
ing forming fabric. This causes an inherent orientation of fibre aligned
to a greater or lesser extent along the direction of travel of the moving
forming fabric, known as the machine direction (MD). The direction
perpendicular to the MD, which is also parallel to the slice and the
axes of the various rolls on the machine, is referred to as the cross di-
rection or CD. The term in-plane is often used to describe the plane
formed by the MD and CD directions. The zed direction or ZD refers
to the direction that is perpendicular to the MD-CD plane. This may
also be called out-of-plane and refers to the structure or properties
through the thickness of the paper. These are the common terms that
describe the orientation of machine-made paper (Fig. 7.4). The
added tensile strength introduced by orienting the fibres preferen-
tially in the MD can be useful for runnability of lightweight papers,
such as newsprint, that are printed using continuous web-fed presses.
However, excessive orientation may be a disadvantage for high-quali-
ty papers, such as office papers printed by sheet-fed processes, since
differences in expansion and contraction (dimensional stability) be-
tween MD and CD can cause jams, misfeeds and misregistration of
imaging. To reduce machine direction orientation, paper machines
are sometimes agitated to introduce a lateral shake of the forming
fabric. The forming section is oscillated in the cross direction, adding
to the turbulence of the pulp suspension on the wire. The significance
of fibre orientation will be discussed in the context of anisotropic (not
the same in the MD, CD and ZD) shrinkage of the web.

6 Paper Drying in the Manufacturing Process

 Paper and Water

ZD MD
CD

Fig. 7.4 An image of paper on a paper

machine illustrating the three perpen-
dicular directions. Machine direction
(MD) is aligned parallel with the move-
ment of the web through the paper ma-
chine. Cross machine direction (CD) is
aligned parallel to the wind-up reel. The
zed direction (ZD) is parallel to the thick-
ness direction of the web. It is perpendic-
ular to the principal plane of the web
formed by the MD and CD vectors.
Considering the forming of hand-made papers, the dipping action
that gathers the fibre from the vat involves a continuous motion in
one direction that can cause fibres to remain oriented with slight
preference parallel with the short side of the mould. The vatman
seeks to overcome this bias by shaking the mould alternating be-
tween two perpendicular directions as the water is filtered from the
aqueous suspension. Therefore, sheets made by hand usually have
little or no perceivable orientation of fibres unless it was intentionally
introduced into the sheet by the vatman during drainage. A sheet
that has no apparent directionality, either by orientation of fibres or
by non-uniform straining or restraint during processing, is said to be
isotropic, i.e. having the same properties in any in-plane direction.
Papermakers refer to such an ideal sheet as being ‘square’.
The wet web initially has very low strength that results from fibre
entanglement and friction between contacting fibres. In the second
part of the forming section, vacuum is applied to the back of the
forming wire using wet suction boxes followed by vacuum boxes, and
finally by the couch roll. The suction boxes cause the fibres to be
drawn towards the wire and compress the wet web structure. When
air fully infiltrates the web a situation known as breakthrough occurs,
where the rate of dewatering slows and the wet suction boxes no
longer effectively drain the water from the wet web. It then becomes
necessary to apply a mechanically induced vacuum to draw more vol-
ume of the air-water mixture from the fibrous mat. This occurs in the
vacuum boxes and the couch roll. A dandy roll that rotates in contact
with the top surface of the wet web may also be used to imprint the
web with watermarks or ‘laid and chain’ patterns. The movement of

Paper Drying in the Manufacturing Process 7

Paper and Water

the fibres under the dandy roll may also improve formation and water
removal, especially on older paper machines. After breakthrough, the
presence of air-water interfaces, especially at fibre-fibre crossings
where residual water forms menisci, contributes significantly to the
web strength. These forces that result from the surface tension of wa-
ter and the capillary forces within the structure are referred to as
Campbell forces (Campbell 1959) (Fig. 7.5). Interfibre attraction in-
creases as free water is removed and as the air-fibre-water interface is
extended. These consolidating forces are greater for pulps with higher
a b c
specific surface area, such as those that are more highly refined
Fig. 7.5 Surface tension and capillary (beaten) or that have a greater fraction of fine particles. At the end of
forces acting on the fibre suspension the forming section, the wet web remains relatively thick and uncom-
during water removal. Figures (a)–(c)
pressed. The apparent density (kg/m3) of the structural network at
show the increase of fibre-fibre attraction
as water drains from the web. Air water this point is relatively low because the interfibre contacts result only
surfaces are created after breakthrough in from the small compressive forces induced by Campbell forces and
(b), and then reduced with the evapora- vacuum applied to the bottom side of the forming wire. Fibres do not
tion of water in (c).
fully conform to one another and the potential interfibre contact area
is not fully developed. The wet web exits the forming section at about
18% solids having nearly 92.5% of the water in the original dilute stock
removed.
In the press section, the wet web is mechanically compressed to
further reduce the water content. More significantly, the fibre net-
work is densified and the fibres conform to one another in close con-
tact. This increases the contact area between fibres, where interfibre
bonds can form as more water is removed. As the solid content of the
web approaches 25%, the surfaces are brought close enough together
for hydrogen bonds to begin to form between the fibre surfaces. Al-
though some hydrogen bonding is expected to occur, the strength
of the web is still dominated by surface tension or capillary forces
(Campbell forces) that are inversely proportional to the thickness of
the water film remaining between fibres. Press sections are designed
to affect the greatest amount of mechanical dewatering of the wet
web without shearing the network structure or disrupting the fibre
cell walls. If the latter does occur, the strength potential of the paper
would be seriously compromised. The web exits the press section
and the wet end at about 50% solids so that about 99.5% of the water
has been removed at this point in the process.

7.1.2 Dry end The dry end consists of the dryer section, size press, machine calen-
der and winder (see Fig. 7.2). The main purpose is to reduce the
moisture content of the web to a level where the paper is at equilibri-
um with ambient humidity conditions, typically 3–7wt%. Over-drying
to lower levels would cause the paper to become brittle and lose the

8 Paper Drying in the Manufacturing Process

 Paper and Water

dryer cylinders flexibility needed for bending and folding. The paper would also be-
(upper tier)
pocket
come susceptible to severe dimensional deformations such as cock-
ventilator ling and curl when water reabsorbs into the paper as it equilibrated to
typical humidity conditions. Dry end operations are important for
optimizing the mechanical strength and for conditioning the surface
pocket
properties of paper. The dry end may include sections for surface
treatment such as a size press or an on-machine pigmented coating
web web
dryer
station. The influences these have on the behaviour and properties of
fabric paper are beyond the scope of this chapter.
The dryer section of a conventional paper machine used in the
dryer cylinder
(lower tier) manufacture of printing and writing papers consists of a series of
Fig. 7.6 Diagram of a drying section large-diameter, steam-filled cylinders. The web enters this section
detail showing three cylinders that illus- with a solid content approaching 50%. It is conveyed through the
trate the pathway of the paper web. dryer section in a serpentine pattern, wrapping around the massive
rotating cylinders. In a modern paper machine it is common to have
30–60 dryer cylinders with diameters of 1.5–1.8m. Transport of the
web is facilitated by a permeable dryer fabric, which supports and re-
strains the web and provides a backing force to hold the paper tightly
against the cylinder surfaces. This permits heat energy to be efficient-
ly conducted to the web. Water vapour escapes the web to pass
through or condense in the dryer fabric. Water may also evaporate
felt
paper as the web is drawn between rolls in what is called the dryer pocket
unrestrained
(Fig. 7.6). The figure illustrates how the web is brought into contact
with the heated surfaces of the dryer cylinders, alternating exposure
from one side to the other. This approach ensures a balanced rate of
a restrained
removal of water from both sides of the paper. This helps prevent a
sidedness that could result in a humidity-sensitive curl in the finished
felt pressure paper. As the web passes between the upper and lower tiers of the
conventional dryer, it travels in an unsupported open draw where
both sides of the paper are fully exposed (Fig. 7.7). This permits in-
creased evaporation of water from both sides of the paper. It also in-
troduces a region where machine tension and web shrinkage can have
felt significant effects on the web structure. The development of the sin-
paper gle-tier configuration in more recent dryer sections eliminates open
b vacuum pressure
draws by fully supporting the web with the dryer fabric and restrain it
Fig. 7.7 Diagram of two different dryer using vacuum transfer rolls (see Fig. 7.6). The benefit this provides,
cylinder configurations: double-tier dryer and the importance of web tension and restraint, will be addressed in
(top) and single-tier dryer (bottom). The
a later section.
black arrows mark the areas where the pa-
per web is restrained by the dryer fabric Various methods are used to accelerate the drying process through
that holds it against the heated cylinder increased air exchange at the surface of the web, such as pocket ven-
surface. The blue arrows indicate the main tilation rolls, blow boxes and vacuum rolls. Control of the rate of
direction of water removal from the paper
contacting the cylinder. heating is essential for machine runnability to prevent excessive ad-
Source: Chance (1994). hesion of the wet web to cylinders in the early stages of drying that

Paper Drying in the Manufacturing Process 9

Paper and Water

would cause web breaks. It is also important to optimize the drying

sequence so that drying at later stages can be maximized when re-
maining water is more tightly bound to the fibres.
When the wet web enters the dryer section, the fibrous structure
is already partially consolidated due to capillary contraction in the
forming section and the mechanical compression applied in the press
section. The web still contains 45–55% moisture as free water in the
voids between the fibres, as imbibed water within the interfibre
bonds and fibre cell walls, and as bound water (or adsorbed water)
closely associated with the hydroxyl-rich fibre surfaces and fine capil-
laries (see page 000). The strength of the web is dominated by Camp-
bell forces (Lyne and Gallay 1954, Rance 1954), interfibre entangle-
ment and interfibre friction, which may involve some hydrogen bond-
ing (Baum 1991). While the fibre lumens are partially collapsed, the
cell wall thickness remains unchanged from its initial swollen state
(Nanko and Ohsawa 1989). As the drying process continues, the
fibrous network undergoes significant transformation in structure and
properties as bonds strengthen and fibres shrink. The combination of
moisture and elevated temperature causes the cellulose to become
more pliable under the backing forces of the dryer fabrics. The web is
densified, causing even more interfibre bonding to occur.
The rate of drying slows as free and imbibed water are removed by
thermal evaporation since more energy is required to liberate the re-
maining water from fine capillaries and that which is strongly bound
to the cellulosic surfaces. The drying process involves three phases
that reflect the heating sequence used to efficiently remove water in
its various forms while optimizing machine runnability. There is a dis-
tinct relationship between the web temperature, moisture content
and the drying rate (Fig. 7.8). In the initial heat-up period, the se-
quence of dryer temperatures is kept low to prevent the wet web
from sticking excessively to the surface of the drying cylinders.
Temperature is incrementally increased in the drying cylinders to
raise the web temperature from 40 ± 5˚C as it enters, to 70 ± 5˚C at
about one-third of the way through the dryer section. This marks the
beginning of the second phase, called the constant rate period. The
web temperature remains constant as free water is evaporated. The
web reaches a critical moisture content of about 20wt% when the
free water is fully depleted. The resistance to heat transfer increases
as the contact surface between web and cylinder decreases. While
the pore structure of the fibres collapses at this moisture content, air
and water vapour replace liquid water between fibres and in the cell
walls so that heat is transferred by vapour transport rather than by
conduction. The rate of transfer depends on how open and tortuous

10 Paper Drying in the Manufacturing Process

 Paper and Water

heat-up-period constant rate period falling rate period

temperature

drying rate moisture content

Fig. 7.8 Relationship between the web

temperature, moisture content and the
drying rate of paper.
Source: Kiiskinen et al. (2002).
time

the remaining pore network is within the paper structure. This brings
on the falling rate period of drying. The temperature of the web in-
creases as the imbibed and adsorbed water is evaporated from the
web. Rapid shrinkage of the fibres occurs as the point of critical mois-
ture content is exceeded during drying. Fibre shrinkage causes di-
mensional changes in the web structure and the introduction of in-
ternal stresses that remain in the paper even after it is cooled to
room temperature. The response of the wet web to the drying pro-
cess is significantly dependent on the nature of fibrous structure and
the individual fibres.
As the web is conveyed through the drying process it is subjected
to a variety of internal and externally imposed forces that strain and
restrain the fibrous network as it dries. Tension is applied to the web
in the machine direction to pull it away from the dryer cylinders, pre-
vent fluttering and to reduce the occurrence of web breaks. This ten-
sion strains or elongates the web in the MD, permanently deforming
the network structure and, of course, the fibres contained within.
Elaboration on the effect this has on the network and its properties
will be provided in a later section. The dryer fabric and adhesion to
the dryer cylinders restrains the web, thereby inhibiting shrinkage.
However, this is usually not uniform across the width of the paper
machine. Thus, it is possible to have paper that has anisotropy of me-
chanical behaviour that is due solely to the stretching and restraining
differences in the dryer section, and not fibre orientation as was
pointed out earlier. The shrinkage potential of fibres on drying and
the resistance to that shrinkage imposed by other fibres in the net-
work, by drying restraint and by machine tension, can cause stress to
be frozen into the structure when the web is cooled and dried to am-

Paper Drying in the Manufacturing Process 11

Paper and Water

bient conditions. These residual stresses that remain in the finished

product are readily released with the reintroduction of water or water
vapour (humidity). Release of internal stress, the breaking of intra-
and interfibre bonds, and the swelling of fibres may result in out-of-
plane deformation such as cockle or curl, surface roughening, or di-
mensional changes. Before proceeding to a detailed discussion of
how the web structure responded to drying and subsequent rewet-
ting, it is useful to first examine the response of individual fibres to
drying.

7.2 Drying of individual In their natural state, wood fibres are fully hydrated with water satu-
lignocellulosic fibres rating the cell wall. By their chemistry and structure, wood fibres are
hygroscopic in that they are easily wet by and absorb water. The
porous structure of the fibre is made more porous when lignin and
hemicelluloses are removed from the cell wall during the chemical
pulping processes. Water remains absorbed between fibres, in the
fibre lumen (centre), within the pores of the cell wall and on the fibre
surface. As described in earlier chapters, the conditioning of fibres
during the pulping and beating or refining processes leaves the fibres
in a swollen state with the cell wall delaminated and fibrillated. The
extent of swelling depends on the fibre species and origin, and on
fibre conditioning.
Beating and refining result in an increase of the cell wall surface
area, water absorption and swelling. The shrinkage that fibres exhibit
when they are dried directly result from changes in the internal struc-
ture of the cell wall. Stone and Scallan (1966) studied the effects of
drying on the porous structure of the fibres using nitrogen adsorption
surface area measurements. They found that the lamellae in the wa-
ter-swollen secondary wall S2 layer (see page 000) are drawn together
in progressively thicker aggregations that result in a decrease in the
total pore volume (Fig. 7.9). The pore volume that remains once the
perspective view

Fig. 7.9 Lamellae (consisting of macro-

a fully swollen b) partially swollen c) unswollen
fibrils) drawing together in thicker ag-
gregations as a fibre dries. In the upper
row, a perspective view shows lamellae
from wet (a) to dry (c). The bottom row
shows lamellae in the same process in
cross-section (axial view).
Source: Stone et al. (1966). axial view

12 Paper Drying in the Manufacturing Process

 Paper and Water

fibre is dried is negligible compared to the original swollen state. The

microfibrils that comprise the S2 layer are principally oriented along
the fibre axis, mostly less than 30˚ off axis, referred to as the microfi-
bril angle. This explains the response of the fibre as water is removed
from the cell wall, where shrinkage across the width, called transverse
or lateral shrinkage, is much greater than shrinkage along the length,
referred to as axial or longitudinal shrinkage. This model also ad-
dresses the dependence on fibre conditioning in which increased de-
lamination of the cell wall causes greater initial cell wall thickness and
thus greater shrinkage as the S2 layer collapses with the removal of
water.
Stone and Scallan (1966) also found that as much as 80% of the
water contained in the cell wall is molecularly associated with the cel-
lulose. This occurs in spaces within the microfibrils that compose the
lamellae, the so-called microreticular pores. Molecular mechanisms
to explain the changes in mechanical and viscoelastic properties en-
countered as single fibres are dried under tension have been suggest-
ed by a number of investigators. Hudson (1963) proposed that the
changes occurred as a result of conformational changes of the cellu-
lose in the cell wall for more favourable molecular alignment and the
elimination of microcompressions. Microcompressions appear as
creping or wrinkling of the fibres when they undergo unrestrained
shrinkage along the principal axis, or when fibres are subjected to ex-
fibril ternal compression, for example at fibre crossings in a network struc-
ture (Fig. 7.10). The shrinkage potential of a fibre is defined as the
amount it would shrink if dried unrestrained with no applied external
dislocation
forces.
Jentzen (1964) and others (Spiegelberg 1966, Hill 1967) observed
that fibres dried under external axial load showed increased tensile
strength and elastic modulus and reduced strain-to-failure or elonga-
tion. The tensile strength is the maximum load per unit area where
the fibre breaks. The elastic modulus is the resistance to elastic
stretching. Kallmes and Perez (1966) found that the increase in tensile
strength and decrease in strain-to-failure were proportional up to
twice the values for fibres dried unrestrained. Thus the drying of
fibres under tension will have considerable significance for the final
a b strength properties of the network they form. Jentzen attributed the
Fig. 7.10 Structural configuration of a increase in fibre stiffening and tensile strength to the increased orien-
fibre interior free air-dried (left) and tation of the crystallite regions of the cellulose, and more uniform
dried under tension (right). Drying distribution of stress among the fibrils. The effect was greater for
under tension results in fibre stiffening
fibres with low-yield chemical pulps, for pulps with a higher content
due to the increased alignment of fibrils
within the fibre cell wall. of hemicelluloses or for those that were more heavily beaten or
Source: Retulainen et al. (1998). refined. Either of these conditions enables the cell wall elements to

Paper Drying in the Manufacturing Process 13

Paper and Water

rearrange more easily under the tensile load (see Fig. 7.9). Kim et al.
(1975) divided the effects of drying on single fibres into two mecha-
nisms. For fibres with low microfibril angle (<10˚), kinks and other dis-
tinct dislocations generated in pulping and refining are removed by
drying tension and the fibre strength is restored. For fibres with high
fibril angles, strength is also imparted by a reduction of the fibril an-
gle. None of these phenomena can occur without the presence of
water. A more comprehensive explanation of prevailing mechanisms
forwarded by Salmén et al. (1987) emphasizes the importance of the
plasticizing action of water in softening the hemicelluloses surround-
ing the crystalline and amorphous regions of cellulose. Under the
right conditions of temperature and humidity, the hemicelluloses are
softened and deform so that external forces are able to align crystal-
lites and orient the disordered zones (Fig. 7.11). When the fibres are
dried and cooled, the result is the increased stiffness and load-bear-
ing capacity (tensile strength) of the fibres.
When fibres are dried, transverse, axial and thickness shrinkage
and collapse of the lumen change the fibre dimensions (Fig. 7.12).

Fig. 7.11 Fibril section showing the con-

figuration of the crystalline and amor-
phous regions of cellulose (beige lines)
and hemicellulose (green lines) during
drying. The fibre in wet condition is fully a b c
relaxed (a). Axial restraint tensions the
fibril during initial drying stages (b). Dry-
ing causes hemicellulose surrounding the
fibril to be aligned with the direction of
fibril restraint (c). wet moist dry (~50%RH)
Source: Salmén et al. (1987, Waterhouse
2002).

transverse
or lateral

axial or
longitudinal

bonded
area

free fiber
segment
Fig. 7.12 SEM photograph of a paper
surface indicating areas where the fibre
experiences transvere, axial and longitu- axial
dinal shrinkage during drying. At ×400, wrinkles
the bottom edges represents 250µm.

14 Paper Drying in the Manufacturing Process

 Paper and Water

The transverse shrinkage of individual, unrestrained fibres was first

measured by Page and Tydeman using soft X-ray micro-radiography¹
(Page and Tydeman 1963, Tydeman et al. 1966). For chemically
pulped softwood fibres, they reported the average shrinkage in width
to be 18% for unbeaten and 29% for beaten fibres. The average
thickness shrinkage was 60% for unbeaten and 75% for beaten fibres.
These values are significantly larger than those for the axial (length-
wise) shrinkage of individual unrestrained fibres that is typically of the
order of 1–2% (Nanko and Wu 1995).
More recently, Nanko et al. studied the shrinkage of individual
kraft chemically pulped fibres and thermomechanically pulped
(TMP) fibres in web structures as a function of moisture content us-
ing confocal scanning laser microscopy² (Nanko and Ohsawa 1989,
Nanko et al. 1991). They studied the transverse shrinkage in unre-
strained, free fibre segments and also in segments restrained by
bonding at fibre crossings (Fig. 7.13, Videos 7.2 and 7.3). Nanko and
Ohsawa (1989) suggested that the fibre morphology (form and struc-
ture) passes through five stages from the saturated state through dry-
ness (Fig. 7.14). The stages are as follows:
• For solid contents up to 50–55%, the morphology of the fibres
remains relatively unchanged as free water and water at the
surface (terms; see previous comment) evaporate.
• As solid content increases from above 50 % to above 60 %, water
evaporates from the lumen and increased capillary forces cause

1 Soft X-ray microradiography is a non- which has been treated with a fluorescent
standard method that measures the atte- dye. The light from the focal spot passes
nuation of low-energy (0-5 keV) X-rays as back through the lens and the beam split-
they pass through the specimen. The mass ter to a photodetector. A confocal aper-
of the material in the path of the beam ture (pinhole) is positioned in front of the
causes a reduction of transmitted X-rays photodetector to filter out all of the light
to reach the detector, which for Page and emitted from the focal planes above or be-
Tydeman was X-ray film. The procedure is low the focal plane at a certain depth wi-
similar to transmission electron microsco- thin the specimen. A two-dimensional (X-
py, although instead of a lower energy Y in-plane) image of a narrow focal range
electron beam, the higher energy of the X- (Z-thickness), called an optical section, is
rays permits thicker or denser samples to obtained by raster scanning the sample
be imaged. underneath the sensor. The microscope
sensor is then moved slightly closer or fur-
2 Confocal laser scanning microscopy ther away from the sample and a new opti-
(CLSM) is a method used to map the sur- cal section is acquired. For semi-transpa-
face of opaque samples and the subsur- rent materials such as paper, optical sec-
face, three-dimensional structure of trans- tions can be obtained from several fibre
parent and semi-transparent materials, thicknesses beneath the top surface. A
such as paper. In the CLSM, a laser light three-dimensional reconstruction of the
beam is focused through a beam splitter fibrous structure can be obtained by
and objective lens on to the specimen, stacking two-dimensional optical sections.

Paper Drying in the Manufacturing Process 15

Paper and Water

the fibre to collapse. Compression at the fibre crossings begins to

occur.
• For solid contents above 60% to above 70%, water begins to
evaporate from the cell wall and axial or longitudinal wrinkles that
are aligned along the principal axis of the fibre begin to form.
• At solid contents from above 70% to above 80%, transverse shrink-
age of the fibres in unbonded segments begins to occur and fibres
continue to flatten. Significant web shrinkage coincides with the
onset of transverse fibre shrinkage at the fibre crossing regions.
• For solid contents greater than about 80%, substantial transverse
shrinkage of fibre begins to occur at fibre crossing regions. Shrink-
age of the fibres in the unbonded regions continues, and axial
wrinkles, aligned along the principal axis of the fibre, become more
distinctive. This region includes the fibre collapse point (FCP).
• For solid contents greater than about 90%, only a small amount of
dimensional change is seen to occur up to the final solid content of
the product paper.

Nanko found fibre shrinkage to be minimal up to a specific solid con-

tent, called the fibre collapse point ( FCP), which for the pulps in his
study was close to 86 wt% solids in stage five above. Fibre shrinkage
increased dramatically when fibres were dried further. Other investi-
gators observed a similar increase in fibre shrinkage around this solid
content for different fibres (Rance 1954, Tydeman et al. 1966).The
point at which rapid increase in shrinkage begins was attributed to the
depletion of free or imbibed (capillary) water up to the FCP. At that
critical point, the water that is strongly associated with the cellulose
(Zeronian 1985) either by being adsorbed on the fibre surface, or
which fills the fine porous structure within the cell wall, begins to
evaporate. This causes collapse of the pores, significant contraction of
the cell wall and transverse shrinkage of the fibre. The extent of
shrinkage at the FCP has significant influence on shrinkage of the
fibrous network observed when the wet web dries to this solid con-
tent in the dryer section. This will be discussed in more detail in the
next section.
The total water contained within the cell wall, indicative of the
porous structure accessible to the water, determines how much the
fibres will shrink during drying when the pore structure collapses. It is
also related to the ability of the fibres to reabsorb water once they
have been dried beyond the FCP. An indirect method of quantifying
the amount of water contained within the cell wall, known as the wa-
ter retention value (WRV), was introduced by Jayme (1944). A wet
pad of pulp is centrifuged and the moisture content of the sediment

16 Paper Drying in the Manufacturing Process

 Paper and Water

1.6 pulp is determined. Since the amount of retained water was found to
fibre saturation point (g/g)

be proportional to the extent of beating and other properties associ-

1.2
ated with the swelling of the cell wall, the WRV has been considered
an indicator of the imbibed and bound water associated with the cell
unbleached kraft wall structure. Jayme applied this method to examine the irreversible
0.8 stiffening of fibres, called hornification, which will be discussed in de-
thermomechanical
tail in a later section.
Stone et al. (1968) used the solute exclusion method³ for measur-
0.4
0 1 2 3 4 5 ing the internal pore structure of the water-swollen fibre walls, there-
number drying and rewetting cycles by determining the amount of water contained within the cell wall,
termed the fibre saturation point ( FSP). For example, a high FSP
Fig. 7.15 Change of water sorption ca-
pacity (fibre saturation point or FSP) of
value indicates that the cell wall is open and porous and contains a
different types of pulps after repeated substantial amount of water. Multiple drying and rewetting of me-
drying cycles. The unbleached kraft pulp chanical and chemical pulps alters the FSP (Fig. 7.15). For the ther-
fibre responds to repeated wetting and momechanical pulp fibres, the lignin and hemicelluloses that are
drying with a significantly reduced F SP
due to fibre cell wall collapse. The ther-
present within the cell wall limit swelling and the subsequent collapse
momechanical pulp fibre is prevented of the layered structure during the drying. Therefore, little change in
from collapsing by the presence of lignin. the FSP is observed. The FSP for never-dried pulp increases with
Source: Laivins and Scallan (1993). decreasing yield as lignin and hemicelluloses are extracted from the
cell wall, leaving an open porous structure (Fig. 7.16). Upon drying,
the pore structure of the cell wall collapses. With rewetting, the pore
1.6 space within the fibre is never fully restored to its original state. This
fibre saturation point (g/g)

is evident from the lower FSP values. While it is of little consequence

in conservation treatment that beating/refining can reopen the pore
1.2
structure through mechanical action and restore the FSP of recycled
chemical pulps to a state close to that of never-dried virgin fibres,
two important conclusions may be drawn. First, shrinkage is much
0.8
more pronounced for chemical pulp fibres. This should result in
more pronounced internal stress in the formed sheets as fibres move
0.4
Never-dried and change dimension. Thus, when they are rewet in use or during
Dried and rewet
conservation treatment, the dimensional changes in the fibres and
100 80 60 40 fibrous network will occur as the fibres swell and internal stresses are
pulp yield (%)

Fig. 7.16 The water sorption capacity

(fibre saturation point or FSP) of never- 3 The solute exclusion technique seeks to the volume of water displaced from the
dried pulps of different composition. determine the size and volume of the pore pores contributes to the dilution of the so-
Pulps of high lignin content or high yield structure of a water-swollen material such lution. The change in probe concentration
absorb less water than pulps freed of their as the fibre cell wall. The method quanti- reflects of the volume of the pores accessi-
lignin content. The effect of lignin removal fies the amount of penetration of a series ble to that size probe for pores of the pro-
on the FSP is especially prominent after of probe molecules of a range of molecular be size or larger. Incrementally larger pro-
pulps were dried and then are rewetted, sizes (diameters). The probes are inert bes are tested so that a plot of the distri-
as pulps of low lignin underwent perma- with regard to the pore surfaces of the bution of pore sizes can be generated. The
nent loss of pore volume during drying specimen. The fibres are immersed in an fibre saturation point ( F S P) was formally
which causes them to absorb less water aqueous solution of known concentration defined by Stone et al. (1968) as the volu-
when rewet (see also Fig. 5.14). of the probe. For small-diameter probes me attributed to pores with a diameter of
Source: Laivins and Scallan (1993). that penetrate the water-swollen structure, 56 nm or less.

Paper Drying in the Manufacturing Process 17

Paper and Water

released. Mechanical pulps will be more resistant to shrinkage as the

lignin-hemicellulose gel within the cell wall limits internal hydrogen
bonding. These fibres will also be subject to swelling and the release
of stress when rewet, although to a lesser extent as compared to
chemical pulps. Secondly, rewetting of chemical fibres in conserva-
tion treatments will never fully recover the internal pore structure
that they once had when first used to form the paper. Such fibres are
said to undergo ‘hornification’ during the drying process, a term that
will be discussed in more detail below. The dimensional changes of
the fibres can result in significant changes to the fibrous structure, a
subject that will now be considered.

7.3 Drying The characteristics of a fibrous network as it dries are only partially
of the fibrous network determined by the nature of the fibres that comprise it. There are a
number of other factors that significantly influence the rate of drying
and the properties of the dried structure that forms once the drying
process is completed. Since the fibres are distributed randomly
(stochastically) in the web, all papers have some in-plane variability of
mass, viewed as flocs or, in some situations, streaks. The rates of dry-
ing may be quite different for floccy and lightweight regions. Stresses
build as the regions and fibres contained within shrink at different
rates (Corte and Herdman 1975). This results in competing strains
and the build-up of internal stresses over a range of dimensions. With
no restraining, the web will likely undergo out-of-plane deformation
appearing as cockle or curl. The anisotropic drying behaviour of indi-
vidual fibres can cause directionality in the web shrinkage if the fibres
are preferentially oriented in the machine direction during formation,
either in the paper machine wet end or by the hands of the vatman.
We will see that this directionality (anisotropy) can influence the
shrinkage patterns of the fibrous network that result in out-of-plane
deformations such as curl or waviness. While formation and orienta-
tion are determined by the manner in which the fibres are distributed
in-plane when the paper is formed, other factors such as the exter-
nally applied stresses of web tension and drying restraint will have a
key role in the development of the properties of the network. These
will be discussed in the sections that follow. It is first important to un-
derstand how individual fibre shrinkage is transferred to the shrinkage
of the web structure.

7.4 Network shrinkage Page and Tydeman (1962) advanced a model describing the mecha-
from fiber shrinkage nisms for shrinkage of the fibrous network based on the shrinkage be-
haviour of individual fibres and the transferral of fibre contraction
through the fibre-fibre bonds. Their key observation was that interfi-

18 Paper Drying in the Manufacturing Process

 Paper and Water

bre bonds form before significant shrinkage of the fibres occurs. This
is the so-called ‘adhesion before shrinkage’ concept. Thus the posi-
tions of fibre crossings along the swollen fibre axes are preserved by
hydrogen bonding in the contacting areas where they cross (Fig. 7.17).

MD
Fig. 7.17 Adhesion before
shrinkage concept accord-
ing to Page and Tydeman.
The interfibre hydrogen
CD bonding develops before
shrinkage of the fibres
occurs during drying of the
web. The position of the
fibre crossings established
thickness
in wet web (a) are preserved
in the contacting areas
during drying while the
sheet shrinks to its final di-
mensions (b).
Source: Page and Tydeman
(1962).

As described in a previous section, the transverse shrinkage of a fibre

is considerably larger than its axial shrinkage. This transverse shrink-
age is passed on to the crossed fibre through contraction of the
bonded area (Fig. 7.18). The compressive forces exerted on the
crossed fibre are opposed by the axial stiffness of that fibre so that
shear stresses arise in the bond area that result in fibre deformation.
Since the compressive forces are acting on only one side of the fibre,
Fig. 7.18 The cross-sectional (transverse) bending strain is expected if the fibre is not sufficiently restrained by
shrinkage of the fibre is much greater
than its axial (longitudinal) shrinkage. the fibrous network (Rance 1954). When the crossed fibre has many
This causes microcompressions in axial bonding sites on the top and bottom sides and only a small amount
direction on crossing fibres. of free fibre length, then the restraint imposed by the network will
Source: Baum (1991).
cause axial compression in the bonded area. Microcompressions or
fine scale buckling of the cell wall may occur in the crossed fibre. This
causes a measurable shrinkage along the fibre axis (axial shrinkage).
The shear stresses may also delaminate the cell wall structure and
pucker the lamellae in addition to axial compression (see Fig. 7.18).
Considering all of the crossings that occur along a fibre located in a
fibrous structure, the potential for axial shrinkage is the summed
shrinkage at all of the crossings. Most printing papers have sufficient
grammages (basis weights) and density so that most fibres are bond-
ed on the top/and or bottom sides over nearly all of their length (see
Fig. 7.12). Therefore, axial shrinkage of a fully bonded fibre can theo-
retically approach the transverse shrinkage of about 20%. The pres-
ence of neighbouring fibres in the network adds to the complexity of

Paper Drying in the Manufacturing Process 19

Paper and Water

simple shrinkage analysis since each fibre may experience similar

contractions that oppose or redirect the strains. If no external re-
straining forces are applied to the network, a nominal shrinkage of
about 12% could be expected in drying. As the compression of the
fibre occurs at bond sites, the site first pivots to align parallel to the
principal plane of the web. The contraction then exerts a stress on the
adjacent free fibre segments. This strains (stretches) segments,
thereby removing dislocations, kinks and wrinkles. As discussed in the
previous section, fibres that are dried under restraint or axial load ex-
hibit increased tensile stiffness and strength. This, combined with the
increased stiffness of the densified bonded sites, confers the strength
and stiffness to the entire network structure of paper. Additional re-
straint of the network imposed by external forces will further add to
network strength and stiffness. This subject will be addressed in sec-
tion 7.8.

7.5 Structural factors In work by Nanko et al. (1991), the shrinkage behaviour in bonded and
that control shrinkage unbonded regions was directly measured using confocal laser scan-
ning microscopy. As with single fibres discussed in the previous sec-
tion, the transverse shrinkage increased significantly as the fibre col-
lapse point (FCP) of about 85wt% solid content was approached.
The unbonded regions were observed to shrink first as they are not
subject to the transverse restraining forces of a contacting fibre.
When the bonded regions were observed to shrink, they did so to a
lesser extent due to the axial compression resistance of the crossed
fibre. The theory of Page and Tydeman (1962) identifies the major
structural factors that control shrinkage in the fibrous network as:

• The intrinsic potential shrinkage of the fibres

• The resistance of the fibres to axial compression during shrinkages
• The strength (of the network) during shrinkage and the extent of
fibre-fibre bonding
• Fibrillation, the presence of which has shrinkage forces associated
with it.

The shrinkage potential of fibres (factor 1) depends on the fibre com-

position determined by species and origin and the extent of pulp
conditioning, i.e. chemical pulping and beating or refining that affects
swelling of individual fibres. The shrinkage potential of a wet web in-
creases with decreasing pulp yield as the removal of lignin opens the
porous structure of the cell wall for increased swelling. Pulps with
higher content of hemicelluloses, such as those composed of flax
fibres, also have a greater shrinkage potential since hemicelluloses are

20 Paper Drying in the Manufacturing Process

 Paper and Water

associated with the amorphous (non-crystalline) regions of the mi-

crofibrils. The hydrophilic nature of most hemicelluloses causes the
rapid sorption of water. Beating and/or refining delaminate and fibril-
late the cell wall, which also promotes swelling and increases the
shrinkage potential of fibres. This increased water in the cell wall is
commonly quantified indirectly by measuring the water retention val-
ue (WRV).
The third factor addresses bond strength, which is influenced by
the fibre surface chemistry. The cellulosic surfaces of low-yield
chemical pulps will hydrogen bond more readily than the high-yield
mechanical pulps that have lignin and extractives contaminating the
exposed surfaces. The strength of the bond will be enhanced by
strength promoters such as starch or carboxymethyl cellulose. Addi-
tives such as mineral fillers or rosin size particles tend to interfere
with hydrogen bonding and reduce bond strength. Byrd (1974)
demonstrated that web contraction decreased when debonding
agent was added to chemically pulped handsheets. The strain-to-fail-
ure of the dried sheets also decreased as bonds were weakened. Byrd
found the opposite trends occurred when bonds were fortified with
highly beaten pulp gel.
It is also important to appreciate that the transfer of the transverse
shrinkage from one fibre into the axial shrinkage of another is directly
dependent on the bonded area as quantified in paper science as the
relative bonded area (RBA). Relative bonded area is a dimensionless
expression of the average degree of bonding of the network, defined
as the ratio of the bonded surface of fibres and their total external
surface area. For papers that have low interfibre bonding area or
where the bonding in the region is modified with additives that pro-
mote or interfere with hydrogen bonding, the transfer of shrinkage
from one fibre to another will be affected. This directly affects the
shrinkage of the network as well. The bonded area may be influenced
by fibre stiffness and dimension as well as beating and wet pressing.
Thus it is not merely the shrinkage potential of the fibres that is im-
portant, but also the ability of the bonding of the network to transfer
individual fibre shrinkage to other fibres within the network.
The degree of beating/refining and wet pressing both increase the
shrinkage potential of the web because they both increase the interfi-
bre bonding area. Beating and refining cause an increased flexibility
of the fibres so that they conform to the contours of other fibres
within the network, increasing the RBA. The swelling of fibres also re-
duces the compressive modulus (stiffness) so that the crossed fibres
will deform more easily under axial compression. Pressing provides
increased densification of the network so that more fibres are

Paper Drying in the Manufacturing Process 21

 Paper and Water

brought into contact and the areas of contact are increased. Consid-
ering the importance of the shrinkage associated with fibre crossings,
more fibre bonding will contribute to the shrinkage potential of a
pulp.

7.5.1 Fibre orientation In machine-made papers, fibres preferentially align along the ma-
chine direction (MD) as a result of the forming process in the wet
end. For publication grades such as newsprint, the fibres may have
15 1.5–2.5 times the probability of ordering in the M D as compared to
shrinking (%)

the CD. This creates anisotropy of shrinkage and mechanical strength

CD
properties of the paper. Gates and Kenworthy (1963) studied this as-
10
pect of drying shrinkage for chemical pulps. They found that shrink-
age in the cross machine direction was generally greater than that in
the machine direction (Fig. 7.19). The strain-to-failure, or the amount
MD
the paper stretched to the point of breaking, was also greater in the
5 CD, although tensile strength was observed to be greater in the MD
(Fig. 7.20). This behaviour is explained using the so-called ‘adhesion
before shrinkage’ model of Page and Tydeman (1962) discussed
0 above. One can easily conceptualize that the fibres aligned in the
30 40 50 60 70 80 90 100 cross machine direction will have a statistically greater number of
solids content (%)
fibre crossings along their length than those aligned in the MD. When
Fig. 7.19 Relationship between solids fibre shrinkage occurs, fibres aligned in the CD will experience more
content of pulp and web shrinkage in sites where axial compression can occur. This causes more overall
machine (MD) and cross machine (CD)
shrinkage of the network in the cross machine direction unless exter-
directions. The shrinkage in the cross
direction is generally greater in C D than nal forces, such as drying restraint (Fig. 7.19), interfere. In the dried
M D for machine-made paper. network, the fibres in the CD will have more regions of creping and
Source: Hansson et al. (1989). microcompressions as compared to the MD fibres. When the paper
is strained in the CD the compressed bonded regions will stretch so

3.5
stress (kN/m)

3
MD
2.5 LWC base paper

2
Fig. 7.20 Relationship between strain-to-
failure – that is, the force required to
1.5
break the tensioned paper – and the
shrinkage experienced by the paper dur-
1
ing its drying on the paper machine. In
the MD, the paper has a higher tensile 0.5
strength because more fibres are more
oriented in that direction. 0
Source: Kajanto and Niskanen (1998). 0 0.5 1 1.5 2 2.5 3 3.5
strain (%)

22 Paper Drying in the Manufacturing Process

 Paper and Water

that a greater strain-to-failure of the paper will occur. Gates and Ken-
worthy (1963) found the strain-to-failure (extension at break) is lin-
14 early related to the percent shrinkage (Fig. 7.21). Essentially the
stretch (%)

shrinkage (negative strain) that occurs during drying is recovered un-

12
der tensile loading. Lower tensile strength is observed in the C D
10
since paper derives a significant amount of strength from the tensile
strength of individual fibres and fewer fibres are oriented in the CD
8 to bear the load (Fig. 7.20). The MD tension of the paper machine al-
so imposes directionality on the network, which will influence the
6 mechanical properties. This will be discussed in more detail below.
4 Fig. 7.21 Relationship between stretch and shrinkage of machine-made paper
MD CD
fourdrinier comparing the MD and CD directions. The more the paper web is allowed to
2 vat machine shrink during drying, the more it will be extended before it ruptures when tensioned.
handsheets The extension at break is linearly related to percent shrinkage.
0 Source: Gates and Kenworthy (1963).
0 2 4 6 8 10 12 14
shrinkage (%)

7.5.2 Development As fibres shrink during the drying process, their straining (movement)
of residual stresses exerts stresses (forces/area) on the network through all of the fibres
that they cross and form bonds with. The fibres become increasingly
rigid as their cell walls collapse, hydrogen bonds form and the water
that once imparted flexibility is evaporated. Hemicelluloses or dry
strengthening agents such as starch will solidify and reinforce contact
points between fibres, microfibrils or fillers that may be present. A
significant distribution of internal stress will still remain. Additional
stresses may be induced if the web is mechanically restrained from
shrinkage or if it is stretched in the MD by the tension of the paper
machine. The finished paper will therefore contain ‘frozen-in’ stress-
es (Ivarsson and Steenberg 1947) that are a function of the character-
istics of the fibres contained in the network, the drying conditions
and any external forces that may have been imposed on the network
during drying. In his review of the subject, Waterhouse (2002) de-
scribed a body with residual stresses to be ‘… in a state of stress even
though it has no external forces or constraints acting on its bound-
ary’. Based on this widely accepted perspective, residual stress is dis-
tributed at various structural levels encompassing stress at the
molecular level, within microfibrils and fibres, at fibre-fibre bond
sites, and between flocs in the fibrous network.
When a wet web is dried with no external forces, referred to as un-
restrained drying (also free drying), many of the residual stresses are
relieved by the deformation of the network either by dimensional
shrinkage or as out-of-plane deformation (cockling or curling). The

Paper Drying in the Manufacturing Process 23

 Paper and Water

total shrinkage that occurs in this situation is referred to as the poten-

tial shrinkage (or free shrinkage). This is dependent on the shrinkage
potential of the fibres and the other factors that influence shrinkage
and drying stress discussed in the previous section. Paper made on a
2.5
drying force / sample width (kN/m)

paper machine or by hand typically have imposed external forces that

2.0
bleached kraft oppose the shrinkage, or may even elongate the web during drying.
sheet density 750 kg/m³
The shrinkage that actually occurs during the drying process is re-
1.5 ferred to as the drying shrinkage (also allowed shrinkage). Smith
(1950) first defined the difference between potential shrinkage and
1.0
allowed shrinkage as the dried-in strain. This is a useful parameter for
predicting the tensile strain-to-failure. Smith also found that the
0.5
hygroexpansion of paper is inversely proportional to the dried-in
0 strain. In general increasing dried-in strain, as in the case of a fully
30 40 50 60 70 80 90 100
dry solids content (%)
restrained sheet, will increase dimensional stability and reduce the
strain-to-failure.
Fig. 7.22 Relationship between drying
If the wet web is restrained during drying by holding the edges sta-
force versus solid content during the
drying of the paper web restrained along tionary to maintain the in-plane dimensions, the net stress within the
its edges. The drying force, which is a network will increase as moisture content decreases. This stress,
measure for the stress exerted upon the which acts in opposition to the shrinkage of the web, can be mea-
restrained paper sheet during drying, in-
sured at the sample edges. The final stress measured when drying is
creases as the solids content of the web
increases. complete is defined as the drying stress. An example of the develop-
Source: Htun and de Ruvo (1977). ment of drying stress is shown in Fig. 7.22 for bleached softwood kraft
pulp (Htun and de Ruvo 1977). The drying stress is first detected at a
solid content of about 40%. Drying stress increases slowly until about
15
final drying force (kN/kg)

85% solid, which coincides with the fibre collapse point (FCP). The
drying stress then increases rapidly as more shrinkage of the fibers
within the network occurs. In that same study, Htun and co-workers
10 also found that tensile strength, elastic modulus (tensile stiffness) and
compressive strength were all linearly related to the drying stress of
paper for situations where the paper was fully restrained from shrink-
5 age during drying.
From numerous investigations it can be inferred that the four
structural factors that control shrinkage, as identified by Page and Ty-
0
deman (1962) and discussed above, will directly affect drying stresses.
0 150 200 250 Ivarsson (1954) and others (Brecht and Pothmann 1955, Brecht et al.
WRV (g/100g)
1956, Byrd 1974) studied the drying stresses of papers and found
Fig. 7.23 Relationship between the dry-
ing force versus the degree of swelling
them to be dependent on the level of beating and also the shrinkage
of fibres as measured by their ability to observed during unrestrained drying. Hansson et al. (1989) showed a
hold water (water retention value or direct relationship between the swelling of fibres, as measured by the
WRV ). The WRV is dependent on the water retention value, and the drying stress (Fig. 7.23). Ivarsson
degree of beating. The drying force is a
(1954) showed that drying stress was almost proportional to the
measure for the stress exerted upon the
restrained fibre during drying. grammage (basis weight) of a paper. Byrd (1974) demonstrated that
Source: Hansson et al. (1989). the drying stresses were directly affected by the strengthening or

24 Paper Drying in the Manufacturing Process

 Paper and Water

weakening of the interfibre bonds using chemical additives. Since

drying stress appears to be directly related to the network shrinkage,
one might assume for webs with significant fibre orientation that in-
creased shrinkage in the CD would cause greater drying stress in the
CD than in the MD. On the contrary, the opposite is observed. This
results from the greater number of fibres oriented in the MD, and
the axial stiffness of a fibre is greater than the transverse stiffness
(Hansson et al. 1989).
Another approach used to characterize the in-plane internal
stresses was introduced by Kubát and co-workers (Johanson and Ku-
bát 1964, Johanson et al. 1967), and reviewed by Htun (1986a). The
Kubát internal stress level is determined by generating a series plots
from stress relaxation experiments where the change in stress is mea-
sured as the sample is held at constant strain. The measured stress
level increases with beating, pressing, and by drying under restraint or
tension. The Kubát internal stress is zero for sheets dried unre-
strained. Htun and de Ruvo (1977) further asserted that the Kubát in-
ternal stress is essentially equal to the drying stress for paper sam-
ples, regardless of conditions of refining, wet pressing and sheet
structure. This equivalency is useful for simplifying the relationships
between drying and the mechanical properties of the paper.

7.6 Drying of the web The previous sections have shown that when a web is unrestrained
in papermaking during drying, and allowed to shrink, a relatively extensible paper is
formed with strength that is less than optimal. Historical handmade
papers were often subjected to restraint of some form in order to im-
prove the paper properties, especially planarity, stiffness, strength
and dimensional stability. In contemporary machine papermaking,
the design of web transport and drying strategies are also used to
control and optimize these properties. The paper machine also intro-
duces artefacts such as fibre orientation, web tension and fabric pat-
terning that are integral parts of the process and are difficult to elimi-
nate entirely. In this section, those aspects of the process that apply
external forces to change the web structure during drying will be dis-
cussed. A drying load is an external force that acts in opposition to
the drying stress. It may originate from the adhesion forces between
the wet web and the drying surface that restrain the web from shrink-
age. Drying load may also be imposed by tension forces of the dryer
fabric that compress the web against the dryer cylinder that adds to
the in-plane restraint of the fibrous network. The tension of the web
as it is drawn through the paper machine will introduces a machine
directional (MD) load that may restrain dry or even stretch (strain)
the web in the MD.

Paper Drying in the Manufacturing Process 25

Paper and Water

7.6.1 Wet straining Tension applied to the web is important for maintaining runnability
and draws during the papermaking process. It is especially important as the un-
supported web passes in the spaces between machine components,
known as open draws. The force applied to the web in the machine
direction will stretch or draw the web. This will actually cause subtle
acceleration of the web as it travels from the couch roll to the reel.
Machine direction tension is closely controlled by varying the rota-
tional speeds of the machine rolls and dryer cylinders. In general,
faster machines require greater tensions to maintain stability of the
unsupported web. Wet straining of the web may occur as it leaves the
forming section at the couch roll to enter the press section (as on
older machines) or more commonly between the press section and
the dryer section. Stretching at the latter can be as much as 2.5–3.0%
(Kiiskinen et al. 2002). The solid content of the web is typically less
than 50% when it leaves the press section. In this hydrated state, the
fibre structure is deformable and local separation or slippage of the
internal structure can occur. Straining of the fibrous network tends to
move, straighten and align fibres with the risk of separating contacting
fibres that are in contact. In fact, the thickness of the web increases,
causing the density to decrease. Schultz (1961) and Parsons (1972)
observed that wet straining of more than 3% causes the light scatter-
ing coefficient of dried paper to increase, which is an indication that
the relative bonded area (RBA) has decreased. Since only a small
amount of hydrogen bonding has occurred at that low solid content,
the straightening and separation of fibres in wet straining reduced the
potential for bonding that was developed in the press section. The
loss of internal bond strength of the dried paper was observed when
the wet stretch exceeded 1% in the open draw in pilot trials conduct-
ed by Juppi and Kaihovirta (2003). However, the tensile strength, ten-
sile stiffness and compressive strength in the machine direction will
also increase, by as much as 20% with wet straining of a web at 50%
solids (Htun 1986b, Hansson et al. 1989). Silvy (1971) found that MD
web strength increased when the web was wet strained up to about
2%, above which no strength gain is realized. Schultz (1961) found a
similar relationship where a maximum for tensile strength was ob-
served at 4% wet straining. This added strength is an obvious benefit
for the papermaker since strength of the web in MD is sought for
runnability in web-fed printing processes. This is especially beneficial
for publication paper grades made at lower basis weights and from
mechanical fibres that form weaker networks, such as newsprint,
lightweight coated and supercalendered paper grades. Wet straining
reduces the extensibility or strain-to-failure of the papers. The CD
strength properties will also be reduced.

26 Paper Drying in the Manufacturing Process

 Paper and Water

7.6.2 M D drying restraint In a conventional dryer section, the web is drawn in the machine di-
rection as it passes from the upper to the lower tiers of dryer cylin-
ders. The web is pressed to the surface of the cylinder by the dryer
fabric as it passes around the cylinders (see Fig. 8.6a). However,
there is an open draw between the tiers where the web is unsupport-
ed. Sufficient machine tension must be applied to pull the web away
from the cylinder surface, where in the early stages of drying the web
may experience some adhesion. Also, at high machine speeds the ve-
locity of the web and the air currents surrounding the fast-moving
web may cause it to flutter. Fluttering can be the source of web
breaks unless adequate tension is applied to the web in the MD. For
faster machine speeds, additional tension must be applied, which
adds to web stretching in the open draws. To maintain tension, the
rotational speed of cylinders is increased. This is done in groups, with
6-10 groups in the dryer section, with each successive group operat-
ing at slightly increased rotational rate. The stretching, or draw, expe-
rienced between each group is typically 0.2–0.3% (Kiiskinen et al.
2002). This small amount of MD straining results from the applied
drying load (web tension) that slightly exceeds the drying stress that
is characteristic of the fibrous structure of the web. This straining is
effectively uniform across the paper machine, as demonstrated by
Rutland (1992). The MD shrinkage is a function of the position across
the paper machine, labelled as ‘MD Allowed’ (Fig. 7.24). Using the
rewet web as an approximation of the dimensions of the web in the
pre-dried state, the shrinkage that occurred during the drying pro-
cess is estimated. The near zero values for ‘MD Allowed’ indicate
that when this sample is rewet, for instance during conservation pro-
cesses, very little hygroexpansion (expansion of the dried web on ex-
posure to humidity or water) will occur. Restraining the MD shrink-
age to near zero or even stretching the web causes the MD tensile
2.8
shrinkage (%)

C D potential
2.4

2.0
C D allowed
1.6

1.2

0.8
Fig. 7.24 Shrinkage across the width of MD potential
the paper web on the paper machine. 0.4
Shrinkage is greater in the cross direction MD allowed
0
(CD) than in the machine direction
(MD). In CD, shrinkage is smaller in the –0.4
centre of the sheet than along the edges. 1 3 5 7 9 11 13 15 17
Source: Rutland (1992). front cross machine position back

Paper Drying in the Manufacturing Process 27

Paper and Water

strength to [Link] ‘MD Potential’ represents the potential

shrinkage that would occur if the wet web was dried unrestrained. The
potential shrinkage of the web is a function of many variables dis-
cussed in preceding sections. The difference between the plots is the
dried-in strain. With the high dried-in strain, the MD extensibility will
be low. The results for the cross machine will be discussed in a sec-
tion below.
It was shown in a preceding section that the preferential alignment
of the fibres in the machine direction that occurs in the forming sec-
tion contributes significantly to the anisotropy of papers. The differ-
ences in the MD and CD properties could be related to the
anisotropy of individual fibres and the strength and shrinkage of the
network they form when fibres are preferentially oriented. The MD
uniaxial stress applied to the web by paper machine tension also con-
tributes significantly to the anisotropy of paper, referred to as internal
stress orientation. Setterholm and Kuenzi (1970) and Htun and Fellers
(1986) observed that tensile strength was influenced more by the ori-
entation of the fibres, while the tensile stiffness (elastic modulus) and
strain-to-failure were influenced more by the internal stress orienta-
tion induced by directional restraint in drying. Therefore, the science
of making paper relies on the control of these two variables, fibre ori-
entation and internal stress orientation, to obtain a sheet with the de-
sired level of directionality. Unfortunately, these two parameters also
have a complex role in the dimensional stability of papers and their
response to rewetting.

7.6.3 Drying strategies The drying strategy that determines the rate of drying and when re-
straint is applied to the web in the drying sequence is critical for de-
veloping properties that are important for end use, such as strength,
elasticity and extensibility. Compromises may need to be made in or-
der to optimize the properties important to a specific grade of paper.
For instance, cross machine strength may be sacrificed to improve
strength in the machine direction, or the load-bearing strength may
be developed at the expense of elasticity and toughness. While de-
tailing the various strategies used to manufacture paper is beyond the
scope of this chapter, it may be of use to the conservator to appreci-
ate that additional control is possible by applying a specific drying
strategy. Up to this point, only a comparison between unrestrained
and various levels of restraint has been considered. However, drying
strategies may include the application or omission of restraint de-
pending on the solid content the web is dried to. Htun and de Ruvo
(1983) investigated this aspect of paper drying in the simplest cases,
where full restraint was initially applied and then removed at a specific

28 Paper Drying in the Manufacturing Process

 Paper and Water

solid content, and where the paper was dried unrestrained to a solid
content and then restraint applied. The extensibility, expressed as
strain-to-failure, can be controlled by changing the drying strategy
(Fig. 7.25). In the case of ‘restrained then free’ (RF) drying, the longer
that restraint is maintained through the drying process, the less ex-
tensible the paper becomes. In contrast, the sooner restraint is ap-
plied in ‘free then restrained’ (FR) drying, the lower the extensibility.
While these observations are consistent with what has already been
described for the unrestrained and restrained drying situations, it
should be noted that the relationships are not linear with solid con-
tent. More complex strategies that combine restrained and free
shrinkage with wet straining were investigated by Htun (1986b).

10
strain to failure (%)

Fig. 7.25 Relationship between different restraint strategies for paper webs during
8 drying and the subsequent response of paper to tensile stress, expressed as strain-
to-failure. The two curves can be read as follows: the upper curve, labelled ‘free-re-
6 strained’, illustrates paper webs that were first allowed to dry to specific solids content
without restraint, and then were restrained for final drying stages. The sooner restraint
4
is applied during drying, the lower the extensibility of the sheet. The lower curve, la-
2 belled ‘restrained-free’, illustrates paper webs that were first restrained during drying
to specific solids content, and were released for the final stages of drying. The longer
0 restraint was maintained during drying, the lower the extensibility of the sheet.
40 60 80 100
solids content (%) Source: Htun and de Ruvo (1983).

The sequence of drying temperatures can also be used to control

paper properties. Castellan et al. (1985) studied how changing the
rate of temperature increase of the web influences the strength and
strain-to-failure properties of chemically pulped papers. They found
that the most rapid increase in temperature gave the lowest strength
and elasticity, although the strength properties depended very little
on the temperature programme. A gradual increase in temperature
was found to the extensibility of the paper.

7.6.4 CD shrinkage While the web is fully restrained in the machine direction as it passes
through a conventional cylinder dryer section, the shrinkage in the
cross machine direction remains essentially uncontrolled. As the web
winds between the upper and lower tiers of the dryer section (see
Fig. 8.6a), there is no external force preventing shrinkage in the CD
in the open draws. Since there is no drying load applied in the cross
machine, shrinkage will proceed unopposed from the edges toward
the centre of the web. This directional shrinkage becomes less as the
drying stress is dissipated in the network structure and the structure
itself imparts restraint from further directional shrinkage. Structural
restraint never approaches full restraint so, regardless of the distance

Paper Drying in the Manufacturing Process 29

Paper and Water

from the edge, the CD shrinkage is always greater than the MD

shrinkage for papers made using conventional cylinder drying sec-
tions. A typical cross machine profile showing the actual CD shrink-
age realized at each position is shown in Fig. 7.23 as ‘CD Allowed’.
This pattern in the CD shaped as a ‘smile’ is observed for all com-
mercial papers. The CD strain-to-failure shows a similar pattern, as
shrinkage is recovered in tensile loading, while the CD tensile
strength is essentially uniform across the width of the machine
(Chance 1994). The ‘C D Potential’ plotted in Fig. 7.24 shows that the
unrestrained shrinkage is greater than the MD restrained, although
greater shrinkage still occurs at the edges. This is attributed to the ef-
fects of fibre orientation causing greater CD shrinkage, as discussed
in a previous section.
The greater shrinkage at the edges causes a cascade of artefacts to
occur that create a non-uniformity of the paper when taken from vari-
ous positions across the web. Since the web shrinks more at the
edges, the grammage (weight per unit area) of the paper is observed
to be higher at the same positions. To compensate for this, the slice
may be narrowed at the edges so that less fibrous suspension, hence
less fibre, would be distributed to the edges. The result was a uniform
profile of the grammage across the machine. However, by narrowing
the slice, flow patterns of the suspension changed so that the fibre
orientation is no longer oriented exactly along the MD, but rather off
axis at an angle. The front and back sides of the machine are at oppo-
site angles and the centre remains aligned in the MD. The hygroex-
pansivity of the paper reflects these subtle differences in fibre orien-
tation. To overcome this problem, most contemporary paper ma-
chines are designed so that the fibrous suspension is diluted at the
edges, or wherever grammage is too high, so that the flow patterns
are not disrupted while the amount of fibres in a given position are
controlled.
Single-tier dryer sections were developed in order to address the
shrinkage profile, and to introduce restraint of CD shrinkage (see
Fig. 7.7b).In such systems the open draws between the upper and
lower tiers are eliminated. The web is continuously supported by the
dryer fabric as it passes around the vacuum roll. The vacuum roll of-
fers additional restraint while sweeping water vapor from the web.
Chance (1994) showed that the shrinkage of the web could be re-
duced from 3.3% in a conventional dryer to 1.3% shrinkage using the
single-tiered configuration. Support of the web enables the paper
machine to be operated at faster speeds without increasing the web
tension and the internal stress orientation. This results in better con-
trol over the web properties.

30 Paper Drying in the Manufacturing Process

 Paper and Water

7.6.5 Development of The preceding sections have focused on how drying affects the me-
surface properties chanical behaviour of paper. The surface properties are substantially
influenced by the manner in which the web is dried. Lindem (1991)
investigated the relationships between surface roughness and gloss
and shrinkage allowed during drying. The roughness was found to
monotonically increase with allowed shrinkage. The Parker Print Surf
roughness used in this test is based on the passage of air under a ring
in contact with the paper surface. This is referred to as an air leak
method. The gloss was found to decrease linearly as a function of
shrinkage. Restrained drying causes the bonds to align and the fibre
segments between bonds to stretch and straighten under the tension
of web shrinkage. The paper surfaces are also plasticized at high
moisture and temperature early in the drying sequence. When the
web is brought into contact with the smooth heated surface of the
dryer cylinder, the fibres in the structure conform to the planar sur-
face, levelling out peaks and out-of-plane deformation. By drying the
web unrestrained, the fibres and fibre flocs freely strain and bend un-
der the drying stresses, causing large-scale out-of-plane deformation,
as curling and cockling. It will also occur on a finer scale as twisting
and bending of fibres and microfibrils occurs. This will appear as
roughening of the surfaces, which changes both appearance and tac-
tilility.

7.7 Historical drying The mechanisms that cause shrinkage and the development of the
of paper mechanical properties of the fibrous network as it dries also apply to
the drying of papers made by hand throughout history. It is beyond
the intent of this chapter to provide the details of historical drying
and the specific effects that may be caused in papers dried by the
various methods. However, it may be useful to examine some exam-
ples of the more common methods, and how they may enhance the
properties. Loft drying was widely used in Europe and North America
before the introduction of cylinder drying on the paper machine
in 1821. The use of restraint in drying and moulding the paper on to
a smooth surface are universal techniques to enhance the paper
quality.

7.7.1 Loft drying Pressing establishes hydrogen bonding before sheet shrinkage occurs
during loft drying. In loft drying spurs of four or five sheets are draped
over ropes or wooden poles suspended in the drying loft (Fig. 7.26).
Loft drying allows free shrinkage of the paper, which then is free of
dried-in strain. The only potential restraint in traditional loft drying is
caused by the rope on which the paper was suspended along its cen-
tre, and areas of contact it has with other sheets lying adjacent in the

Paper Drying in the Manufacturing Process 31

Paper and Water

Fig. 7.26 Loft drying in a Western hand

paper mill. Damp sheets are hung in
gatherings or spurs on ropes extending
across the loft.

spur. In the moistened state the sheets in the spur adhere due to the
surface tension forces of the Campbell effect. The additional thick-
ness of the consolidated spur restrains the sheets from out-of-plane
buckling that produces cockling or curling. As the solid content of the
sheets exceeds 50%, the sheets no longer adhere to each other and
free shrinkage similar to that shown in Fig. 7.22 will occur. The poten-
tial restraint of handmade sheets resulting from these factors is negli-
gible compared to the restraining forces exerted on the web by the
paper machine. Finishing the loft-dried handmade paper with a glaz-
ing hammer, platen press or calender roll is necessary to improve pla-
narity and smoothness.

7.7.2 Contact restraint Drying methods for handmade papers used in Asia (China, Japan, Ko-
drying rea and India) rely on contact with a rigid surface to restrain dry the
paper (Hunter 1947). The simplest method is to dry the paper on the
mould on which it was formed. The paper adheres to the mould by
surface tension forces, or by the penetration of the fibres into the
woven or laid pattern. While this may provide added mechanical re-
straint of the paper, the imprinted pattern and fibre pull-off that oc-
curs on the wire side yields a roughened surface requiring finishing to
improve smoothness. The paper may be spread or ‘pasted’ against
smooth boards, masonry wall or metal plates exposed to the sun. This
will provide restraint at some point in the drying process where the
adhesion between the paper and the surface no longer resists the

32 Paper Drying in the Manufacturing Process

 Paper and Water

drying stress. The smooth surface may act as a cast for the softened
fibres on the paper surface so that, when dried, the paper surface be-
comes smoother as well. In Japanese papermaking, formed papers
were brushed on to a wall that was heated by fire on the reverse side
(Fig. 7.27). Although contact restraint improved planarity and in
some cases smoothness on one side, the drying restraint needed to
attain the high strength properties of machine-made paper was ab-
sent.

Fig. 7.27 Drying single sheets of hand-

made paper in Japan. Damp sheets are
brushed into contact with boards that are
exposed to sun to accelerate the drying
process.

7.8 Rewetting and Paper that is composed of lignocellulosic fibres is extremely hygro-
humidity response scopic, in that it readily absorbs water. However, in the manufacture
of paper of printing and writing grades, it is usually the practice to include ad-
ditives that change the paper’s response to water. These additives
could slow the penetration of water into the structure, such as sizing
agents, or may even reinforce the structure from degradation that
occurs when exposed to water, such as strengthening aids. Although
the hygroscopic cell walls were once swollen with water when the wet
web was first formed, the drying process causes the collapse and
contraction of the cell wall so that much of the pore structure is ren-
dered inaccessible or disappears entirely. Therefore, the water sorp-
tion characteristics of paper are usually quite different than the pulps
from which they were formed.
This section will focus on how the drying conditions affect the
response of the paper to the reintroduction of water, as humidity or
by wetting. Of specific interest is the hygroexpansivity of the paper,
which is the dimensional change that the paper undergoes as a func-
tion of absorbed moisture. It is these dimensional changes and the
release of residual stress that cause deformation defects such as
cockling, curling or surface roughening.
In the discussion of the drying of paper provided in the preceding
sections, no mention was made of the chemical additives, such as
sizing agents and wet and dry strengthening agents that significantly
affect the interactions of paper and water as it is reintroduced. While
an adequate treatment of these additives could take up a separate
chapter, it is important for the reader to understand the important
functions these additives provide. The hydrogen bonds that formed

Paper Drying in the Manufacturing Process 33

Paper and Water

between fibres can be easily broken if water penetrates the bond ei-
ther at the edges or through the cell wall as the fibre re-imbibes wa-
ter. Sizing agent is usually added to slow the flow of water into the
structure and may thereby protect the bond from softening and fail-
ure. Sizing may be applied as an additive to the pulp water suspen-
sion, known as an internal sizing, or directly on to the paper as a sur-
face sizing. Internal sizing agents treat the fibre and bond surfaces,
making it more difficult for water to wet the surfaces. This slows the
rate of water penetration and can protect the fibres and bonds from
swelling. Surface sizes fill some of the pore structure on the surface
and may also increase the surface’s hydrophobicity (lower tendency
to be wet by water). Usually neither sizing method renders the paper
fully waterproof. Rather, these treatments slow the rate of wetting so
that water is absorbed into the paper over the course of minutes or
hours. The drying conditions that influenced drying stress and shrink-
age will then affect how the paper responds. It is the reversibility of
the dried-in strain and the release of residual stresses that are the
subject of this section.
Salmén et al. (1985) observed that the conditions of drying re-
straint did not affect the moisture sorption characteristics of paper.
This was determined from the water sorption isotherms, where paper
moisture content is plotted as a function of relative humidity at a
fixed temperature. The isotherms for the restrained and unrestrained
samples are superimposed, even though the hygroexpansion of the
two samples differed substantially. This indicates that the access of
the structure to the penetration of water vapour was unaffected by
straining the samples during drying.

7.8.1 Cockling Cockling is an out-of-plane defect that usually occurs in random po-
sitions in the paper. It occurs when regions of 5-50 mm in diameter
undergo hygroexpansion and the network structure cannot accom-
modate the expansion within the principal plane. The non-uniformity
in dried-in stresses, possibly resulting from the structural unevenness
of the paper formation, are satisfied by local buckling of the sheet, re-
sulting in out-of-plane deformation. Cockling tends to decrease with
increased grammage since the increased bending stiffness of the
sheet resists the out-of-plane buckling (Kajanto and Niskanen 1998).
Fig. 7.28 shows the artificially generated cockles created by restrain-
ing the surrounding region and permitting the centre region to ex-
pand. Conversely, the centre region may be restrained from expan-
sion while the surrounding region expands (Brecht 1958).

34 Paper Drying in the Manufacturing Process

 Paper and Water

Fig. 7.28 Demonstrations after Brecht (1958) of the influence of non-uniform drying
on the planar dimensions of the sheet. Areas of the paper that were allowed to dry
first were restrained by the still expanded wet areas that were kept damp. The first-
dried areas remained partially expanded. The last-dried areas were able to shrink fully
and force the first-dried areas out of plane. In (a), the area within the circles was dried
first, in (b) the area around the circles was dried first. Source: Buchschuster (2007).

7.8.2 Dimensional stability Dimensional stability of paper refers to the resistance of paper to hy-
groexpansivity, the change in dimension, or strain, associated with a
change in water content, or thermoexpansivity, which is the dimen-
sional change that accompanies a change in temperature. In this sec-
tion we will only consider the hygroexpansivity of paper. Uesaka
(2002) provided a comprehensive review of dimensional stability as it
strain (%)

0.12 B
D pertains to paper. Soon after paper is made it reaches equilibrium
0.08
with its surrounding relative humidity. The final moisture content of
0.04 the paper determined in the dryer section was selected so that only a
0
A
small change in moisture content would be required for that equilib-
–0.04 rium to be met. If, at some point in time, the relative humidity sur-
–0.08 C
irreversible shrinkage
rounding the paper changes, the paper will either gain or lose mois-
E

4 6 8 10 12
ture content. A plot of these two variables is the water sorption iso-
moisture content (%) therm (see page 000). If the paper gains moisture, the fibres and the
Fig. 7.29 Dimensional stability of paper fibrous structure should swell, resulting in an increase in the linear di-
in M D under changing climatic condi- mensions of the sheet. The increased moisture may also weaken bonds
tions plotted as the relationship be-
tween the percentage strain and the and relieve residual stress. So in the situation where the sheet was
moisture content of the paper. Strain is stretched during drying, there is a chance the shrinkage will initially
taken as a measure for change in dimen- occur as the residual stress is relieved. Thereafter, increased sheet
sion or hygroexpansivity. The uppermost
curve shows the initial expansion of the expansion will be proportional to sheet moisture content (Fig. 7.29)
humidified paper and its subsequent (Uesaka et al. 1992). The irreversible shrinkage that appears in that
shrinkage during drying, at which point plot results directly from the dried-in strain introduced by MD ten-
dried-in strain imparted to the paper
during manufacture is released. This
sion during the papermaking process. After the drying strain is recov-
shrinkage of the paper is irreversible (see ered, the sheet then swells and contracts directly with moisture con-
also Fig. 13.12). The lower curve shows tent in subsequent cycles. A similar plot for the hygroexpansion in
that during a subsequent humidification
the cross machine shows no initial decrease in strain (Fig. 7.30) (Ue-
and drying cycle, sheet swells and con-
tracts directly with its moisture content. saka et al. 1992). Since the web is effectively unrestrained in the CD,
Source: Uesaka et al. (1992). there is no dried-in strain to recover. Thus, the expansion and con-

Paper Drying in the Manufacturing Process 35

 Paper and Water

0.8 traction appear to be superimposed with no irreversible shrinkage.

strain (%)

0.7
0.6
The hygroexpansion coefficients are the slopes of the lines, defined
0.5 as the change in dimension divided by the change in humidity. The
0.4 hygroexpansion coefficient for the MD shown in Fig. 7.29 is
0.3
0.2
0.036%/% RH. The CD hygroexpansion coefficient from Fig 7.30 is
0.1 0.140%/% RH. The reasons for the difference are attributed to those
0 structural factors that control shrinkage discussed in the context of
–0.1
4 6 8 10 12 network shrinkage (section 7.4). In this case the increased transverse
moisture content (%)
hygroexpansion of the fibres expands the bonded regions that expe-
Fig. 7.30 Dimensional stability of paper
in CD under changing climatic condi-
rienced microcompression during drying shrinkage. This results in in-
tions plotted as the relationship be- creased CD expansion. Salmén et al. (1985) found that for oriented
tween the percentage strain and the sheets with a restrained tensile strength ratio MD/CD of 2.5, the hy-
moisture content of the paper. In the
groexpansion coefficients were six times greater in the CD than in
cross direction, there is no dried-in strain
to recover and therefore the expansion the MD.
and contraction of the paper relates di- The in-plane hygroexpansivity of isotropic handsheets dried freely
rectly to its moisture content.
is more than twice that of the same handsheets dried under restraint
Source: Uesaka et al. (1992).
(Fig. 7.31). This graph shows the effect of increasing density, an indi-
cation of interfibre bonding, on the hygroexpansivity. The unre-
strained sheet, freely dried, shows a slight increase in hygroexpansivi-
in-plane hygroexpansivity (%/% M C)

ty with density, suggesting the increased bonding caused more drying

0.06 shrinkage and more hygroexpansivity in the paper.
dried freely
The hygroexpansivity of machine-made papers changes as a func-
tion of density (Fig. 7.32). For the sheet dried without restraint, the
0.04 differences in MD and CD hygroexpansivity are clearly shown. These
are attributed to the free shrinkage that occurs in the CD, which al-
lows a greater expansion to occur as water is reintroduced into the
0.02 dried under restraint
structure. Since the MD was restrained, its hygroexpansivity is much
lower. The axial stiffness of the fibres, preferentially oriented in the
0
MD, along with the MD machine tension that restrained initial shrink-
250 500 750 1000
density (kg/m³)
age would also contribute to the dimensional stability in the MD. Ad-
ditives such as fillers, that are dimensionally stable, or debonding
Fig. 7.31 Relationship between the den-
agents, that reduce shrinkage by interfering with bond strength, will
sity of isotropic paper (handsheet, no
directionality) and its in-plane hygroex- also improve the dimensional stability of papers. However, in both
pansivity (lateral expansion of the sheet cases the strength of the paper is reduced, a compromise that may
dimensions). The paper dried freely is not be acceptable for the papermaker.
more labile to hygroexpansion than the
paper dried under restraint. The Z-direction (ZD) hygroexpansivity is always much greater than
Source: Salmén et al. (1985). the in-plane (MD or CD) values. This change in the structure has sig-
nificant influence on several paper properties. First, bending stiffness
is highly dependent on thickness, increasing by the cubed power. The
decreased interfibre bonding, associated with a thickness increase,
will increase the scattering coefficient (opacity). Increase in thickness
will also cause the Z-directional strength to decrease so that the pa-
per is more subject to delamination.

36 Paper Drying in the Manufacturing Process

 Paper and Water

Fig. 7.33 shows the out-of-plane (ZD) hygroexpansivity as a func-

in-plane hygroexpansivity (%/%M C)

tion of sheet density, i.e. bonding, for the same papers shown in Figs
0.12
CD 7.29 and 7.30 (Salmén et al. 1985). The values for the ZD dried freely
× MD
dried freely increase with density, showing the effect of the swelling of the fibre
dried under cell walls transferred directly to expansion of the web structure. Z D
restraint
0.08 hygroexpansivity is about 10 times greater than the in-plane average
hygroexpansivity. For restrain-dried sheets, this factor is 30 times
×
× × greater for the ZD, showing not only the effects of fibre swelling, but
0.04 × ×
× × × ×
also the release of internal stress that breaks bonds and allows more
×
out-of-plane deformation of the fibres.
0 Sung et al. (2005) investigated the Z-directional hygroexpansivity
250 500 750 1000
density (kg/m³)
of printing papers by mapping the change in thickness as a function
of relative humidity from 50% to 80% RH, then from 80% to 90%
Fig. 7.32 Relationship between the den-
sity of anisotropic paper (machine-
RH. In the study of newsprint and copy paper, they found that ma-
made paper with directionality) and chine-calendered or super-calendered papers exhibited a hysteresis
its in-plane hygroexpansivity (lateral when subjected to an increase then decrease in moisture content
expansion of the sheet dimensions).
(Fig. 7.34). This irreversibility of hygroexpansion was attributed to re-
The graph shows that there is a marked
difference in the MD and C D hygroex- lease of the extreme stress imparted by the calendar densification. It
pansion of the paper that was dried is also a recovery of the collapsed lumen and recovery of the fibres to
without restraint. more tubular shape as demonstrated by Forseth and Helle (1997).
Source: Salmén et al. (1985).
Newsprint has significant initial expansion (0.58%/% RH). This ex-
pansion is not fully reversed and the thickness remains 10% greater
than initially. The copy paper has a hygroexpansivity of 0.16%/%RH,
and only a small amount of irreversible swelling. One would expect
this since this sample had 20% mineral matter, one benefit of which
is that it adds dimensional stability to the paper. While in-plane di-
out-of-plane hygroexpansivity (%/%MC)

1.2 dried mensional instability is easily perceived, by the appearance of cockles

under restraint
or change in length dimensions, out-of-plane (Z-directional) expan-
0.8
sion that accompanies reintroduction of water can be difficult to

100 Fig. 7.34 The thickness of copy paper

thickness (µm)

0.4 dried freely

and newsprint as a function of the
90 copy
moisture content as controlled by
changing the relative humidity. The
0
80 newsprint irreversible expansion of both papers
250 500 750 1000
density (kg/m³) is evident in the plot. Thickness was
70 measured at 250,000 locations within
Fig. 7.33 Relationship between the den- a 12 mm square region of the paper
sity of oriented machine-made paper using opposing laser profilometers.
and its out-of-plane (Z-directional) hy- 60
The method is non-contacting and
groexpansivity (thickness expansion of has a precision of 1 µm in three di-
the sheet). The graph shows the direct 50
mensions.
relationship between hygroexpansivity Source: Sung et al. (2005).
and bonding for papers dried without 40
restraint. 0 5 10 15 20
Source: Salmén et al. (1985). moisture content (%)

Paper Drying in the Manufacturing Process 37

Paper and Water

detect. However, the profound influence it will have on the paper’s

properties, such as opacity, bending stiffness and mechanical
strength, can be easily measured.

7.8.3 Hornification Jayme (1944) introduced the term ‘irreversible hornification’ to de-
scribe the loss of rewettability and swellability of low-yield chemical
pulps once they have been dried. Minor (1994) and Laivins and Scallan
(1993) provided reviews of the subject. This phenomenon originates in
the changes in the cell wall and microstructure of the fibre that were
discussed in section 7.2. The collapse of the pore structure within the
cell wall, the formation of hydrogen bonds between lamellae and in-
creased crystallization of the cellulose molecules cause the fibres to
stiffen. The rate of water absorption into the fibre also slowed sub-
stantially. This has important consequences for papermakers as they
select the raw materials for their process. Pulps used for papermaking
are most commonly derived from one of three categories: never-
dried pulps, dried pulp lap, or recycled fibres. Never-dried pulps, as
the name implies, remain wet from the tree, through the pulping and
bleaching process and into stock preparation. The pulp may be ob-
tained directly from a pulp mill in the vicinity of the paper machine.
The pulp may also be received as wet pulp lap, which is received as
bales of wet pulp sheets. In both cases, the fibres remain swollen and
flexible, which provides good bonding and sheet strength. Such pulps
will have high shrinkage potential for reasons discussed in section 7.2.
Shipping costs may be reduced if the pulps are first dried. The fibres
contained in dried pulp lap will be hornified. It is necessary to refine
or beat the fibres in order to condition the cell wall to promote
swelling, flexibility and improved bonding associated with more
hydrated fibres. Recycled fibres will also be hornified and may also
require conditioning to promote bonding for sheet strength. The
fibre saturation point (water contained within the cell wall) for never-
dried pulp is higher than that for pulp which was dried and rewet (see
Fig. 7.16).
Hornification has implications for the conservators for at least two
reasons. First, the furnish used to make a paper, whether it is never
dried or from dried fibres that have been rewet, will influence the
strength properties of the paper. Consideration of the source fibres
may allow a better understanding of how a paper will respond to con-
servation treatments that require exposure to water. It may also eluci-
date the strength and stiffness properties observed for a given speci-
men. Secondly, in rehydration of paper during conservation treat-
ments, the fibres never fully absorb water to their original state. Thus
the releases of internal stress, the swelling of fibres and the breaking

38 Paper Drying in the Manufacturing Process

 Paper and Water

of all hydrogen bonds may take extensive amounts of time to fully

achieve. Therefore, the length of time that a paper is exposed to wa-
ter will influence the extent to which the rehydration proceeds. Oth-
er variables will influence this, including the fibre types, the pulping
process and the extent of refining or beating. Sizing agents, surface
treatments and web density will also play a significant role in the rate
of rehydration. Thus, hornification describes fibres that, once dried
below the fibre collapse point, never fully recover the properties they
had in never-dried pulps.

Summary This chapter has focused on the hidden forces that are frozen into
the paper when it is manufactured. These forces, referred to as inter-
nal stresses, are responsible for many of the mechanical characteris-
tics of paper that give it valuable end-use properties. The introduc-
tion of water or heat usually disrupts these forces, causing an imbal-
ance that produces dimensional changes at small and large scales.
The result is irreversible contortion of the structure so that the
smooth planar form that it had when first made is lost. In order for
the paper conservator to consider solutions to the treatment of pa-
per exposed to wet or humid conditions, or before using water in
conservation methods, a familiarity of the conditions that give rise to
internal stresses and the result of their release is important.
In discussing the manufacturing process, a justification for the co-
pious amounts of water required to obtain a uniform distribution of
the fibres is given. This sets the stage for the manner in which fibres
are organized in the fibrous structure. The relative non-uniformity of
the fibrous structure is typical of all papers, although excessive floc-
culation of fibres is considered a disadvantage. The fibres are de-
posited stochastically in stratified layers, giving rise to a difference in
the properties within the plane of the web (M D-CD) and through
the thickness (ZD). Fibres also tend to align preferential along the
machine direction, or direction of flow, of the papermaking web,
causing significant differences in the machine direction and cross ma-
chine direction properties. The small- and large-scale positioning of
the fibres has a dramatic influence on the shrinkage that occurs when
water is removed and the expansion when it is reintroduced. This
chapter has discussed how the dewatering processes change along
the paper machine, from removing large quantities of water from the
fibrous suspension in the wet end, to the final conditioning of the
web in the dry end, to bring it close to a state that will be at equilibri-
um with typical humidity conditions. Superimposed on this discus-
sion are the structural changes, mostly shrinkage that the web under-
goes as it goes from the wet to the dry state. The influence of the

Paper Drying in the Manufacturing Process 39

Paper and Water

processes that act in opposition to web shrinkage, notably machine

tension and the dryer fabric, is also discussed. Analogies between
machine-made paper and historical handmade papers are provided,
since all of the underlying mechanisms that control shrinkage are the
same for both.
The elongated shape, complex microstructure and hygroscopicity
of papermaking fibres cause characteristic dimensional changes when
water is removed and reintroduced. Upon drying, hollow fibres col-
lapse into ribbon structures and shrink many times more in the trans-
verse direction (across the fibres) than axial (along the length). When
fibres are bonded into the web structure, this shrinkage affects the
entire structure by causing large-scale shrinkage, and fine scale inter-
nal stress as shrinkage is restrained. Preconditioning of fibres through
pulping and refining or beating causes fibres to accept more water
during the early stages of the papermaking process in order to im-
prove bonding. However, during drying, this water is yielded so that
fibres shrink more. This in turn significantly affects the dimension
changes in the fibrous structure, as well as the internal stresses that
build within. The intrinsic potential shrinkage of the papermaking
fibres is an important parameter when considering the dimensional
response of papers. During manufacturing, the fibres in the web col-
lapse and dry, increase their stiffness and strengthen, known as
hornification. Although still hygroscopic, this process is not fully re-
versible, and so the original state of swelling and flexibility that the
fibres once exhibited in the papermaking dilute stock fed to the
headbox. Conservators must consider that water does cause fibres to
swell, and internal stress to be released, but it is not possible for
fibres to attain the original state of hydration while still locked in the
fibrous structure of paper.
Finally, a discussion of the dimensional changes that a sheet ex-
hibits when rewet by water is given. The release of stress and the soft-
ening of interfibre bonds cause the web to expand to greater and
lesser extents at different locations. When the web cannot accommo-
date this swelling within the structure, buckling occurs. The web ex-
periences out-of-plane deformation, most often known as cockling.
The paper is left in a state much different than what it was in when
first manufactured. The only recourse for a conservator is to re-enact
the papermaking process by rewetting the sheet and drying the paper
under similar restraint to that which was done originally. However,
because of hornification of the fibres, this re-enactment will never be
as convincing as the real process.

40 Paper Drying in the Manufacturing Process

 Paper and Water

References Baum, G. A. (1991). Sheet structure considerations – paper as an engineered

material. In: Paper Machines Operations (B. A. Thorpe, ed.), 3rd edition,
pp. 54–84. Atlanta, GA: Joint Textbook Committee of the Paper Industry,
TA P P I P R E S S .

Brecht, W. (1958). Beating and hygrostability of paper. In: Fundamentals of

Papermaking Fibres (F. Bolam, ed.), Vol. 1, pp. 241–262. Kenley: Technical
Section of the British Paper and Board Maker’s Association.

Brecht, W. and Pothmann, D. (1955). Der Einfluß der beim Trocknen von
Ppaier herrschenden Zugkräfte auf das Verhalten von Papier. Das Papier 9,
429–437.

Brecht, W., Gerspach, A. and Hildenbrand, W. (1956). Trocknungsspannun-

gen in ihrem Einfluss auf einige Papiereigenschaften. Das Papier 10, 454–458.

Buchschuster, P. (2007). Von den (Un)Gleichmäßigkeiten der Trocknungs-

schrumpfung. Unpublished report, Stuttgart: Staatliche Akademie der
Bildenden Künste Stuttgart.

Byrd, V. L. (1974). Web shrinkage energy: an index of network fiber bonding.

Tappi Journal 57, 87–91.

Campbell, W. B. (1959). The mechanism of bonding. Tappi Journal 42,

999 –1001.

Castellan, A., Doignie, J. C. and Pommier, J. C. (1985). Effect of different dry-

ing temperature programmes on some mechanical properties of an un-
bleached kraft paper. Svensk Papperstidning 88, R44–R47.

Chance, J. L. (1994). Overview of the dryer section. In: 1994 Practical Aspects
of Pressing and Drying Seminar, pp. 271–297. Atlanta, GA: TA PPI PRESS.

Corte, H. and Herdman, P. T. (1975). Zur Schrumpfung von Papier bei der
Trocknung. Das Papier 29, 288–295.

Forseth, T. and Helle, T. (1997). Effect of moistening on cross-sectional

details of calendered paper containing mechanical pulp. Journal of Pulp and
Paper Science 23, J95–100.

Gates, E. R. and Kenworthy, I. C. (1963). Effects of drying shrinkage and fiber

orientation on some physical properties of paper. Paper Technology 4,
485–494.

Hansson, T., Fellers, C. and Htun, M. (1989). Drying strategies and new
restraint technique to improve cross-directional properties of paper.
In: Fundamentals of Papermaking, Vol. 2 (C. F. Baker, ed.), pp. 743–781.
London: Mechanical Engineering Publications.

Paper Drying in the Manufacturing Process 41

Paper and Water

Hill, R. L. (1967). The creep behavior of individual pulp fibers under tensile
stress. Tappi Journal 50, 432–440.

Htun, M. (1986a). Internal stress in paper. In: Paper – Structure and Properties
(J. A. Bristow and P. Kolseth, eds), pp. 227–239. New York: Marcel Dekker.

Htun, M. (1986b). The control of mechanical properties of drying restraints.

In: Paper – Structure and Properties (J. A. Bristow and P. Kolseth, eds),
pp. 311–325. New York: Marcel Dekker.

Htun, M. and de Ruvo, A. (1977). Relation between drying stresses and inter-
nal stresses and the mechanical properties of paper. In: Fibre-Water Interac-
tions in Papermaking, Vol. 1, pp. 477–487. London: Technical Division of the
British Paper and Board Industry Federation.

Htun, M. and de Ruvo, A. (1978). Correlation between the drying stress and
the internal stress in paper. Tappi Journal 61, 75-77.

Htun, M. and de Ruvo, A. (1983). The influence of drying strategies on the re-
lationship between drying shrinkage and strain to failure of paper. In: The Role
of Fundamental Research in Papermaking, Vol. 1 ( J. Brander, ed.), pp. 385–398.
London: Mechanical Engineering Publications.

Htun, M. and Fellers, C. (1986). The in-plane anisotropy of paper in relation

to fiber orientation and drying restraints. In: Paper – Structure and Properties
(J. A. Bristow and P. Kolseth, eds), pp. 327–345. New York: Marcel Dekker.

Hudson, F. L. (1963). The effect on load elongation properties of wetting

paper and redrying under cross direction tension. Svensk Papperstidning 66,
60–63.

Hunter, D. (1947). Early papermaking processes and methods. In: Paper-

making: The History and Technique of an Ancient Craft (D. Hunter, ed.),
2nd edition, pp. 170–202. New York: Alfred A. Knopf.

Ivarsson, B. W. (1954). Introduction of stress into paper sheet during drying.

Tappi Journal 37, 634–639.

Ivarsson, B. W. and Steenberg, B. (1947). Paper as a visco-elastic body.

III. Application of Eyring’s and Halsey’s theory to stress-strain diagrams of
paper. Svensk Papperstidning 50, 419–432.

Jayme, G. (1944). Mikro-Quellungsmessungen and Zellstoffen.

PapierFabrikation /Wochenblatt für Papierfabrikation 6, 187–194.

Jentzen, C. A. (1964). Effect of stress applied during drying on some of the

properties of individual pulp fibers. Tappi Journal 47, 412–418.

42 Paper Drying in the Manufacturing Process

 Paper and Water

Johanson, F. and Kubát, J. (1964). Measurement of stress relaxation in paper.

Tappi Journal 67, 822–832.

Johanson, F., Kubát, J. and Pattyranie, C. (1967). Internal stresses, dimensional

stability and deformation of paper. Svensk Papperstidning 70, 333–338.

Juppi, K. and Kaihovirta, J. (2003). The effect of the dryer section on paper
quality. Pulp and Paper Magazine of Canada 104, 58–61.

Kajanto, I. and Niskanen, K. (1998). Dimensional stability. In: Paper Physics,

Papermaking Science and Technology, Book 14, (K. Niskanen, ed.), pp. 223–
259. Helsinki: Fapet Oy, Finnish Paper Engineers’ Association and TA PPI.

Kallmes, O. J. and Perez, M. (1966). Load/elongation properties of fibres. In:

Consolidation of the Paper Web, Vol. 1 (F. Bolam, ed.), pp. 507–528. London:
Technical Section of the British Paper and Board Makers’ Association.

Kerekes, R. J. and Schell, C. J. (1992). Characterization of fibre flocculation by

a crowding factor. Journal of Pulp and Paper Science 18, 32–38.

Kiiskinen, H., Paltakari, J. and Pakarinen, P. (2002). Drying and paper quality.
In: Papermaking Part 2, Drying, Papermaking Science and Technology, Book 9
(M. Karlsson, ed.), pp. 332–368. Helsinki: Fapet Oy, Finnish Paper Engineers’
Association and TA PPI.

Kim, C. Y., Page, D. H., El-Hosseiny, F. and Lancaster, A. P. S. (1975). Me-

chanical properties of single wood pulp fibers – the effects of drying stress
on strength. Journal of Applied Polymer Science 19, 1549–1561.

Laivins, G. V. and Scallan, A. M. (1993). The mechanism of hornification of

wood pulps. In: Products of Papermaking, Vol. 2 (C. F. Baker, ed.),
pp. 1235–1260. Leatherhead: Pira International.

Lindem, P. E. (1991). Paper surface properties versus stress during drying.

In: Proceedings of the 1991 International Paper Physics Conference, Vol. 1
(R.E. Mark, ed.), pp. 327–348. Atlanta, GA: TA PPI PRESS.

Lyne, L. M. and Gallay, W. (1954). Fiber properties and fiber-water relation-

ships in relation to the strength and rheology of wet webs. Tappi Journal 37,
581–596.

Minor, J. L. (1994). Hornification - Its origin and meaning. Progress in Paper

Recycling 3, 93–95.

Nanko, H. and Ohsawa, J. (1989). Mechanism of fibre bond formation. In:

Fundamentals of Papermaking, Vol. 2 (C. F. Baker, ed.), pp. 783–811. London:
Mechanical Engineering Publications.

Paper Drying in the Manufacturing Process 43

Paper and Water

Nanko, H. and Wu, J. (1995). Mechanisms of paper shrinkage during drying.

In: Proceedings of the 1995 International Paper Physics Conference, Vol. 1
(C. J.T. Dodson, ed.), pp. 103–113. Montreal: Canadian Pulp and Paper
Association.

Nanko, H., Asano, S. and Ohsawa, J. (1991). Shrinking behavior of pulp fibers
during drying. In: Proceedings of the 1991 International Paper Physics Confer-
ence, Vol. 2, (ed.), pp. 365–373. Atlanta, GA: TA PPI PRESS.

Page, D. H. and Tydeman, P. A. (1962). A new theory of the shrinkage, struc-

ture and properties of paper. In: The Formation and Structure of Paper, Vol. 1
(F. Bolam, ed.), pp. 397–414. London: Technical Section of the British Paper
and Board Makers’ Association.

Page, D. H. and Tydeman, P. A. (1963). Transverse swelling and shrinkage of

softwood tracheids. Nature 199, 471–472.

Parsons, S. R. (1972). Effect of drying restraint on handsheet properties.

Tappi Journal 55, 1516–1521.

Rance, H. F. (1954). Effect of water removal on sheet properties.

Tappi Journal 37, 640–454.

Retulainen, E., Niskanen, K. and Nilson, N. Fibers and bonds (1989). In: Paper
Physics, Papermaking Science and Technology, Book 16 (K. Niskanen, ed.),
pp. 55–87. Helsinki: Fapet Oy, Finnish Paper Engineers’ Association and
TA P P I .

Rutland, D. F. (1992). Dimensional stability and curl. In: Mill Control and
Control Systems: Quality and Testing, Environmental, Corrosion, Electrical
(M. Kouris, ed.), 3rd edition, pp. 132–151. Atlanta, GA: The Joint Textbook
Committee of the Paper Industry and TA PPI PRESS.

Salmén, L., Fellers, C. and Htun, M. (1985). The in-plane and out-of-plane
hygroexpansional properties of paper. In: Papermaking Raw Materials, Vol. 2
(V. Punton, ed.), pp. 511–527. London: Mechanical Engineering Publications.

Salmén, L., Fellers, C. and Htun, M. (1987). The development and release of
dried in stresses in paper. Nordic Pulp and Paper Research Journal 2, 44–48.

Schultz, J. H. (1961). The effect of straining during drying on the mechanical

and viscoelastic behavior of paper. Tappi Journal 44, 736–744.

Setterholm, V. C. and Kuenzi, E. W. (1970). Fiber orientation and degree of

restraint during drying: effects of tensile anisotropy of paper handsheets.
Tappi Journal 53, 1915.

44 Paper Drying in the Manufacturing Process

 Paper and Water

Silvy, J. (1971). Effects of drying on web characteristics. Part 2 - The condi-

tioning of the structure during drying: effects of web characteristics. Paper
Technology 12, 445-451.

Smith, S. F. (1950). Dried-in strains in paper sheets and their relation to curl-
ing, cockling and other phenomena. The PaperMaker and British Paper Trade
Journal 119, 185–192.

Spiegelberg, H. L. (1966). The effect of hemicelluloses on the mechanical

properties of individual pulp fibers. Tappi Journal 49, 388–396.

Stone, J. E. and Scallan, A. M. (1966). Influence of drying on the pore struc-

tures of the cell wall. In: Consolidation of the Paper Web, Vol. 1 (F. Bolam, ed.),
pp. 145–166. London: Tech. Section of the British Paper and Board Makers’
Association.

Stone, J. E., Scallan, A. M. and Abrahamson, B. (1968). Influence of beating on

cell wall swelling and internal fibrillation. Svensk Papperstidning 71, 687–694.

Sung, Y. J., Ham, C. H., Kwon, H. L., Lee, H. L. and Keller, D. S. (2005) Appli-
cations of thickness and apparent density mapping by laser profilometry.
In: Advances in Paper Science and Technology, Vol. 2 (S. I’Anson, ed.),
pp. 961–1007. Frecheville Court: Pulp Paper Fundamental Research Society.

Tydeman, P. A., Wembridge, D. R. and Page, D. H. (1966). Transverse shrink-

age of individual fibers by micro-radiography. In: Consolidation of the Paper
Web, Vol. 1 (F. Bolam, ed.), pp. 119–144. London: Technical Section of the
British Paper and Board Makers’ Association.

Uesaka, T. (2002). Dimensional stability and environmental effects on

paper properties. In: Handbook of Physical Testing of Paper (R. E. Mark,
C. E. Habeger, J. Borch and M. B. Lyne, eds), 2nd edition, pp. 115–171.
New York: Marcel Dekker.

Uesaka, T., Moss, C. and Nanri, Y. (1992). The characterization of hygroex-

pansivity of paper. Journal of Pulp and Paper Science 18, J11–J16.

Waterhouse, J. F. (2002). Residual stresses in paper and paperboard. In:

Handbook of Physical Testing of Paper (R. E. Mark, C. E. Habeger, J. Borch and
M. B. Lyne, eds), 2nd edition, pp. 527–562. New York: Marcel Dekker.

Zeronian, S. H. (1985). Chapter 5: Intercrystalline swelling of cellulose.

In: Cellulose Chemistry and Its Applications, (T. P. Nevell and S. H. Zeronian,
eds) pp. 138–158. New York: John Wiley & Sons.

Paper Drying in the Manufacturing Process 45