Hydrophobic Surface Sizing of Testliner: March 2019

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/334326749
Hydrophobic surface sizing of testliner
Thesis · March 2019
CITATIONS READS
2 1,856
1 author:
Misla Lagus
Åbo Akademi University
2 PUBLICATIONS 2 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Surface sizing View project
All content following this page was uploaded by Misla Lagus on 16 July 2019.
The user has requested enhancement of the downloaded file.

HYDROPHOBIC SURFACE SIZING OF
TESTLINER
Misla Lagus
Faculty of Science and Engineering
Åbo Akademi University
2019
M. Lagus: Hydrophobic surface sizing of testliner II
ÅBO AKADEMI – FACULTY OF SCIENCE AND ENGINEERING

Abstract
Subject: Surface sizing of testliner

Author: Misla Lagus
Title of work: Hydrophobic surface sizing of testliner
Supervisor: Martti Toivakka Assistant supervisor: Kimmo Huhtala
Abstract
The demand for inexpensive, sustainable packaging is rising continuously, and
surface sizing of testliner is gaining importance. In this study, the effect of sizing
method, base paper properties, and surface sizing agent (SSA) properties on starch
pickup, paper strength, and sizing effect were examined.
The starch pickup was shown to depend on surface size solids content and base
paper porosity. The strength of the base paper was tied to the porosity, a more
porous base paper was weaker, as expected. Surface sized paper strength depended
on the base paper strength and the starch amount added.
Differences in sizing effect were witnessed both between different SSAs and
between different base papers. Wettability and starch absorption of base paper
affected the sizing effect. SSA particle size, SSA hydrophobicity, and poly
aluminum chloride (PAC) usage also affected the sizing effect.
Keywords: surface sizing, testliner, packaging, surface sizing agent, base paper
properties, hydrophobicity, sizing effect
Date: 13.3.2019 Pages: 72

M. Lagus: Hydrophobic surface sizing of testliner III
ÅBO AKADEMI – FAKULTETEN FÖR NATURVETENSKAPER

OCH TEKNOLOGI
Abstrakt
Ämne: Ytlimning av testliner

Författare: Misla Lagus
Arbetets titel: Vattenavstötande ytlimning av testliner
Handledare: Martti Toivakka Biträdande handledare: Kimmo Huhtala
Abstrakt
Efterfrågan på billiga, hållbara förpackningar stiger kontinuerligt, och ytlimning av
testliner blir samtidigt allt viktigare. I denna studie undersöktes effekten som
baspapprets egenskaper, ytlimningsmedlets egenskaper och ytlimningsmetoden har
på stärkelseupptagningen, pappersstyrkan och ytlimningseffekten.
Stärkelselösningens torrhalt och baspapprets porositet påverkade

stärkelseupptagningen. Porositeten påverkade baspapprets styrka, ett mer poröst
papper var svagare, som förutsett. Både baspapprets styrka och stärkelsemängden
påverkade det ytlimmade papprets styrka.
Skillnader i ytlimningseffekten kunde ses både mellan de olika baspappren och

mellan de olika ytlimningsmedlen. Ytlimningseffekten påverkades av baspapprets
vätbarhet och stärkelseabsorption. Ytlimningsmedlets partikelstorlek,
ytlimningsmedlets hydrofobitet och användningen av poly aluminiumklorid hade
också inverkan på ytlimningseffekten.
Nyckelord: ytlimning, testliner, förpackning, ytlimningsmedel,

baspappersegenskaper, hydrofobitet, limningseffekt
Datum: 13.3.2019 Sidantal: 72

M. Lagus: Hydrophobic surface sizing of testliner IV
Acknowledgements
First and foremost I would like to thank my assistant supervisor Kimmo Huhtala for
guiding the work, for listening to my ideas and suggestions, for motivating me to finish
writing the thesis, and most of all for always asking me to aim a little bit higher.
Sheryl Hildén, Kai Dahlskog and Jussi Salonen, thank you for our insightful
conversations and for sharing your industry knowledge.
I would like to thank Mia Ahokas for providing invaluable insight and advice during
the writing process.
Sanna Salonen, thanks for your innovative ideas and advice on laboratory practicalities.
I would further like to thank everyone working at CH-Polymers laboratory facilities in

Raisio. I will always be grateful for how every one of you have shared your expertise
and knowledge with me without restraint. Special thanks to Taru Ojala-Lehtonen and
Minna Lamminen for your continued friendship, and for teaching me most of the
laboratory analyses.
I would like to thank Professor Martti Toivakka and Assistant Professor Mari Nurmi
for trusting me to complete this master’s thesis and for the final touches.
I am also grateful to everyone involved in sending base paper samples from the
participating paper mills and to Business Finland Oy (former TEKES) for providing
funding for the work inside the BIOBIND project.
M. Lagus: Hydrophobic surface sizing of testliner V
Table of contents
Abstract ...................................................................................................................... II
Abstrakt .................................................................................................................... III
Acknowledgements .................................................................................................. IV
List of tables and figures ........................................................................................ VII
List of abbreviations ................................................................................................ IX
Introduction ................................................................................................................ 1
I LITERATURE REVIEW................................................................ 3
1. Surface size application method ........................................................................ 3

1.1. Pond type size press ................................................................................... 3
1.2. Metered film press ..................................................................................... 5
1.3. Novel and alternative methods ................................................................... 8
2. Surface size .......................................................................................................... 9

2.1. Starch ......................................................................................................... 9
2.1.1. Starch in the paper industry .................................................................. 10
2.1.2. Dispersion and conversion ................................................................... 11
2.1.3. Retrogradation ...................................................................................... 12
2.1.4. Modification ......................................................................................... 13
2.2. Sizing agent .............................................................................................. 15
2.2.1. Internal sizing agents ............................................................................ 15
2.2.2. Surface sizing agents (SSA) ................................................................. 17
2.3. Other surface size additives ..................................................................... 19
2.4. Improving print quality with surface sizing ............................................. 20
3. Testliner base paper ......................................................................................... 21

3.1. Testliner ................................................................................................... 21
3.2. Raw material ............................................................................................ 22
3.3. Papermaking process................................................................................ 22
4. Hydrophobization and water penetration theory.......................................... 25

M. Lagus: Hydrophobic surface sizing of testliner VI
II EXPERIMENTAL PART ............................................................. 27
5. Materials............................................................................................................ 27
6. Methods ............................................................................................................. 29
6.1. Laboratory surface sizing method ............................................................ 29
6.1.1. Surface sizing equipment ..................................................................... 29
6.1.2. Experiment layout for testing the method ............................................ 30
6.1.3. Adjusting the laboratory surface sizing parameters ............................. 31
6.1.4. Surface sizing method used for hydrophobicity tests ........................... 32
6.2. Experiment layout for hydrophobization efficiency testing .................... 33
6.3. Paper analysis ........................................................................................... 33
6.3.1. Hydrophobicity ..................................................................................... 33
6.3.2. Structure and porosity .......................................................................... 34
6.3.3. Strength properties ............................................................................... 35
6.3.4. Other paper analyses ............................................................................ 35
6.4. Surface sizing agent (SSA) analysis ........................................................ 36
III RESULTS AND DISCUSSION .................................................... 37
7. Differences in SSA efficiency ........................................................................... 37
8. Effect of surface sizing method parameters ................................................... 39
9. Effect of base paper properties ....................................................................... 44

9.1. Base paper properties ............................................................................... 44
9.2. Starch pickup............................................................................................ 46
9.3. Paper strength ........................................................................................... 47
9.4. Sizing effect ............................................................................................. 49
10. Effect of SSA properties on sizing effect ..................................................... 52
Conclusions ............................................................................................................... 54
Swedish summary - Svensk sammanfattning ........................................................ 56
References ................................................................................................................. 56
M. Lagus: Hydrophobic surface sizing of testliner VII
List of tables and figures
Table 1. Benchmark settings for surface sizing parameter testing ............................. 30

Table 2. The surface sizing method used in further experiments ............................... 32
Table 3. Viscosity of surface size with Starch 1 and Starch 2 ................................... 41
Table 4. Basic base paper properties .......................................................................... 44
Table 5. Hg-porosimetry ............................................................................................ 45
Table 6. DWR and ultrasonic dynamic penetration tester results .............................. 46
Table 7. Starch pickup ................................................................................................ 46
Table 8. Z-direction strength and SCT strength ......................................................... 48
Table 9. SSA sizing effect .......................................................................................... 49
Figure 1. Pond type size press ...................................................................................... 4

Figure 2. Two-sided metered size press ....................................................................... 7
Figure 3. Amylose and amylopectin chemical structure (modified from Holik, 2013)9
Figure 4. Structure of internal sizing agents............................................................... 16
Figure 5. Sizing agent particle in emulsion ................................................................ 17
Figure 6. Structure of SMA (top) and SAE (bottom) ................................................. 19
Figure 7. Structure of an unsized European testliner ................................................. 21
Figure 8. Two-ply Fourdrinier/hybrid forming section schematic ............................. 23
Figure 9. Laboratory Mathis pond-type size press ..................................................... 30
Figure 10. All surface sizing test points on testliner 1 ............................................... 37
Figure 11. Sizing effect of SSA 31 on the different base papers ............................... 38
Figure 12. Sizing effect of SSA 30 on the different base papers ............................... 38
Figure 13. Size press pressure and 18% starch pickup (base paper TL + S).............. 40
Figure 14. Correlation between size pickup at 12% solids content and solids content
required for 40 kg/t pickup for different base papers. ................................................ 41
Figure 15. The effect of starch viscosity on surface size pickup ............................... 42
Figure 16. The effect of starch viscosity on Bendtsen porosity ................................. 43
Figure 17. Hg-porosimetry pore size and volume ...................................................... 45
Figure 18. DWR with 12% starch correlation to 12% starch pickup ......................... 47
Figure 19. Testliner 1 strength increase with increased starch amount ...................... 47
Figure 20. Gurley air permeability and Z-direction strength ..................................... 48
M. Lagus: Hydrophobic surface sizing of testliner VIII
Figure 21. SCT for 40 kg/t starch sized base paper depends on base paper SCT ...... 49
Figure 22. DWR 12% starch correlation with sizing effect ....................................... 50
Figure 23. t(MAX) and sizing effect .......................................................................... 50
Figure 24. Parameter W and sizing effect .................................................................. 51
Figure 25. Sizing effect and SSA particle size, no PAC ............................................ 52
Figure 26. Sizing effect and SSA particle size, with PAC ......................................... 52
Figure 27. Sizing effect and SSA film water contact angle, no PAC ......................... 53
Figure 28. Sizing effect and SSA film water contact angle, with PAC ..................... 53
M. Lagus: Hydrophobic surface sizing of testliner IX
List of abbreviations
AKD Alkyl ketene dimer

ASA Alkenyl succinic anhydride
CD Cross direction
COD Chemical oxygen demand
DWR Dynamic water retention
FWA Fluorescent whitening agent
KL Kraftliner
MD Machine direction
PAC Poly-aluminum chloride
PPS Parker print-surf
SAE Styrene acrylic ester
SCT Short-span compression test
SMA Styrene methacrylate
SSA Surface sizing agent
Tg Glass-transition temperature
TL Testliner
M. Lagus: Hydrophobic surface sizing of testliner 1
Introduction
The focus of this thesis is hydrophobization of testliner through surface sizing. It is a

relevant topic because corrugated boxes are low-cost, sustainable packaging materials
that are being used more and more because of the prevalence of online shopping.
Packaging materials are also gaining more traction in the paper industry as the market
share of graphical papers continues to decline. The use of recycled fiber-based testliner
is further rising compared to virgin-fiber based kraftliner because recycling of paper
and board has never been practiced in Europe to the degree it is today, with the
Americas and Asia following the trend (CEPI, 2016; Holik, 2013; ERPC, 2016).
Surface sizing, or applying starch and surface sizing agent (SSA) to the paper surface,
is an old process. The application on testliners is nevertheless complex because of the
large number of changing parameters in the hydrophobization. Previously, and in some
cases currently, hydrophobization of liners would be achieved through internal sizing,
i.e., adding the sizing agents to the wet end of the paper machine. However, especially
recycled testliner fibers can contain a wide range of paper chemistries that may
complicate the wet end internal sizing process, leading to deposits and costly
shutdowns. Surface sizing allows for a simplification of the wet end, prevention of the
uneconomical application of sizing agents to the inside of the paper is, and precise
fine-tuning of the surface properties (Bajpai, 2015; Zeng, 2013).
The exact mechanisms for hydrophobization in surface sizing are currently not well
understood, and laboratory studies seldom correlate well with practical on-machine
tests. Furthermore, there is no SSA that would provide the same level of
hydrophobization on all testliners. Through a better understanding of the surface sizing
process, wasteful SSA application amounts could be avoided, and SSA’s could be
developed more efficiently. A reliable laboratory method would facilitate SSA
development further and improve performance prediction possibilities.
In this thesis, the goal is to increase our understanding of the hydrophobization in

testliner surface sizing. In Chapters 1 through 4, the surface sizing process and relevant
materials are discussed from a literature point of view. The methods and materials used
in this study are presented in Chapters 5 and 6. The results of the thesis are covered in
Chapters 7 to 10, starting with observed differences in sizing effect and continuing to
the effect of surface sizing method, base paper and SSA parameters on sizing effect.
I LITERATURE REVIEW
The focus of this work is hydrophobization in surface sizing of testliner. In order to

introduce the surface sizing process, the following literature review covers surface size
application methods, starch, surface sizing agents (SSA), and testliner base paper.
Relevant theories of hydrophobization and water penetration are also introduced.
1. Surface size application method
The surface sizing operation is placed in the middle of the drying section of the paper
machine. A solution of starch, surface sizing agent and potential additives is applied
to the surface of the paper web, using either a pond type size press, metered film press,
or a novel method presented in the following chapter. The three surface size
ingredients and the application method can be varied, comprising the endless
individual configurations of each paper machine. Further, the pickup and efficiency of
the surface size is dependent on concentration (solids content), starch type,
temperature, viscosity, pH, sheet moisture, internal sizing and its development, and
machine speed (Hagiopol, Johnson, 2012).
1.1. Pond type size press
The pond type size press, or flooded size press, is the older and nowadays less used of
the two most common surface size application methods. The paper web passes
between two rotating cylinders with a pond of surface size between them (Figure 1).
Wet pickup in the pond type size press is usually 30-50% of the paper weight, i.e., 300
to 500 kg/t. Typically, the pond size press is run at approximately 10% solids,
corresponding to 30-50 kg/t dry pickup (Holik, 2013; Maurer, 2009; Gess, Rodriguez,
2005).
Figure 1. Pond type size press (Andersson, Ridgway & Gane, 2013)
Surface size is transferred to the paper through wetting and capillary transfer in the
pond, and pressure penetration in the nip. Wetting and capillary transfer are mainly
controlled by the formation and chemical properties of the paper. The formation
includes porosity, basis weight, density, surface smoothness, moisture content, and
temperature. Increased porosity increases the pickup, and a rougher surface raises
pickup when more liquid is metered through the nip. Chemical properties depend on
the raw material and the presence and development of internal sizing. The internal
sizing influences the fiber surface energy, limiting the capillary transfer of surface size,
thus shifting the transfer mechanism towards pressure penetration. Without internal
sizing, the starch pickup is mostly dependent on pore wall wetting. The low solids
content in the pond type size press means that the paper is rewetted considerably, and
the absence of internal sizing may lead to web breaks especially at slower speeds
(Maurer, 2009; Gane, 2001; Holik, 2013; Andersson, Ridgway & Gane, 2013; Gess,
Rodriguez, 2005).
Pressure penetration in the nip is hydrodynamic force-dependent. The hydrodynamic

force, and thus pickup, increases with softer, larger size press rolls, higher nip pressure,
and faster machine speed. When machine speed is increased the transfer mechanism
will shift from mainly pond interactions to more pressure penetration. During the
change, starch pickup will pass through a minimum before the hydrodynamic pressure
increases (Holik, 2013; Maurer, 2009).
Other than the pickup, the holdout of surface size should be considered. The added
surface sizing agents are designed to make the surface of the paper hydrophobic, a
higher level of holdout is often desired to prevent the wasteful application of SSA to
the middle of the liner. However, surface size starch contributes to the strength of the
paper, and holdout should be optimized for both strength development and
hydrophobization depending on the requirements. The holdout is generally increased
by lower paper porosity and higher molecular weight or viscosity of starch (Maurer,
2009; Gess, Rodriguez, 2005).
Problems and defects in pond size presses can occur because of web detachment from
the rolls and pond instability. Upon leaving the size press nip, the paper web must be
pulled apart from one of the size press rolls. To prevent defects, the starch viscosity,
roll diameter, web tension and web take-off angle should be optimized. At high
machine speeds, high turbulence may occur in the pond causing instability, e.g.
splashing. Stability problems can be alleviated with foils dipped into the pond on each
side, and by using bigger roll diameters (Maurer, 2009; Gane, 2001; Holik, 2013).
Pond size press pigmentation has been studied showing promising results of decreased
drying energy demand, reduction in fluorescent whitening agent (FWA) addition
amount, and better holdout at the surface due to higher solids content. However,
metered size press pigmentation has been found more efficient for fiber reduction
(Wygant, Lucidi & Kendrick, 2009).
1.2. Metered film press
The metered film press is the other of the two most common surface size application
methods and is widely used today. It was developed to reduce the splashing and
rewetting problems with the pond-type size press. Instead of passing the paper through
a pond where starch is picked up, a set amount of starch is metered on an applicator
roll and pressed into the paper (Maurer, 2009; Lehtinen et al., 2000). There are two
widely used methods of starch metering which will be presented next.
The gate roll film press

In the gate roll film press, the starch amount is metered by film splitting between three
rolls: the metering (= gate) roll, the transfer roll, and the applicator roll. The metering
roll is in contact with a starch pond, from where a film is picked up. The film is split
once in the nip between metering and transfer roll, and a second time between transfer
and applicator roll. The residual film is pressed onto the paper in a nip between the
applicator roll and the backing roll. If the starch is applied on both sides of the paper,
either a simultaneous or tandem configuration can be chosen. In a simultaneous gate
roll film press, the backing roll is the applicator roll for the other side. In a tandem gate
roll film press, the other side of the paper is treated in a separate station after the first
one (Maurer, 2009).
Pond instability and splashing are common problems with the gate roll film press. They
can be controlled by running the metering roll at slow speeds, or by using higher solids
content and higher viscosity starches. The higher solids content, in turn, allows paper
webs of a higher moisture content to be sized without risk of web breakage, reducing
drying energy and time both before and after the film press. These advantages can
enable higher machine speeds, further increasing the machine efficiency (Maurer,
2009).
Rod metered film press

In a rod metered film press (Figure 2) surface size starch is applied to the metering roll
with a short dwell coater or a pond, and a rod meters the starch to the preferred film
thickness. Surface size pickup is controlled by machine speed, surface size viscosity,
film split between paper and roll, and roll hardness. Film split between metering rod
and roll, rod profiling, and rod pressure are also relevant (Holik, 2013; Maurer, 2009).
Figure 2. Two-sided metered size press (Forsström, 2003)
The method has the same advantages over pond type size presses as the gate roll film
press, i.e., the higher solids content of the starch allows a reduction in drying energy.
Solids contents up to 30% have been studied and shown to be runnable, giving a higher
holdout of starch, resulting in decreased paper porosity, decreased oil absorption,
decreased internal strength of paper, and increased surface strength and bending
stiffness. However, the effect of higher viscosity starch on the SSA efficiency has not
been evaluated (Lipponen, 2005).
Sources of problems with metered film presses are starch viscosity and film splitting
behavior. Low viscosity starch may lead to fluid spraying in the metering. In the film
splits between rolls and between roll and paper, misting and orange peel pattern can
occur. For the roll-paper split, the problem may be prevented by bringing the film-
splitting closer to the roll (Maurer, 2009).
Metered film press pigmentation has been studied with promising results. Fiber
content could be reduced in the final product for savings in fiber price, and the size
press could be used as a precoating device. However, adding pigments to the surface
size adds viscoelastic behavior to the “sizing color”, and the formulations need
optimized rheology and water retention for runnability (Gane, 2001; Wygant,
Kendrick & Walter, 2008).
1.3. Novel and alternative methods
New methods for surface size application that have been introduced are calender water
boxes, spray application, and foam application. In the calender water boxes, the starch
is usually diluted and applied at a low level with lubricants. Surface strength of the
paper is increased, and fiber swelling after coating is limited (Maurer, 2009). Spray
application may become more important in the future, with possible benefits including
higher starch solids contents, no wearing parts and easy control of surface size amount
(Holik, 2013; Valmet, 2017). Adding foaming agents to the surface size and applying
it as a foam keeps the surface size efficiently on the surface of the paper. The
advantages could be lower coat weights needed and higher solids contents, but the
foaming agent may also hinder the performance of the SSA (Sievänen, 2010).
2. Surface size
The surface size applied to the paper consists of starch, surface sizing agent (SSA) and
potential additives. The SSA decreases water absorbency, the starch increases the
strength of the paper, and together they improve printability by controlling ink
penetration (Zeng, 2013). In this work, hydrophobization is the focus point. Strength
is, however, a far too important part of the surface sizing result to be ignored, and thus
the following chapter will cover both starch and surface sizing agents, other additives
will be introduced briefly.
2.1. Starch
Starch is a naturally occurring biopolymer which is present in a variety of plants. For

use in the paper industry, the main raw materials are corn, potato, wheat and tapioca.
In the plants, starch exists as semi-crystalline granules consisting of two polymers of
D-glucose units, amylose (20-30%) and amylopectin (70-80%), both built up with α-
1,4-bonds (Figure 3). Amylose is the smaller of the two, despite being present at only
20-30 mass-% there are about 150 times as many linear amylose molecules as larger
branched amylopectin molecules. The ratio of the two constituents, branch points in
amylopectin, and granule size of the starch depend on the source (Holik, 2013).
Figure 3. Amylose and amylopectin chemical structure (modified from Holik, 2013)
2.1.1. Starch in the paper industry
In the paper machine, starch can be added to the wet end, sprayed on the forming wire,
and added as surface size. It is by weight the third most used material after cellulose
fibers and mineral pigments in papermaking and paper conversion (Maurer, 2009).
Wet end starch

Wet end starch is typically added at levels up to 1.5% of cationic starch or up to 5%
of starch with a cationic polymer. Higher addition levels can disrupt the charge balance
in the wet end. The strength of the paper is improved when the free hydroxy groups in
the starch interact with fiber surface cellulose, participating in hydrogen bonds.
Retention and formation are improved when the gel-like starch network encapsulates
particles and fines. Cationic starch is retained efficiently due to ionic interactions with
the anionic fibers, while anionic or native starch needs to be used with a cationic
polymer, like polyvinylamine, which improves retention by forming aggregates or
through precipitation. Anionic starch can also be used in conjunction with cationic
starch to stabilize the charge balance (Maurer, 2009; Holik, 2013).
Spray starch
Spray starch is sometimes added to the paper web on the forming wire at addition
levels of up to 20% of the dry product. The starch is usually unmodified and uncooked;
it is assumed that cooking and gelling take place throughout the drying process.
Benefits of spray starch include retention where it is sprayed (surface or between
layers), improved bonding between layers, lower manufacturing costs, and increased
strength. However, when making recycled linerboard at high machine speeds the
starch may not have enough time to penetrate into the sheet and gelatinize, causing
plugged wires and accumulation of starch in discharges (Maurer, 2009; Holik, 2013;
Ryu, Lee, 2007).
Surface size starch

Surface size starch is added to the paper at the size press or film press for pick up levels
of 30-70 kg dry starch per ton of paper, usually about 40 kg/t. The starch binds large
vessels and loose fibers with hydrogen bonds, increasing the strength of the paper up
to 60%. Stiffness, dimensional stability, and air leak density are also improved. Native
starch is not suitable due to high viscosity and tendency to retrograde; the starch must
be modified or converted either by the supplier or at the paper mill before use (Zeng,
2013; Maurer, 2009; Holik, 2013).
2.1.2. Dispersion and conversion
Native starch is rarely used in surface sizing because of the high molecular weight and
corresponding high viscosity. It is either converted, modified, or both to control the
viscosity, charge and bonding ability. Starch is usually supplied dry in package or bulk
container to avoid transporting water. Dispersion and conversion are done on-site at
the paper mill. Common conversion methods include enzymatic conversion, thermal
conversion, thermos-chemical conversion and acid hydrolysis. Chemical modification
(usually oxidation, hydroxyethylation or cationization) can be done either at the
supplier or the paper mill (Holik, 2013; Gane, 2001; Maurer, 2009).
Dispersion
Dispersion is carried out by heating the starch during agitation for a set amount of time,
either continuously or in a batch process. The starch granules swell, and the semi-
crystalline structure breaks up, forming a gel-like network. Minimum requirements are
20 minutes at a temperature of 95 ºC, but granular fractions will persist in the starch
suspension. Jet cooking with pressurized steam at higher temperatures (120-140 ºC)
disperses most residues. Excess steam and pressurized release of starch can increase
dispersion efficiency further, lowering the viscosity. Jet cooking is especially popular
with pre-modified starches (Gess, Rodriguez, 2005; Maurer, 2009; Holik, 2013).
Conversion
Conversion lowers the molecular weight and viscosity of the starch to ease the
application. Thermal conversion happens during efficient thermal dispersion or jet
cooking, heat and shear disperse the starch for complete hydration and lower viscosity.
Thermo-chemical conversion combines thermal conversion with chemical

modification (Holik, 2013; Gane, 2001; Gess, Rodriguez, 2005).
In enzymatic conversion, enzymes are added to the starch dispersion in a batch or

continuous process. The enzymes break down the starch under controlled pH, agitation
and heating until the desired viscosity is achieved, then the temperature is increased
rapidly to kill the enzyme. The method is simple and inexpensive, but reproducibility
is low which can lead to variations in viscosity (Gess, Rodriguez, 2005; Maurer, 2009).
Acid hydrolyzed starch is depolymerized with acids, e.g. hydrochloric acid. The
method is inexpensive, but the resulting depolymerized starch has an unstable viscosity
and is sensitive to retrogradation, and therefore temperature and pH control is required
(Maurer, 2009).
2.1.3. Retrogradation
During cooling and storage, the semi-crystalline structure of starch may recover in a
process called retrogradation. Amylose aggregates form particles, while amylopectin
forms a gel where linear parts of the molecules align. The retrogradation of amylose
can start if dilute starch is kept at a temperature of 67-90 ºC, and amylopectin can start
thickening below 55 ºC. Retrogradation is facilitated by starch molecular weights
between 6500 and 160 000, slightly acidic pH (6.5), high amylose content,
destabilizing ions (for example aluminum and calcium), contamination, well-dispersed
starch and hydrolyzed starch (Maurer, 2009).
To avoid retrogradation, the temperature of dispersed (cooked) starch should be kept

at 60-65 ºC or above 95 ºC. Starch modification by oxidation or esterification can
reduce retrogradation, and preservatives can be used to prevent microbial action that
lowers the pH. Redispersion of amylose particles requires breaking of bonds at high
temperatures (115-120 ºC), whereas for amylopectin a temperature of 55 ºC is
sufficient to dissolve the gel (Maurer, 2009).
2.1.4. Modification
Chemical modification of starch is increasingly popular, in 2009 about two-thirds of

the starch used in size press application was modified (Maurer, 2009). The higher cost
of modified starch compared to native on-site converted starch probably limits the use
in testliner production, however, the benefits may outweigh the cost. Advantages of
modified starches include improved stability, lower viscosity, retrogradation
resistance, improved film formation and binding properties. The most common
methods for modification are oxidation (oxidized starch) and derivatization (ethylated
starch and cationic starch). Cationic starch has superior retaining and fixing properties,
while anionic starches are compatible with a range of additives, including FWAs
(Maurer, 2009; Jonhed, 2006).
Oxidized (anionic) starch

Oxidized starch is produced by reacting starch with sodium hypochlorite or peroxide
(or less commonly with ozone or sodium periodate). The hydroxyl groups in starch are
transformed to carboxyl groups, and starch is depolymerized through ring-opening.
Starch properties (e.g. viscosity) can be controlled by adjusting the reaction conditions
or by choice of raw material (different amylose to amylopectin ratio) (Gess, Rodriguez,
2005; Vanier et al., 2017; Jonhed, 2006).
Oxidized starch has improved water holding, film formation, film strength, and film
integrity properties compared to native starch, giving the final paper high surface
strength and improved ink holdout. The new bulky carboxyl groups hinder
retrogradation sterically, reducing the gelatinization temperature. In surface sizing,
oxidized starches have been shown to penetrate deep in the paper structure and
decrease opacity (Gess, Rodriguez, 2005; Jonhed, 2006; Lee et al., 2002).
The anionic character of the starch influences the interaction with other papermaking
substances. Broke recycling brings the anionic starch to the wet end, where it prevents
pigment and cellulose agglomeration through efficient dispersion, and reduces
retention. This leads to anionic trash in the effluent water and increased chemical
oxygen demand (COD). To combat the retention problem, slightly oxidized starch, i.e.,
bleached starch, has been developed for surface sizing. Bleached starch is produced
with a low degree of oxidation (carboxyl content less than 0,3%) and partial
crosslinking. However, the dispersing properties of anionic starch still worsen
retention and, therefore, the use of anionic starch in surface sizing is generally
declining (Lee et al., 2002; Jonhed, 2006; Gess, Rodriguez, 2005; Maurer, 2009).
Cationic starch
Cationic starch is produced by a nucleophilic substitution reaction with tertiary or
quaternary amines. The reaction with tertiary amines requires acidic pH, but
quaternary amines work across acid to alkaline pH values. Hydroxyethylated cationic
starch is produced for specialty products (Maurer, 2009).
In surface sizing, cationic starch has been found to increase the holdout of the surface
size on the paper surface. The higher holdout, in turn, improves opacity, brightness,
print gloss, and ink print density (Shirazi et al., 2005; Lee et al., 2002).
The primary advantage of cationic starch over anionic and nonionic starch is improved
binding and retention ability. The cationic starch interacts with the anionic fibers
ionically, which produces stronger binding compared to hydrogen bonds. During
broke processing, the cationic starch binds fillers and fines, reducing the suspended
solids and COD of the wastewater significantly. Because of environmental regulations
and discharge limits, the use of cationic starch in surface sizing is rapidly growing (Lee
et al., 2002; Maurer, 2009; Jonhed, 2006).
2.2. Sizing agent
The purpose of sizing is to make paper water resistant to some degree. In general, the
mechanism is to coat the hydrophilic fibers with hydrophobic sizing agents. The sizing
agents can be either amphipathic molecules, having one hydrophilic and one
hydrophobic part, or highly hydrophobic molecules. The sizing agents can be applied
to the paper either in internal sizing, in surface sizing, or in a combination of the two.
In internal sizing, the sizing agent is added to the paper stock at the wet end of the
paper machine, while surface sizing agents are added to the starch at the size press.
The sizing agents can be characterized according to application method (internal or
surface) and functioning principle (reactive or non-reactive). Reactive sizing agents
form a chemical or physical bond with the cellulose fibers during drying, characterized
by a developing hydrophobicity. Non-reactive sizing agents need a high level of
hydrophobicity to make the paper moisture resistant without the bond to the fibers, but
the hydrophobicity effect is instant (Holik, 2013; Hagiopol, Johnson, 2012).
2.2.1. Internal sizing agents
The most important property for internal sizing agents is the retention. Otherwise, the
sizing agent will flow out with the effluents. Reactive internal sizing agents form a
bond with cellulose and are thus readily retained. Cationic non-reactive sizing agents
are attracted to the anionic fibers because of their charge, but anionic or non-ionic non-
reactive internal sizing agents need a retention aid. Other than retention, the
hydrophobic character, molecular weight, molecular orientation, glass-transition
temperature (Tg), paper surface friction and pH compatibility are all essential factors
for the hydrophobization process (Hagiopol, Johnson, 2012).
The three most popular internal sizing agents are all reactive: rosin, AKD (alkyl ketene
dimer) and ASA (alkenyl succinic anhydride) (Figure 4) (Hagiopol, Johnson, 2012).
Figure 4. Structure of internal sizing agents. From left to right: fortified rosin, AKD and ASA
(modified from Holik, 2013)
One of the oldest sizing agents, rosin is extracted from resin that occurs naturally in
trees. Extraction is carried out from the black liquor in the pulping process. Rosin
requires an acidic pH to be effective, limiting the use of calcium carbonate pigments
because of dissolution. Rosin is anionic and needs alum (aluminum sulfate, Al2(SO4)3)
to bind to the fibers. Later, rosin sizing agents have been modified, reacted further
(using either the double bond or the carboxyl group), and saponified to produce a range
of rosin-based sizing agents with increased hydrophobicity (e.g., fortified rosin)
(Hagiopol, Johnson, 2012; Zeng, 2013).
AKD and ASA were invented after rosin sizing for alkaline and neutral papermaking.
They quickly became popular because the pH allowed calcium carbonate pigment to
be used. AKD and ASA are highly reactive with cellulose fibers, but also with the
surrounding water, which can quickly “kill” the sizing agent through hydrolysis. AKD
creates sticky by-products in the hydrolysis reaction, lowering machine efficiency. The
reactive sizing agents are nonpolar, and thus need emulsifiers and stabilizers to be
water soluble (Figure 5). Emulsion stability issues are common, especially ASA
hydrolyzes fast and must be dispersed right before use. AKD size needs a high
temperature curing at drying and grade-dependent curing times to impart
hydrophobicity. ASA size is more reactive than AKD and does not require curing, but
elevated stock temperatures and higher pH increase efficiency (Hagiopol, Johnson,
2012; Holik, 2013; Zeng, 2013).
Figure 5. Sizing agent particle in emulsion (Holik, 2013)
There are many other polymeric internal sizing agents, and new ones are being
researched continuously to combat the problems of dispersion stability, retention and
hydrolysis reactions. Common synthesis methods for non-reactive sizing agents are to
add hydrophobic tails or cationic charges to existing polymers. Reactive sizing agents
can be synthesized utilizing cellulose-reactive groups, for example, epoxy groups,
aziridines, carbamoyl chloride groups, succinic anhydride groups, and ketene dimer
groups (Hagiopol, Johnson, 2012; Gess, Rodriguez, 2005).
2.2.2. Surface sizing agents (SSA)
Surface sizing agents (SSA) are applied with the starch on the surface of paper. Thus,
wasteful application to the inside of the paper sheet is prevented. The paper machine
wet end chemistry is also simplified because of the exclusion of sizing agents. Surface
sizing agents are usually non-reactive, i.e., they do not participate in reactions that alter
their molecular structure when applied to the paper. To be effective, SSAs need a high
degree of hydrophobicity and a high coverage on the paper sheet. The hydrophobicity
needs to be balanced with hydrophilic character to enable the mixing of SSA with
starch, and thus the SSA particles are amphipathic (they contain one hydrophobic and
one hydrophilic part). Stabilizers are used to prevent agglomeration of the amphipathic
particles, and to prevent foaming of the surface size, both of which decrease coverage
on the paper (Hagiopol, Johnson, 2012; Holik, 2013).
A high degree of coverage is essential to provide hydrophobicity. Adequate coverage

is assured by adjusting the SSA Tg, minimum film-formation temperature (MFFT) and
the paper machine drying conditions to allow the SSA particles to spread on paper
surface. A film is usually not formed, instead a lattice of hydrophobic SSA particles
and hydrophilic starch. The efficiency can further be improved by increasing the
hydrophobic character of the SSA, decreasing particle size (more coverage for same
SSA dosage) and using different types and amounts of stabilizers to prevent
agglomeration in the surface size and excessive penetration in the paper sheet. For
example, zirconium salts have been used as complexing agents to prevent penetration
of the SSA (Pandian, van Calcar & Wolff, 1994; Hagiopol, Johnson, 2012).
There is a multitude of SSA chemistries available for virtually all types of

papermaking conditions. Some internal sizing agents can also be applied as SSA. For
example, while AKD has been used in surface sizing, it is only stable for a couple of
hours at starch temperatures (about 60 °C). The most common SSAs are polymeric
SSAs, usually with a styrene base to provide hydrophobicity. While the polymeric
SSAs are mostly used in surface sizing, they may also be used as internal sizing agents
or in combination with traditional internal sizing chemistry (Gess, Rodriguez, 2005;
Holik, 2013).
A polymeric SSA is commonly synthesized by emulsion copolymerization, usually

through a free radical polymerization that starts with homogeneous nucleation.
Essential factors for the synthesis are reaching the target Tg, control of the particle size
distribution, and control of the stabilizer type and concentration. Glass-transition
temperature can be estimated theoretically taking into account the ratio of monomers
and the monomer’s individual Tg. Particle size can be controlled by adjusting the
number of nuclei at the beginning of the reaction and ensuring their protection. The
polymer composition is influenced by the process type and the monomer ratio. The
structure is influenced by Tg and composition (homogeneous or, e.g., core-shell). The
chemical composition distribution can be affected by reaction conversion, monomer
ratio, and monomer reactivity. The copolymer will orientate so that the more water-
soluble co-monomer is oriented towards the particle surface (Hagiopol, Johnson,
2012).
The two main, widely used groups of polymeric SSA are styrene-maleic anhydride
copolymers (SMA) and styrene-acrylic ester copolymers (SAE) (Figure 6) (Hagiopol,
Johnson, 2012; Holik, 2013).
Figure 6. Structure of SMA (top) and SAE (bottom) (modified from Holik, 2013)
2.3. Other surface size additives
Other than SSA, prevalent additives in the surface size starch are preservatives for the
starch, promoters, defoamers, fillers, and pigments, FWAs, co-binders and other
functional additives, for example, to improve printability (Holik, 2013).
Promoters in surface sizing are usually aluminum compounds, e.g., poly aluminum
chloride (PAC). Promoters can be used with rosin sizing or ASA sizing to fixate the
molecules to the fibers, but SMA SSA hydrophobization performance has also been
reported to increase. Aluminum has a high cationic charge which gives it high affinity
toward the fibers, as well as the anionic sizing agents. Fines and fillers will also be
retained with aluminum, preventing SSA particles absorption only in the fillers and
fines (Holik, 2013).
2.4. Improving print quality with surface sizing
Pigment addition to the size press has been shown to increase paper smoothness,
improve optical properties, increase density and stiffness, and improve printability.
The pigment can be used to replace some of the fiber in the paper, the size press can
act as a precoater, and FWA demand can be reduced (Wygant, Kendrick & Walter,
2008).
When pigments are added, the surface size solids content increases considerably, up
to 60%. A high pigment loading may require additional binders and rheology modifiers
in addition to starch to improve runnability (Maurer, 2009).
Fluorescent whitening agents (FWAs) are usually added at the size press to prevent
wasteful application in the bulk of the paper. The FWAs improve opacity and
whiteness of the paper (Holik, 2013).
3. Testliner base paper
The following chapter introduces testliners and the paper machines used to produce
liner in general.
3.1. Testliner
Testliners are produced in grammages of around 70-450 g/m2 globally (in Europe 70-
130 g/m2) with maximum machine speeds above 1500 m/min. The speeds have been
increased through improved designs to achieve high annual production for low basis
weight paper grades. The paper machines for liner production usually have one or two
layer gap formers or one to multi-ply Fourdriniers. The multi-layer design enables
production of white top liners or high-performance liners with kraft pulp additions in
the top layer and also works to decrease sheet two-sidedness (Holik, 2013).
Figure 7. Structure of an unsized European testliner

3.2. Raw material
Testliner paper grades are entirely based upon recycled, secondary fibers giving a
structure shown in Figure 7. The fibers are usually sourced from old corrugated
containers, but recycled newsprint and graphical paper may also occur. Fibers are
reused on average 3.5 times in Europe. (ERPC, 2016) For each reuse, the fiber
properties deteriorate, thereby making the paper weaker (Holik, 2013).
The strength decrease is a combination of decreased fiber strength and decreased

interfiber bonding. Recycling treatment seems to have a higher impact on interfiber
bonding through a decrease in surface bonding potential, and the first recycling cycle
results in the largest decrease in paper strength. Repeated drying and rewatering also
impact the fibers in the paper, which poses a problem for samples studied in laboratory
scale (Nazhad et al., 2008; Khantayanuwong, 2002).
Recycled secondary fibers are not as absorbent as virgin fibers. Fiber treatment, e.g.,
refining and sorting, is the primary way to increase the absorbency of the paper for
more efficient starch pickup in the paper. Internal sizing will also impact how the
surface size penetrates the paper or stays at the surface (Holik, 2013).
More fines in paper furnish increase the density of the paper and have an impact on
porosity. Fines naturally fill up pores as well as pull fibers closer together decreasing
the thickness of the sheet. Fibrils contain a tenfold surface area compared to flake-like
fines. The presence of fines on the paper surface can be significant for SSA
performance (Sirviö, Nurminen, 2004).
3.3. Papermaking process
Recycled paper stock processing

Because of the already subpar properties, it is crucial to process the recovered paper
stock properly before it ends up in the paper machine chest. Processing consists of
repulping, separation, treatment of fibers, and treatment of contaminants. Repulping is
an energy-intensive mechanical process of disintegrating the stock to individual fibers.

These fibers then need to be separated from contaminants, for example, studs from
used boxboard. Separation techniques include screens, hydrocyclones, and flotation
cells. Then, the clean fibers are treated with refiners and in some cases bleach, and
finally, chemicals are added to reduce or negate the effect of residual contaminants.
The processed stock is fed to the paper machine headbox through the approach flow
system, where all components of the stock are mixed, metered and supplied
continuously. From the headbox, the suspension is distributed onto the wire, where the
formation of the paper web takes place (Höke, Höke & Schabel, 2010; Holik, 2013).
Figure 8. Two-ply Fourdrinier/hybrid forming section schematic (MultiLayer machine,Voith)
Formation
Different wire configurations impact the final paper structure, which further will have
an impact on surface size penetration and performance. For testliners, standard
configurations are the Fourdrinier, the gap former and the hybrid former. The
Fourdrinier is a long planar wire where almost all dewatering happens on the bottom
side, leading to a pronouncedly two-sided paper with higher compaction of fibers on
the wire side, leaving fines and filler at the top. However, shoeboxes can disturb this
two-sidedness with centrifugal force. A gap former wire means that the fiber
suspension is a jet delivered between two wires and the dewatering is equal to both
sides. The paper will have high symmetry with fillers and fines in the middle. However,
pressure pulses can be used to mix the distribution. A hybrid former is a Fourdrinier
wire with a second wire loop on top. The two wires are adjoined after some dewatering
has already taken place before the press section (Figure 8). This configuration gives
higher Z-direction symmetry than the Fourdrinier and also increases the drainage
efficiency (Forsström, 2003; Holik, 2013).
Pressing, drying and calendering

After formation, the paper enters the press section where the sheet passes through
several press nips where more water is pressed out mechanically. Paper is densified,
porosity decreases, fillers and fines are carried to the surface with the water and paper
roughness is affected. The surface densification affects liquid (and starch) absorption
properties (Holik, 2013).
Dryers evaporate the rest of the water by heat, and after this, the surface size is usually
applied to the paper web. Coating takes place after surface sizing if it is required, and
finally the paper may be calendered to increase its smoothness (Holik, 2013).
4. Hydrophobization and water penetration theory
The goal of hydrophobic surface sizing is to increase the water repellence of the paper
surface, i.e. to control water interaction with the paper surface. The interaction depends
on the fiber surface chemistry, or interfacial energies (γ), and on the paper
microstructure (Holik, 2013). Paper microstructure is often modelled as a collection of
capillaries, although in reality the situation is far more complex.
Droplet contact angles (θ) indicate whether the paper is easily wettable (θ < 90º,
cos θ > 0) or liquid repellent (θ > 90º, cos θ < 0). The contact angle is a result of
interfacial energies according to the Young’s equation:
𝛾𝑝𝑎𝑝𝑒𝑟−𝑎𝑖𝑟 − 𝛾𝑝𝑎𝑝𝑒𝑟−𝑙𝑖𝑞𝑢𝑖𝑑
cos 𝜃 =
𝛾𝑙𝑖𝑞𝑢𝑖𝑑−𝑎𝑖𝑟
When the interfacial energy between the air and the paper is lower than the interfacial
energy between paper and water, it is energetically favorable for the paper to have an
interface with air rather than with water, and the paper is hydrophobic. The sizing
agent should thus either decrease the surface free energy (𝛾𝑝𝑎𝑝𝑒𝑟−𝑎𝑖𝑟 ) or increase the
interfacial energy (𝛾𝑝𝑎𝑝𝑒𝑟−𝑙𝑖𝑞𝑢𝑖𝑑 ) (Holik, 2013).
The influence of paper microstructure for wetting driven capillary absorption can be
described by the simplified Washburn equation
1
ℎ≈ × cos 𝜃,
𝑟
where r is the radius of the capillaries, h is the water penetration depth into the paper
and θ is the contact angle. According to the simplified Washburn equation, the water
penetrates deeper into the structure the smaller the capillary radius, if the paper is easily
wettable (θ < 90º, cos θ > 0) (Holik, 2013).
In surface sizing, especially when carried out by conventional size presses, the starch
is forced into the paper by the nip pressure during application. This pressure
penetration is controlled not by the surface energetics but rather by the driving pressure,
the compressed pore structure of the paper (permeability) and the starch viscosity.
II EXPERIMENTAL PART
The objective of this thesis was to determine what properties of the base paper, the
sizing agent, or the sizing process have an impact on the sizing effect, measured as
paper surface hydrophobicity (Cobb 60s). The study consisted of laboratory surface
sizing experiments on different base papers, paper analysis, surface size analysis, and
sizing agent analysis. The existing laboratory scale sizing method was first studied and
optimized to obtain comparable results. Then, a variety of surface sizing agents were
tested on different base papers.
In this section, all relevant materials, equipment, and methods used in this work are
presented.
5. Materials
Eight base paper mill samples were used in the study: five unsized testliners (TL 1,
TL 2, TL 3, TL 4, and TL 5), one testliner with diluted surface size starch (TL + S)
one unsized kraftliner (KL) and one copy paper grade (C). The base papers were
analyzed, and the results are presented in Chapter 9: Effect of base paper properties.
Starch 1 and Starch 2 were part of Roquette’s “Stabilys A” product line of low
viscosity modified potato starches. Starch 2 had a slightly higher viscosity. The
structure of the starches simulates enzymatically converted starch.
Both starches were available in powder form and cooked with a jet cooker at 134 ºC
at a set level of solids content.
45 different surface sizing agents (SSAs) named SSA 1-45 were tested, including
development products and references of different origin. All SSAs were tested on
testliner 1, SSA 30 and SSA 31 were also tested on the other base papers. Most of the
SSAs were tested both with and without the addition of poly aluminum chloride (PAC),
denoted SSA X and SSA X + PAC. The surface sizing agents were analyzed, or previous
analysis data was used, to compare performance with SSA properties. The results are
presented in Chapter 10: Effect of SSA properties on sizing effect.
6. Methods
6.1. Laboratory surface sizing method
Part of the objective of this master’s thesis was to refine an existing laboratory surface
sizing method and eliminate possible variables to obtain repeatable and comparable
results. The aim was to achieve a constant addition level of surface size to the base
paper (35-45 kg/t) and a repeatable hydrophobization effect of sizing agent dosage.
The following section will describe how the surface sizing method was tested, which
parameters were controlled and the final laboratory surface sizing method.
6.1.1. Surface sizing equipment
Starch was jet-cooked with variations in cooking speed, slurry solids content and
temperature. The cooked starch’s solids content was measured with a CEM 5
microwave oven. The surface size was prepared by mixing cooked starch with hot
water and surface sizing agent with a laboratory magnetic mixer. Temperature,
agitation time and solids content were controlled. The surface size was analyzed
(solids content, pH, temperature, Brookfield 100 rpm) after preparation. The surface
size was applied to the base papers using a Mathis pond-type size press (Figure 9) with
soft rolls (70 shore D). The laboratory size press has adjustable speed (2-16 m/min),
adjustable pressure (0-100 on a scale, the actual pressure was not known) between the
rolls and the rolls can be heated with hot water before pouring the surface size into the
pond. After surface size was added into the pond, four base paper sheets were run
through the sizing solution and dried. The surface sized paper sheets were dried either
in an oven or with an infrared dryer with variations in temperature and drying time.
Figure 9. Laboratory Mathis pond-type size press
6.1.2. Experimental layout for testing the method
To achieve a constant sizing pickup and consistent sizing effect in the size press, the
parameters that can influence the process were first defined. Then, the parameters were
set to benchmark values, which enabled testing of the effect of each one with a number
of experiments on the same base paper (testliner with starch, TL + S) by adjusting one
parameter at the time. The benchmark and tested settings are presented in
Table . For the parameter test, a sizing solution of starch and SSA 1, an anionic, widely
used reference surface sizing agent, was used.
Table 1. Benchmark settings for surface sizing parameter testing
Parameter Benchmark Tested parameter values

Paper moisture content 6-7% 7-8%, 6-7%, oven dried (ca 0%)
Size press speed 4 m/min 2,4, 8 m/min
Size press pressure 20 ca 0, 5, 10, 15, 20, 40
Pond starch volume 250 ml 20, 50, 75, 100, 150, 200, 250, 300 ml
Oven drying temperature 105 °C 60, 80, 105, 125, 145, 180 °C
Oven drying time 3 min 1, 3, 5 min
Surface size solids content 18% 10, 15, 16, 17, 18%
Surface size temperature 60 °C 20, 40, 60, 65, 70, 75 °C
6.1.3. Adjusting the laboratory surface sizing parameters
Paper moisture content

The effect of base paper moisture content was investigated with an exicator cabinet
using saturated sodium chloride solution instead of silica to produce a higher base
paper moisture content than that of the standard conditioned sheets (7-8% and 6-7%,
respectively). Oven-dried base paper was used to investigate the effect of extremely
low moisture content.
Size press speed

Speed was controlled by a wheel on the laboratory size press with a range of 2 m/min
to 16 m/min; the tested values were 2, 4 and 8 m/min.
Size press pressure

The pressure was similarly controllable on the machine with a lever going from 0 to
100. However, the working pressure at lever position 100 was not known. The tested
pressure lever positions were ca. 0 (barely holding the surface size in the pond), 5, 10,
15, 20 and 40.
Starch volume in pond

The effect of starch volume in the pond was tested by using starch volumes of 20, 50,
75, 100, 150, 200 and 300 ml.
Drying conditions
Drying method, temperature and drying time were investigated both with oven and
infrared drying. The moisture content of the sized paper after drying and after
conditioning was measured for some test points. For oven drying, temperatures of 60,
80, 105, 125, 145 and 180 °C were tested with different drying times resulting in
completely dry (moisture content < 1%) and even burnt (t > 105 °C) samples. The
infrared dryer was tested with drying times of 10, 15 and 20 s with the sized paper in
the middle or at the bottom of the infrared dryer rack.
Surface size solids content

For base paper 1, surface size solids contents of 10, 15, 16, 17 and 18% were tested.
For the other base papers, surface size solids content of 12% was used as a guide to
determine optimal solids content for correct pick-up.
Surface size viscosity

Viscosity was altered by using Starch 1 as baseline and testing Starch 2, which had a
slightly higher viscosity but identical chemical structure and properties. Both starches
were jet-cooked and diluted to solids contents appropriate for the base papers studied,
and the highly effective SSA 1 was added.
Surface size temperature

Temperatures tested were ambient (c. 20 °C), 40, 60, 65, 70 and 75 °C. Higher
temperatures would have damaged the size press rolls.
6.1.4. Surface sizing method used for hydrophobicity tests
Results of the parameter testing can be found in Chapter 8: Effect of surface sizing
method parameters. The surface sizing method used for hydrophobicity tests in this
study is presented in Table 2.
Table 2. The surface sizing method used in further experiments
Parameter Value
Paper humidity 6-7%
Size press speed 4 m/s
Size press pressure 5
Starch volume in pond 150 ml
Oven drying time (105 °C) 3 min
Surface size solids base paper dependent
Surface size temperature 60 °C
6.2. Experimental layout for hydrophobization efficiency testing
To study differences in SSA hydrophobization efficiency, 45 different surface sizing

agents were tested on testliner 1. SSA 30 and SSA 31 were also tested on the other
base papers to shed light on differences in hydrophobization efficiency between the
base papers. In every test, the surface size recipes were formulated for a 40 kg/t pickup
with x kg/t starch, y kg/t SSA and z kg/t PAC, so that the sum of x, y, and z was 40.
The dosing of SSA’s tested on testliner 1 was 1, 2, and 4 kg/t SSA. Other base papers
were not as receptive of SSA hydrophobization, and a testing regimen of 1, 3 and 6 kg/t
was used for these base papers. When PAC was used with the sizing agent, it was
added at a constant addition amount of 2 kg/t regardless of SSA dosage. Because the
surface size is applied to both sides of the paper in the pond size press, dosing amounts
should not be considered to reflect the actual case in metered film press surface sizing,
when the SSA is applied on one side only.
6.3. Paper analysis
6.3.1. Hydrophobicity
Cobb 60 s tests were performed according to standard ISO 535 (TAPPI T 441) with
an L&W Cobb Sizing Tester. A 10 cm2 area of the paper was exposed to a 1 cm high
water pillar for 45 seconds. The water was poured off and at the full testing time (60 s)
paper was sandwiched between blotting sheets and rolled with a 10 kg roller. Weight
increase was used to calculate water absorption in g/m2. Two tests were made for each
test point and results were averaged.
Contact angles with water of sized papers (and surface sizing agent films) were
measured with Fotocomp DropPrint software. Ten drops of ion-exchanged water were
dropped from a height of 5 mm on the specimen and photographed at 0,5 s. The
obtained photographs were processed and calculated with the software, and contact
angles were averaged.
An ultrasonic dynamic penetration tester (Emtec PDA.C 02 Module MST Standard)

was used to study base paper and sized paper water penetration. Samples were cut and
fastened on a sample holder with two-sided tape. The device immersed the sample in
ion-exchanged water and immediately an ultrasound signal was passed through the
sample, and the received signal was recorded. During wetting, the air was displaced
with water and the received ultrasound signal changed as a result of different scattering
and absorption mechanisms. Measurement data was processed to extract the following
calculated parameters: intensity at 5 seconds I(5 s), time at 50% intensity t(50), time
at 95% intensity t95, time at maximum intensity tMAX and absorption value A30.
PDA A30 results have been shown to correlate with Cobb 60s values (Lopes et al.,
2010).
6.3.2. Structure and porosity
Porosity of the samples was measured with PASCAL 140/440 mercury porosimeter
at Åbo Akademi University. In mercury porosimetry, the pore structure of a sample
can be determined by the pore volume displaced by mercury. A sample is put in a
vacuum chamber and the chamber is filled with mercury while pressure and volume
are monitored. The resulting data can be used to calculate pore volume and pore
diameters of the sample. Thus, results on both the total porosity and the pore size
distribution are obtained.
Gurley air permeability was measured with an L&W air permeance tester. A sample
was put in the testing device where the time required for a volume of air to flow
through the sample is recorded. Results are given in seconds for flow of 1 liter of air.
Porosity and smoothness were measured with a Bendtsen type tester. The Bendtsen
tester measures air leakage both through the sample (porosity) and between the
measuring head on the sample and the sample surface (smoothness). Results are given
in ml/min.
Roughness/smoothness was also tested with Parker Print-surf (PPS) equipment. The
PPS measurement is based on the resistance to airflow between the sample surface and
a metal band in contact with the paper. The result is a measure of roughness in
micrometer.
6.3.3. Strength properties
Short-span compression (SCT) was analyzed at KCL (Oy Keskuslaboratorio -

Centrallaboratorium Ab). In SCT measurement, a sample cut in machine direction
(MD) is fixed with two clamps having a short distance between them. The clamps are
brought together with increasing force until the samples structural integrity collapses.
The final corrugated board box compression resistance depends directly on the liner
and fluting strength, which is usually measured as SCT or ring crush test (RCT) for
liner and as Concora medium test (CMT) for fluting (Holik, 2013).
Z-direction strength was measured with an L&W Z-direction tester. A sample

(dimensions 10 x 30 cm) was fed into the tester between two rolls of double sided tape.
The tape was attached to the paper by clamping from the topside and backside, and the
force required to separate the clamps with paper tearing is recorded.
6.3.4. Other paper analyses
Dynamic water retention (DWR) was used to measure starch pickup or starch
affinity of the paper. Samples of the base papers were used and a 12% solids starch
solution of starch 1 was kept at 60 °C before filling the DWR testing equipment
cylinder. The device lets starch flow past the surface of the paper sample, and DWR
was measured as weight increase of the paper sample in g/m2.
6.4. Surface sizing agent (SSA) analysis
The charge of the SSA’s was measured with a CAS Charge analyzer. A sample was
diluted to 0,1% strength, and the device measured the required amount of anionic
(PES-Na) or cationic (poly-DADMAC) titrant to neutralize the sample.
Hydrophobicity was measured with contact angle measurements from films. SSA was
diluted to 25% solids, and a 150 µm wet film was drawn on a glass plate. The film was
dried in a 100 °C oven to mimic the surface sizing method. After drying, the film was
allowed to stabilize overnight in a conditioned room (t 23 °C, RH 50%) before
measuring water contact angles with Fotocomp DropPrint equipment and software.
The particle size of the SSA solutions was measured with a Malvern Zetasizer. A
diluted sample is put into the machine, and light scattering is used to measure the
particle size distribution.
III RESULTS AND DISCUSSION
7. Differences in SSA efficiency
In the surface sizing experiments a broad range of sizing effect, measured in Cobb 60 s
of the sized paper, was witnessed on testliner 1 (see Figure 10). Some sizing agents
gave a low Cobb value with a small addition amount, while other SSAs had barely any
effect even at higher addition amounts.
100 Starch 1
80
Cobb 60 s, g/m²
60
40
20
0
0 1 2 3 4 5 6
Amount sizing agent, kg/t dry
Figure 10. All surface sizing test points on testliner 1
Figure 11 and Figure 12 illustrate a significant difference in the sizing effect of SSA 31
and SSA 30 between the different base papers. Testliner 1 seems to respond very well
to sizing, while testliner 4 is most difficult to hydrophobize. The differences in sizing
effect between sizing agents and between different base papers are well known in the
industry. A study by Karademir et al. (Karademir, Ozdemir & Imamoglu, 2007)
investigated the differences in sizing effect for office paper, old corrugated, magazine
paper and newspaper.
140
120 TL 4
100 TL 2
KL
Cobb 60 s, g/m²
80
C TL 3
60
TL + S
40
TL 5
20
TL 1
0
0 1 2 3 4 5 6 7
Sizing agent amount, kg/t dry
Figure 11. Sizing effect of SSA 31 on the different base papers (TL = testliner, KL = kraftliner,
C = copy paper)
160
140
120
Cobb 60 s, g/m²
TL 5
100
TL + S
TL 3
80
60
TL 2
40
20 TL 1 TL 4 KL
C
0
0 1 2 3 4 5 6
Sizing agent amount, kg/t dry
Figure 12. Sizing effect of SSA 30 on the different base papers

8. Effect of surface sizing method parameters
Before actual experiments on the sizing effect of SSAs could be made, the laboratory
surface sizing method was studied. The following chapter details effects witnessed
on surface size pickup and sizing effect when adjusting the surface sizing method
parameters.
Paper moisture content

The moisture content of the base paper was not found to affect either the surface size
pickup or the sizing effect. It was decided to keep the conditioned base paper sheets in
sealed plastic bags before surface sizing so that moisture content variability would be
minimized.
Previously, the sized sheets were brought to the conditioning room and left to stabilize
overnight in piles after surface sizing. The conditioning was found to be ineffective
because the bottom samples had not stabilized quickly enough due to lack of airflow
around the samples. Clips with hangars were used to condition the surface sized sheets
in bunches of approximately ten sheets/clip overnight.
Size press speed

Size press speed was not found to have a measurable effect on starch pickup even
though a pickup increase with increasing paper machine speed has been reported in
the literature (Maurer, 2009; Brander, James.,Thorn, Ian., 1997). However, the
laboratory size press operates at extremely low speeds compared to industrial scale
paper machines. Small changes in the speed probably did not cause measurable effects
at the low speeds used in laboratory. Similarly, it was found in other studies that a
speed increase from 40 to 100 m/min did not increase starch pickup (Shirazi et al.,
2005). No effect on sizing efficiency was found.
Size press pressure

Higher size press nip pressure decreased the surface size pickup in the laboratory size
press. Similar effects for pond type size presses in industrial and laboratory scale have
been reported in the literature (Maurer, 2009; Shirazi et al., 2005). Figure 13 shows
the effect on testliner + starch base paper sized with 18% solids surface size. It was
decided to use as low pressure as possible which could still be monitored to prevent
surface size solids content from becoming too high while achieving the goal of 40 kg/t
starch pickup; this was determined to be “5” on the pressure lever.
60
50
Starch pickup, kg/t
40
30
20
0 10 20 30 40 50
Size press pressure lever position
Figure 13. Size press pressure and 18% starch pickup (base paper TL + S)
Starch volume in pond

The starch volume in the pond did not influence pickup or sizing effect, except when
the small volume of 20 ml left dry patches on the first sheet run through the size press.
It was decided to use 150 ml as starch volume to facilitate Brookfield viscosity
measurements in a 200 ml decanter.
Drying conditions
IR drying and oven drying above 105 °C burned the sized sheets, causing them to
appear overly hydrophobic. The IR dryer was deemed too unpredictable and powerful
to produce even, consistent drying results. Oven temperature and time did not affect
pickup, except when drying continued during conditioning and pickup seemed too high.
The drying method was left as it was, 3 min in a 105 °C oven.
Surface size solids content

Surface size solids content was the parameter that affected pickup the most in the
laboratory size press. For base paper TL + S, an ideal pickup, with the other parameters
adjusted, was obtained at solids content of 17%. For testliner 1, 9% solids content was
found to give 35-45 kg/t pickup, and for testliner 3, 12% solids content was suitable.
When testing further base papers, a correlation between the pickup at 12% solids
content and the needed solids % for 40 kg/t pickup was found (Figure 14). This
correlation could be used to determine appropriate solids % for testing of new base
papers.
70
60 TL 1 y = -3,9022x + 90,158
R² = 0,9157
Pickup starch 12%, kg/t
TL 5
50
TL 2
40 C
TL 4
30 TL + S
20 KL
TL 3
10
0
7 9 11 13 15 17 19 21
Solids content for 40 kg/t, %
Figure 14. Correlation between size pickup at 12% solids content and solids content required
for 40 kg/t pickup for different base papers.
Surface size viscosity

The viscosity of the surface size with starch 1 and starch 2 is shown in Table 3. A more
substantial difference in viscosity was witnessed with higher solids content.
Table 3. Viscosity of surface size with Starch 1 and Starch 2
Base paper 2 Base paper 3 Base paper 1

Solids 9% Solids 12% Solids 18%
Starch Starch Starch Starch Starch Starch
1 2 1 2 1 2
Brookfield 100 rpm mPas 14 16 18 22 31 47
The holdout of surface size was expected to increase with the higher viscosity starch 2,
as described in the literature (Maurer, 2009; Gess, Rodriguez, 2005; Shirazi et al., 2005;
Jeong et al., 2012). When comparing the pickup of surface size on the different base
papers (Figure 15), the increased holdout may explain the differences observed. The
higher holdout of starch 2 may prevent penetration through the highly porous
testliner 1, thus decreasing pickup. For the dense testliner + starch base paper, starch
holdout may be high irrespective of the starch, and thus the higher viscosity may cause
more starch to stay on the paper surface through the nip passage, increasing pickup.
70
Surface size pickup, kg/t
60
Testliner 3
Testliner 1 Testliner + starch
50
40
30
6 9 12 15 18 21
Surface size solids content, %
Starch 1 LV Starch 2 HV
Figure 15. The effect of starch viscosity on surface size pickup (LV = low viscosity, HV =
high viscosity)
The higher viscosity starch decreased the porosity of the papers, as measured with
Bendtsen air permeability (Figure 16), consistent with results obtained on a pilot scale
paper machine in a doctoral thesis study. (Lipponen, 2005) For base paper testliner +
starch, the difference is negligible, probably because of the already low porosity of the
dilute-starch-containing base paper. No significant differences in Cobb 60 s values
were found.
400
Bendtsen porosity, ml/min

300 Testliner 2
200
Testliner 1
100 Testliner + starch
0
6 8 10 12 14 16 18 20
Surface size solids content, %
Starch 1 LV Starch 2 HV
Figure 16. The effect of starch viscosity on Bendtsen porosity
Surface size temperature

Surface size temperature was not shown to affect either pickup or hydrophobicity.
Laboratory surface sizing method evaluation

When testing the effect of the sizing method parameters on the sizing effect, an
extremely well-functioning anionic surface sizing agent was used (SSA 1), and no
effect on Cobb results was found when adjusting method parameters. However, this
surface sizing agent may have distorted the results because of its excellent sizing effect
so that differences may not have been measurable.
9. Effect of base paper properties
In this chapter, the measured base paper properties are presented. Then, the effects of
base paper properties on starch pickup, paper strength and sizing effect are studied.
9.1. Base paper properties
The results of the basic paper analyses are presented in Table 4. The copy paper is
more porous and less rough than the liners, and the testliner with starch is least porous.
Also between the testliners, differences in bulk, porosity and roughness can be seen,
despite the similar raw material composition. PPS and Bendtsen roughness show that
the other base papers have very rough surfaces, especially the kraftliner, except the
copy paper which is smoother. Ash content is lower for the virgin fiber containing
kraftliner, as expected, and higher for the filled copy paper.
Table 4. Basic base paper properties
Base paper properties TL 1 TL 2 TL 3 TL 4 TL 5 KL TL + C

S
Grammage g/m² 84 97 95 150 107 116 90 78
Thickness µm 139 164 147 231 187 202 136 120
Bulk cm³/g 1,7 1,7 1,5 1,5 1,8 1,7 1,5 1,5
Bendtsen porosity ml/min 360 440 300 450 380 370 210 1050
Gurley s 34 28 44 24 31 31 55 10
Cobb 60s g/m² 109 140 96 228 156 136 86 82
Bendtsen
roughness
topside ml/min 1100 1600 800 950 1300 2000 950 330
backside ml/min 1450 1100 800 950 1800 2200 1200 370
PPS
topside µm 8 11 9 8 10 12 9 6
backside µm 11 10 10 9 13 12 12 7
Ash content
450C % 15 12 17 15 14 6 13 21
925C % 10 8 12 11 10 5 10 12
The Hg-porosimetry results (Table 5 and Figure 17) show similarly that the copy
paper is most porous, while the testliner with starch is least porous. The testliners are
equal in porosity, except testliner 4 which is more porous.
Table 5. Hg-porosimetry
Hg-porosimetry TL 1 TL 2 TL 3 TL 4 TL 5 KL TL + C
S
Pore volume mm³/g 530 537 525 605 598 494 395 621
Porosity % 32,0 31,8 33,9 39,3 34,2 28,4 26,1 40,4
250
200
Relative pore volume, mm³/g
150
100
50
0
0,01 0,1 1 10
Pore size diameter, µm
TL 1 TL 2 TL 3 TL 4 TL 5 KL TL + S C
Figure 17. Hg-porosimetry pore size and volume
Dynamic water retention measurements with starch (Table 6) show that testliner 1, 2 and 5,
and the copy paper, absorb the highest amounts of 12% starch. The same testliners also have
the lowest t(MAX) values with the ultrasonic dynamic penetration tester, which means that
their surface wetting time is fastest. The W roughness values are also low for these base papers.
Table 6. DWR and ultrasonic dynamic penetration tester results
DWR TL 1 TL 2 TL 3 TL 4 TL 5 KL TL + C
S
DWR starch 12% g/m² 68 82 47 62 72 52 50 69
Ultrasonic dynamic penetration tester, water
I(5s) % 12,86 10,29 50,48 4,63 10,71 70,83 22,76 32,32
t(50%) s 0,59 1,19 5,04 1,57 0,58 9,80 2,77 3,50
t95 s 0,23 0,35 1,26 0,43 0,15 1,24 0,75 1,00
tMAX s 0,13 0,16 0,52 0,21 0,08 0,45 0,32 0,47
W 0,20 0,28 2,03 0,51 0,01 0,90 0,44 1,19
A 30 21,70 24,30 26,90 26,80 24,90 24,10 26,90 25,70
9.2. Starch pickup
Starch pickup, and therefore solids content for 40 kg/t surface size, varies significantly
between the base papers (Table 7). For all base papers studied, DWR with 12% starch
correlated with 12% starch pickup (Figure 18), except a high starch pickup for
testliner 1. DWR measurements with starch could therefore in many cases be used to
predict starch pickup of the base paper.
Table 7. Starch pickup
Starch pickup TL 1 TL 2 TL 3 TL 4 TL 5 KL TL + C
S
Pickup starch 12% kg/t 60 40 16 25 40 19 20 33
Solids for 40 kg/t % 9 12 20 17 12 17 17 16
70
60 TL 1
12% starch pickup, kg/t

50
40 TL 2
TL 5
30
C
20 TL 4
10 y = 0,9004x - 24,865
TL 3 KL
TL + S
0
40 45 50 55 60 65 70 75 80 85
DWR starch 12%, g/m2
Figure 18. DWR with 12% starch correlation to 12% starch pickup
9.3. Paper strength
Figure 19 illustrates how an increased starch amount applied to testliner 1 increased

the short-span compression (SCT) strength of the base paper. The results indicate that
the laboratory surface sizing method produces a steady strength increase. Further, the
achieved SCT strength of 1,5 kN/m with 40 kg/t is a typical value for commercial
surface sized testliners of similar grammage. (Mondi group, 2018; Saica, 2018)
1,8
1,7
1,6
SCT CD, kN/m
1,5
1,4
1,3
1,2
y = 0,0087x + 1,0848
1,1
1
0 20 40 60 80
Starch amount applied, kg/t
Figure 19. Testliner 1 strength increase with increased starch amount

Table 8. Z-direction strength and SCT strength
Z-direction TL 1 TL 2 TL 3 TL 4 TL 5 KL TL + C
strength S
base paper kPa 638 527 791 420 569 495 850 634
with starch 40 kg/t kPa 750 580
SCT, cd
base paper kN/m 1,1 1,5 1,8 1,7 1,5 2,0 2,1 1,2
with starch 40 kg/t kN/m 1,5 2,0 2,2 2,6 1,9 2,4 2,3 1,6
increase kN/m 0,4 0,5 0,4 0,9 0,4 0,4 0,2 0,4
increase % 36 33 22 53 27 20 10 33
Base paper and surface sized paper strength is presented in Table 8. The kraftliner and
the testliner with starch have the highest SCT strengths, and the testliner with starch
also has the highest Z-direction strength. This is to be expected because of the higher
strength and increased bonding potential of virgin fibers compared to recycled fibers,
and because the added starch serves to glue together the base paper.
900
800
Z-direction strength, kPa
TL + S
700 TL 3
C
600 TL 1
TL 5
500
TL 2
400
KL
TL 4
300
0 10 20 30 40 50 60
Gurley air permeability, s
Figure 20. Gurley air permeability and Z-direction strength
For the liners, Z-direction strength correlates with porosity measured by air
permeability (Figure 20). The copy paper has higher Z-direction strength with higher
porosity, probably because of the different raw material and production process. SCT
strength of sized paper correlates with SCT strength of base paper (Figure 21), which
means that liner can be strengthened by either adding more starch or strengthening the
base paper.
SCT CD basepaper, kN/m 2,5
1,5
KL
TL + S
TL 4
1 TL 3
C TL 5 TL 2
TL 1
0,5
1,4 1,6 1,8 2 2,2 2,4 2,6 2,8
SCT CD with 40 kg/t starch, kN/m
Figure 21. SCT for 40 kg/t starch sized base paper depends on base paper SCT
9.4. Sizing effect
The sizing effect, measured in Cobb 60 s, of SSA 30 and SSA 31 on different base
papers is presented in Table 9. Testliner 1 and testliner 5 had the highest sizing effect
with both sizing agents, while testliner 3, testliner with starch and copy paper had the
lowest sizing effect.
Table 9. SSA sizing effect
Sizing effect - Cobb 60s TL 1 TL 2 TL 3 TL 4 TL 5 KL TL + C

S
Base paper g/m² 109 140 96 228 156 136 86 82
SSA 30, 3 kg/t g/m² 29 71 79 146 23 78 73 74
decrease % 73 49 18 36 86 43 15 10
decrease g/m² 80 69 17 82 134 58 13 8
SSA 31, 3 kg/t g/m² 26 104 90 162 30 103 80 74
decrease % 76 26 6 29 81 24 7 9
decrease g/m² 83 36 6 66 126 33 6 8
90
80
3 kg/t SSA Cobb 60 s improvement, %

TL 5
70 TL 5
60 KL TL 1
TL 1
50
KL TL 2
40
TL 4
30 TL 3
TL 4 TL 2
20
TL 3
10 TL + S
C
TL + S C
0
40 45 50 55 60 65 70 75 80 85 90
DWR 12% starch, g/m²
SSA 30 SSA 31
Figure 22. DWR 12% starch correlation with sizing effect
Both testliner 1 and testliner 5 have high DWR values with starch 12% (Figure 22),
however, so does the copy paper and testliner 2 despite having a lower sizing effect.
When comparing only the testliners, a higher DWR value seems to indicate better
sizing effect.
160
TL 5
140
Cobb improvement 3kg/t, g/m²
TL 5
120
100
TL 4
80 KL
TL 1
TL 1
60
TL 2 KL
40
TL 2
20 TL 4
TL + S TL 3
TL + S C TL 3
0
0 0,1 0,2 0,3 0,4 0,5 0,6 0,7
t(MAX), s
SSA 30 SSA 31
Figure 23. t(MAX) and sizing effect

Figure 23 shows that the t(MAX) parameter in ultrasonic dynamic penetration testing
does correlate with the sizing effect. The t(MAX) parameter signifies the time point
when the surface of the paper is wetted, and could thus indicate whether the paper
surface is already hydrophobic or not. The base paper usually contains trace amounts
of internal sizing agents to control surface size pickup and prevent web breaks, which
could explain the differences in t(MAX). Fiber refining processes, and recycled fiber
prior sizing could also cause differences in the paper surface chemistry.
Similarly, the parameter W in the ultrasonic dynamic penetration testing correlates

with sizing effect (Figure 24). The W parameter is a machine-calculated value for
surface roughness characterization.
160 TL 5
140 TL 5 TL 1
3 kg/t SSA Cobb 60s improvement, g/m2
120 TL 1
100
80 KL
60
TL 4 KL
40
TL 2
TL 4
20 TL 3
TL + S
TL 2 TL + S C TL 3
0
0 0,5 1 1,5 2
Ultrasonic dynamic penetration tester, W
SSA 30 SSA 31
Figure 24. Parameter W and sizing effect

10. Effect of SSA properties on sizing effect
Particle size
A smaller particle size of the sizing agent was found to give better sizing effect, and
PAC addition aids the sizing further (Figure 25 and Figure 26). Smaller SSA particles
should logically cover more of the paper surface because the number of particles is
higher for the same weight of SSA and therefore the specific surface area is higher.
120
TL 1 4 kg/t SSA Cobb 60s, g/m²
100
80
60
40
20
0
0 50 100 150 200 250 300
SSA particle size z-average, nm
Figure 25. Sizing effect and SSA particle size, no PAC
100
90
80
70
60
50
40
30
20
10
0
0 50 100 150 200 250 300
SSA particle size z-average, nm
Figure 26. Sizing effect and SSA particle size, with PAC
Contact angle (hydrophobicity)

In Figure 27 and Figure 28, the sizing effect is correlated to the SSA hydrophobicity,
measured by water contact angle on oven-dried films of 25% solids content SSA.
Figure 27 shows sizing effect with PAC, and Figure 28 without PAC. SSA
hydrophobicity seems to correlate with sizing effect without PAC, however with PAC
there is no clear correlation. The use of PAC may, therefore, either increase the
hydrophobicity of the final surface size solution or have some SSA particles aligned
more favorably for hydrophobicity.
120
100
80
60
40
20
0
0 20 40 60 80 100 120
SSA film water contact angle, °
Figure 27. Sizing effect and SSA film water contact angle, no PAC
100
90
80
70
60
50
40
30
20
10
0
0 20 40 60 80 100 120
SSA film water contact angle, °
Figure 28. Sizing effect and SSA film water contact angle, with PAC
Conclusions
In this study, eight different base papers were analyzed, and 45 different surface sizing
agents (SSA’s) were tested. The goal was to optimize a laboratory scale method for
studying surface sizing and to increase the understanding of the mechanisms that
control hydrophobization of paper by surface sizing.
Differences in base paper properties were found to impact the starch pickup, paper
strength, and sizing efficiency. For SSA’s both the particle size and the particle
hydrophobicity was found to influence the sizing effect.
A range of sizing effects with the same dosage of different SSA’s was witnessed on
testliner 1. Some SSAs gave a low Cobb value with a small addition amount, while
others did not respond to even higher dosages. Differences were also witnessed
between base papers; the same SSA worked well with some base papers while others
were more difficult to make hydrophobic.
The surface sizing method was optimized for 40 kg/t starch pickup for all the base
papers. Nip pressure and surface size solids content affected the starch pickup. Nip
pressure was set to a minimum and solids content was adjusted for each base paper.
The starch pickup for each base paper could then be estimated using DWR
measurements with starch. Surface size viscosity affected the porosity of the surface
sized paper, and porosity of the base paper affected the starch pickup.
SCT strength of surface sized paper depended on base paper SCT strength and the
applied starch amount. A more porous base paper was weaker in Z-direction, as
expected.
The sizing effect correlated with wettability of the base paper, and DWR
measurements with starch. The easier the base paper was to wet and the more starch it
absorbed in DWR measurements, the stronger was the sizing effect.
Sizing effect was further affected by the SSA particle size, SSA hydrophobicity, and
PAC usage. Smaller SSA particles gave a stronger sizing effect, which is logical given
the increased surface coverage due to the larger specific surface area. A more
hydrophobic SSA also generally gave a higher sizing effect, but with the use of PAC,
the hydrophobicity of the SSA lost significance.
The results indicate that base paper properties can be optimized for starch pickup and
surface sizing effect. The costs of applying excess SSA and excess starch to the paper
surface could thus be decreased. Further studies are recommended on adjusting the
base paper properties to improve sizing effect and on SSA particle interaction with the
paper surface, especially in combination with PAC or other promoters. Printability and
gluability of surface sized testliner should also be considered in further studies.
Swedish summary - Svensk sammanfattning
Vattenavstötande ytlimning av testliner
I detta diplomarbete behandlas hydrofobisk ytlimning av testliner. Ämnet är relevant

eftersom den växande internationella näthandeln höjer efterfrågan på billiga, hållbara
förpackningsmaterial, såsom korrugerade förpackningar av wellpapp.
Förpackningsmaterialens andel i pappersindustrin ökar därtill, medan andelen papper
som används för kommunikation minskar. Dessutom återvinns allt mer papper och
kartong, vilket leder till en förökad användning av testliner tillverkad av returfiber
jämfört med nyfiberbaserad kraftliner (CEPI, 2016; Holik, 2013; ERPC, 2016).
Ytlimningsprocessen, dvs. applikationen av stärkelse och ytlimningsmedel till

pappersytan, är en gammal process. Trots detta är ytlimning av testliner utmanande
på grund av det stora antalet variabler som påverkar hydrofoberingen. Tidigare
hydrofoberades liners uteslutande genom mäldlimning, dvs. limningsmedel lades till
i våtändan på pappersmaskinen. Våtändan blir dock lätt komplicerad och kan orsaka
produktionsstopp, speciellt i testlinertillverkning då returfibrerna kan innehålla en
oändlig mängd olika kemikalier. Nyligen har ytlimning blivit allt vanligare på grund
av strävan efter förenklad våtända, besparingar i mängden ytlimningsmedel då
medlet appliceras endast på ytan och bättre möjligheter att finjustera pappersytans
egenskaper (Bajpai, 2015; Zeng, 2013).
Hydrofoberingsmekanismen i ytlimning är inte väl förstådd och laboratoriestudier

korrelerar sällan med verkliga test på pappersmaskiner. Det finns inte heller något
ytlimningsmedel som skulle fungera lika bra på alla testliners. Med en högre
förståelse för ytlimningsprocessen kunde ytlimningsmedel appliceras mer sparsamt
och ytlimningsmedel kunde utvecklas effektivare. En pålitlig laboratoriemetod skulle
underlätta utvecklingen av ytlimningsmedel och förbättra möjligheterna att förutsäga
ytlimningsmedlens effektivitet i praktiken.
Målet för detta diplomarbete var att närma sig en förståelse för
hydrofobiseringsmekanismen i ytlimning av testliner. Först undersöktes och
optimerades en laboratoriemetod för ytlimning och sedan testades 45 olika
ytlimningsmedel på åtta olika baspapper. Baspappren och ytlimningsmedlen
analyserades och resultaten undersöktes för korrelation med stärkelseupptagning och
ytlimningseffekt.
Baspappren som användes i arbetet var fem olimmade testliners, en testliner med
utspädd ytlimningsstärkelse, en olimmad kraftliner och ett kopieringspapper. Två
modifierade lågviskositetsstärkelser från Roquettes ”Stabilys A” produktsortiment
användes, den ena med något högre viskositet. Stärkelsernas struktur simulerar
enzymatiskt konverterad stärkelse. Båda stärkelserna var i pulverform och kokades
med jetkokare i 134ºC till en bestämd torrhalt. Ytlimningsmedlen inkluderade både
kommersiella referensprodukter och utvecklingsprodukter. Två av ytlimningsmedlen
testades på alla baspapper och alla ytlimningsmedel testades på testliner 1, de flesta
både med och utan polyaluminiumklorid (PAC).
Laboratoriemetoden testades och optimerades först för att uppnå en konstant

stärkelseupptagning på 35-45 kg/t för alla olika baspapper och en konstant
ytlimningseffekt. Variablerna som ändrades var limpressens tryck, limpressens
hastighet, ytlimmets torrhalt, papprets fukthalt, stärkelsevolymen i dammen,
torktiden, torktemperaturen och ytlimmets temperatur. Testerna utfördes så att alla
variabler hölls i utgångsläge, utom den som testades.
Variabeltesterna visade att ytlimmets torrhalt och limpressens tryck påverkade

stärkelseupptagningen. Ett högre limpresstryck gav lägre stärkelseupptagning och
högre torrhalt ökade stärkelseupptagningen. Eftersom testlinern med utspädd
stärkelse hade väldigt låg stärkelseupptagning och ytlimmets torrhalt påverkade
upptagningen mest, bestämdes det att ett konstant lågt limpresstryck skulle hållas för
alla baspapper och att stärkelseupptagningen skulle ställas in med hjälp av torrhalten.
För att uppnå 40 kg/t stärkelseupptagning användes torrhalter från 9 % till 17 %

beroende på baspapper. Det kunde också konstateras att stärkelseupptagningen av
12 % ytlim kan användas för att förutspå rätt torrhalt för 40 kg/t upptagning. Ingen
av variablerna påverkade ytlimningseffekten. Ytlimningsmedlet som användes i
metodtesterna var dock väldigt effektivt och det är möjligt att hydrofobiciteten var
för hög för att skillnader i variablerna kunde upptäckas.
För att undersöka skillnader i ytlimningseffekten mellan baspappren och

ytlimningsmedlen ytlimmades de olika baspappren med olika ytlimningsmedel.
Ytlimsreceptet utgick från 40 kg/t ytlimsupptagning med x kg/t stärkelse, y kg/t
ytlimningsmedel och z kg/t PAC så att summan av x, y och z blev 40.
Ytlimningsmedlen doserades med 1, 2 och 4 kg/t på testliner 1, för de övriga
baspappren användes 1, 3 och 6 kg/t på grund av sämre ytlimningseffekt. Då PAC
användes, var doseringen 2 kg/t oberoende av ytlimningsmedlets dosering. Ytlimmet
applicerades på båda sidorna av pappret i laboratorie-filmpressen, och därmed kan
doseringsmängderna inte direkt jämföras med ensidiga filmpressar i industrin.
Baspappersanalyserna utgjordes av vattenabsorption (Cobb 60s och dynamisk

penetrationstestare med ultraljud), kontaktvinklar för vatten, porositet
(kvicksilverporosimetri, Gurley luftpermeabilitet och Bendtsen luftpermeabilitet),
släthet (Bendtsen och Parker Print-surf (PPS)), styrka (kompressionsstyrka och z-
styrka) och stärkelseupptagning med en dynamisk vattenretentionsmätare.
Ytlimningsmedlens laddning, hydrofobicitet genom kontaktvinklar på film och
partikelstorlek mättes. Ytlimmets pH, viskositet, torrhalt och temperatur analyserades
vid ytlimningarna.
Vid ytlimningarna av testliner 1 med alla ytlimningsmedel uppstod markanta

skillnader i limningseffekt. Vissa ytlimningsmedel gav ett lågt Cobb värde med en
liten tillsatsmängd, medan andra ytlimningsmedel hade knappt någon effekt trots
större tillsatsmängder. Mellan baspappren fanns också stora skillnader, testliner 1

verkade vara lätt att hydrofobera, medan testliner 4 hade minst limningseffekt.
Stärkelseupptagningen varierade avsevärt mellan baspappren. För testlinersen

korrelerade totala porvolymen med stärkelseupptagningen. Skillnaderna i porvolym
mellan testlinersen kan bero på olika fiberraffineringsmetoder och grader av
fiberraffinering och skillnader i retention av fyllmedel och finmaterial. Dynamiska
mätningar av vattenretention med 12 % stärkelse korrelerade med upptagningen av
12 % stärkelse för alla andra baspapper, utom testliner 1. DWR mätningar kunde
alltså användas för att förutsäga stärkelseupptagning.
Liner baspapprens Z-styrka korrelerade väl med luftpermeabiliteten, så att ett mer
poröst papper var svagare. Kopieringspappret hade högre Z-styrka trots hög
luftpermeabilitet, men råmaterialet och produktionsprocessen skiljer sig också från
liners. Kraftlinern och testlinern med stärkelse var, som förväntat, starkast i Z-
riktning, eftersom nyfibrer är starkare än returfibrer och har bättre bindningspotential
mellan fibrerna och stärkelsen limmar ihop testlinern för högre styrka. Mätt med
kvicksilverporosimetri korrelerade porositeten med kompressionsstyrkan för alla
baspapper. Kompressionsstyrkan för ytlimmat papper var beroende av baspapprets
styrka och den applicerade stärkelsemängden, så att kompressionsstyrkan ökade
logiskt med större mängder stärkelse. För testliner 1 mättes en kompressionsstyrka
på 1,5 kN/m med 40 kg/t stärkelse, vilket är ett typiskt värde för kommersiella liners
av samma ytvikt.
Limningseffekten, mätt i Cobb 60s, varierade mellan baspappren. Högre

stärkelseupptagning i de dynamiska vattenretentionsmätningarna med stärkelse,
verkade tyda på bättre limningseffekt för baspappren, med undantag för
kopieringspappret och testliner 2. Den dynamiska penetrationstestaren med ultraljud
gav ett vätningsvärde t(MAX) för baspappren och detta värde korrelerade bra med
limningseffekten. t(MAX) anger tiden då papprets yta är i fullständig kontakt med
vattnet och kunde alltså rapportera den hydrofobiska pappersytans egenskaper.

Mindre hydrofobiska ytor, eller baspapper med lägre t(MAX) värde, hade högre
limningseffekt. Skillnaderna i ythydrofobicitet beror antagligen på små mängder
mäldlimningsmedel i baspappren eller andra fiberraffineringsprocesser och
råmaterial.
Av ytlimningsmedlens egenskaper gav en mindre partikelstorlek bättre

ytlimningseffekt och tillsatsen av PAC gav generellt bättre ytlimningseffekt. Mindre
partiklar täcker mer av pappersytan på grund av att partikelmängden blir större för
samma tillsatsmängd och ytarean blir högre. Högre hydrofobicitet på
ytlimningsmedlet, mätt med kontaktvinkelmätningar på film på glasplatta, gav också
bättre limningseffekt utan PAC. Med PAC fanns det ingen tydlig korrelation. PAC
höjer därmed antingen hydrofobiciteten för ytlimmet eller justerar partiklarnas
position för bättre hydrofobicitet.
Dessa resultat antyder att genom pappersanalys kan baspappret optimeras för
stärkelseupptagning och limningseffekt. Kostnaderna för att applicera ett överskott
av stärkelse och ytlimningsmedel kunde därmed minskas. Ytterligare studier
rekommenderas av justering av pappersegenskaper för ytlimningseffekt och av
ytlimningspartiklarnas interaktion med pappersytan, speciellt i kombination med
PAC eller andra främjande tillsatsmedel. Tryckbarhet och limbarhet av ytlimmad
testliner borde också betraktas i fortsatta studier.
References
Andersson, L., Ridgway, C.J. & Gane, P.A.C. 2013, "Defining the mechanism of
sizepress starch penetration in filled unsized fibrous products - a traditional
technology revisited", vol. Nordic Pulp & Paper Research Journal Vol 28, no. 4,
pp. 547-558.
Bajpai, P. 2015, "Chapter 3 - Pulp and Paper Chemicals∗ " in Pulp and Paper Industry,
ed. P. Bajpai, Elsevier, Amsterdam, pp. 25-273.
Brander, James.,Thorn, Ian., 1997, Surface application of paper chemicals, Blackie

Academic & Professional, London; New York.
CEPI 2016, CEPI Key Statistics 2015, Confederation of European Paper Industries.
ERPC 2016, ERPC Monitoring Report 2015, European Recovered Paper Council.
Forsström, U. 2003, Interactions between base paper and coating color in metered
size press coating, Keskuslaboratorio.
Gane, P.A.C. 2001, "Surface treatment: Sizepress tradition, current development and
a pigmented chemical future", vol. Twelfth Fundamental Research Symposium.
Gess, J.M. & Rodriguez, J.M. 2005, The sizing of paper, 3. ed edn, TAPPI Press,
Atlanta, GA.
Hagiopol, C. & Johnson, J.W. 2012, Chemistry of modern papermaking, CRC, Boca
Raton, FL.
Höke, U., Höke, U. & Schabel, S. 2010, Papermaking science and technology. Book
7, Recycled fibre and deinking, 2nd ed., totally updated version edn, Paper
Engineers' Association : Paperi ja puu, Helsinki.
Holik, H. 2013, Handbook of paper and board, 2nd, rev. and enl. ed edn, Wiley-VCH,
Weinheim.
Jeong, Y.B., Lee, H.L., Youn, H.J., Jeong, K.H. & Ryu, H. 2012, "Influence of the
Viscosity of Surface Sizing Starch Solutions on Surface Sizing Effect of
Linerboard", Journal of Korea Technical Association of The Pulp and Paper
Industry, vol. 44, no. 5, pp. 54.
Jonhed, A. 2006, Properties of modified starches and their use in the surface treatment
of paper, Karlstad University, Faculty of Technology and Science, Chemical
Engineering.
Karademir, A., Ozdemir, F. & Imamoglu, S. 2007, "Effects of Alum-Rosin Sizing on

the Properties of Some Wastepaper Grades", vol. Biotechnology, no. 6, pp. 148-
152.
Khantayanuwong, S. 2002, "Effects of beating and recycling on strength of pulp fibers

and paper", vol. Kasetsart Journal of Nat. Sci, no. 46, pp. 193-199.
Lee, H.L., Shin, J.Y., Koh, C., Ryo, H., Lee, D. & Sohn, C. 2002, "Surface sizing with
cationic starch: its effect on paper quality and papermaking process", TAPPI
JOURNAL, vol. 1, no. 1, pp. 34-40.
Lehtinen, E., Klass, C.P., Suomen Paperi-insinöörien Yhdistys & Technical

association of the pulp and paper industry 2000, Papermaking science and
technology. / 11, Pigment coating and surface sizing of paper, Fapet Oy :
Published in cooperation with the Finnish Paper Engineers' Association and
TAPPI, Helsinki.
Lipponen, J. 2005, Surface Sizing With Starch Solutions at High Solids Contents,
Dissertation for the degree of Doctor of Science in Technology, Helsinki
University of Technology, Laboratory of Paper Technology.
Lopes, P., Preciso, C., Sousa, A.M.d. & Ferreira, P. 2010, "Sizing evaluation of
uncoated fine papers", .
Maurer, H.W. 2009, "Chapter 18 - Starch in the Paper Industry" in Starch (Third
Edition), eds. J. BeMiller, , & R. Whistler, Academic Press, San Diego, pp. 657-
713.
Mondi group 2018, 2018-last update, ProVantage Testliner [Homepage of Mondi

Group], [Online]. Available: https://www.mondigroup.com/en/products-and-
solutions/categories/containerboard/provantage-testliner-3/ [2018, 8].
Nazhad, M.M., Li, K., Sutjipto, E.R. & Pongpattanasuegsa, S. 2008, "Effect of
recycling on paper properties", vol. Tappsa Journal.
Pandian, V.E., van Calcar, D. & Wolff, B.W. 1994, Use of zirconium salts to improve
the surface sizing efficiency in paper making, Google Patents.
Ryu, J. & Lee, H.L. 2007, "Improvement of plybond strength of two-ply sheets by
spraying of starch blends", TAPPI JOURNAL, vol. 6, no. 5, pp. 3-8.
Saica 2018, 2/2018-last update, Product information [Homepage of Saica], [Online].

Available: http://www.saica.com/wp-content/uploads/2018/02/ds_ENG.pdf
[2018, 8].
Shirazi, M., Esmail, N., Garnier, G. & van, d.V. 2005, "Starch Penetration into Paper
in a Size Press", Journal of Dispersion Science and Technology, vol. 25, no. 4, pp.
457-468.
Sievänen, H. 2010, Suitability of foam coating on application of thin liquid films,

Lappeenranta University of Technology, Faculty of Technology, Chemical
Technology.
Sirviö, J. & Nurminen, I. 2004, Paper Properties - Systematic Changes in Paper

Properties Caused by Fines, National Business Publications Ltd, Westmount, P.Q.
Valmet 2017, 2017-last update, OptiSizer Spray - spray sizing for best controllability.
Available: http://www.valmet.com/board-and-paper/board-and-paper-
machines/sizing/spray-application/ [2017, 4/3].
Vanier, N.L., El Halal, S.L.M., Dias, A.R.G. & da Rosa Zavareze, E. 2017, "Molecular
structure, functionality and applications of oxidized starches: A review", Food
Chemistry, vol. 221, pp. 1546-1559.
Wygant, R., Kendrick, J. & Walter, J. 2008, "Metered Size Press Pigmentation for
Fiber Reduction", TAPPI/PIMA PaperCon ’08 Conference, May 4-7 2008.
Wygant, R., Lucidi, G. & Kendrick, J. 2009, "Pond Size Press Pigmentation with Platy
Kaolin", TAPPI PaperCon’09 Conference.
Zeng, H. 2013, Polymer Adhesion, Friction, and Lubrication (1), John Wiley & Sons,
Incorporated, Somerset, US.
View publication stats

Hydrophobic Surface Sizing of Testliner: March 2019

Uploaded by

Copyright:

Available Formats

Hydrophobic Surface Sizing of Testliner: March 2019

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Hydrophobic Surface Sizing of Testliner: March 2019

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Hydrophobic surface sizing of testliner

Thesis · March 2019

Surface sizing View project

The user has requested enhancement of the downloaded file.

ÅBO AKADEMI – FACULTY OF SCIENCE AND ENGINEERING

Subject: Surface sizing of testliner

Date: 13.3.2019 Pages: 72

ÅBO AKADEMI – FAKULTETEN FÖR NATURVETENSKAPER

Ämne: Ytlimning av testliner

Stärkelselösningens torrhalt och baspapprets porositet påverkade

Skillnader i ytlimningseffekten kunde ses både mellan de olika baspappren och

Nyckelord: ytlimning, testliner, förpackning, ytlimningsmedel,

Datum: 13.3.2019 Sidantal: 72

I would further like to thank everyone working at CH-Polymers laboratory facilities in

Abstrakt .................................................................................................................... III

List of tables and figures ........................................................................................ VII

List of abbreviations ................................................................................................ IX

1. Surface size application method ........................................................................ 3

2. Surface size .......................................................................................................... 9

3. Testliner base paper ......................................................................................... 21

4. Hydrophobization and water penetration theory.......................................... 25

II EXPERIMENTAL PART ............................................................. 27

III RESULTS AND DISCUSSION .................................................... 37

7. Differences in SSA efficiency ........................................................................... 37

8. Effect of surface sizing method parameters ................................................... 39

9. Effect of base paper properties ....................................................................... 44

10. Effect of SSA properties on sizing effect ..................................................... 52

Swedish summary - Svensk sammanfattning ........................................................ 56

List of tables and figures

Table 1. Benchmark settings for surface sizing parameter testing ............................. 30

Figure 1. Pond type size press ...................................................................................... 4

AKD Alkyl ketene dimer

The focus of this thesis is hydrophobization of testliner through surface sizing. It is a

In this thesis, the goal is to increase our understanding of the hydrophobization in

The focus of this work is hydrophobization in surface sizing of testliner. In order to

1. Surface size application method

1.1. Pond type size press

Pressure penetration in the nip is hydrodynamic force-dependent. The hydrodynamic

1.2. Metered film press

The gate roll film press

Rod metered film press

Figure 2. Two-sided metered size press (Forsström, 2003)

1.3. Novel and alternative methods

Starch is a naturally occurring biopolymer which is present in a variety of plants. For

2.1.1. Starch in the paper industry

Wet end starch

Surface size starch

2.1.2. Dispersion and conversion

Thermo-chemical conversion combines thermal conversion with chemical

In enzymatic conversion, enzymes are added to the starch dispersion in a batch or

To avoid retrogradation, the temperature of dispersed (cooked) starch should be kept

Chemical modification of starch is increasingly popular, in 2009 about two-thirds of

Oxidized (anionic) starch

2.2. Sizing agent

2.2.1. Internal sizing agents

Figure 5. Sizing agent particle in emulsion (Holik, 2013)

2.2.2. Surface sizing agents (SSA)