on the internal sizing mechanisms of paper with akd and ...12738/fulltext01.pdf · paper i...
TRANSCRIPT
On the internal sizing mechanisms of paper with AKD and ASA related to surface chemistry,
wettability and friction
RAUNI SEPPÄNEN
Akademisk avhandling som med tillstånd av Kungliga Tekniska Högskolan framlägges till
offentlig granskning för avläggande av filosofie doktorsexamen fredagen den 30 november
2007 klockan 14.00 i hörsal F3, KTH, Lindstedtsvägen 26, Stockholm.
School of Chemical Science and Engineering, Department of Chemistry, Division of Surface Chemistry KTH, Royal Institute of Technology SE- 114 86 Stockholm Sweden and YKI, Institute for Surface Chemistry SE-114 86 Stockholm Sweden TRITA-CHE-Report 2007:77
ISSN 1654-1081
ISBN 978-91-7178-801-6
Copyright © 2007 by Rauni Seppänen. The following items are reprinted with permission: Paper I: Copyright © 2000 by Nordic Pulp and Paper Research Journal Paper IV: Copyright © 2004 by Journal of Pulp and Paper Science On the front cover: A water droplet on the surface of hydrophobized paper. Printed at Universitetsservice US AB, Stockholm, November 2007.
i
Abstract Paper and board are hydrophobized (sized) to control the spreading and absorption of water-
based inks and retard the absorption and edge penetration of liquid packaging by aqueous
liquids. Alkenyl ketene dimers (AKD) and alkenyl succinic anhydride (ASA) are synthetic
sizing agents that are generally used under neutral or slightly alkaline papermaking
conditions.
The overall objective of this thesis is to improve understanding of the internal sizing of paper
and board by AKD and ASA by establishing a link between the sizing mechanism on one
hand, and properties of sized papers, such as surface chemistry, wettability and friction, on the
other. Fundamental research has been conducted in parallel with more applied research on
laboratory and pilot papers. Significant effort has been expended to study the spreading
behavior of AKD. The main instrument to characterize the surface chemical composition of
AKD and ASA sized papers was X-ray photoelectron spectroscopy (XPS). By combination
with time-of-flight secondary ion mass spectrometry (ToF-SIMS) we have been able to
determine the lateral distribution and the chemical state of the sizing agent at the paper
surface. Combined with contact angle measurements using liquids with different surface
tensions, and other methods to analyze the amounts of size in paper, this has enabled us to
obtain a deeper knowledge of the sizing mechanisms of AKD and ASA.
The results indicate a definitive relationship between the redistribution of AKD at the surface
of pilot papers and the drying profile used during papermaking. However, the spreading was
not complete, as also seen on a model surface. Further spreading of AKD was shown to occur
via surface diffusion in the form of an autophobic monolayer precursor. The spreading rate
increased linearly with temperature and showed an inverse proportionality with respect to the
melting point of the AKD. This monolayer spreading is relatively slow the diffusion
coefficient being of the order 10-11 m2/s. AKD spreading was not hindered by hydrolyzed
AKD (ketone) that spread as well. Moreover, AKD spread on the surface of crystalline
calcium carbonate. In laboratory papers, the extractives present on CTMP fiber surfaces
appeared to have enhanced the spreading of AKD when the fibers were in water.
In spite of a slightly lower retention, ASA covered the surface of unfilled and PCC-filled pilot
papers to a significantly higher extent than AKD. The ASA sized papers, however
demonstrated slightly lower resistance to water. This was attributed to formation of
ii
hydrolyzed ASA products. The results obtained confirm the proposed sizing mechanism for
ASA, where the hydrolyzed ASA plays a key role. The sizing level of these papers stored
wrapped in aluminium foil at 23 °C and 50 %RH was nearly unchanged over prolonged
storage time. In contrast, the papers exposed to ambient conditions suffered from sizing loss,
most likely due to hydrolysis and migration. The reduction of the sizing degree was higher for
the AKD than ASA sized unfilled papers and the catalytic effect of PCC contributed to the
hydrolysis of AKD in PCC-filled papers.
As expected, the use of sizing agent reduced the surface energy of paper. The higher the
sizing degree of paper the lower the surface energy, and thus the higher the resistance to
wetting. This was particularly seen in the contact angles with ethylene glycol having a lower
surface tension than water.
AKD significantly decreased the friction between unfilled papers, whereas ASA had no
impact. This difference was attributed to surface chemical composition. Friction reduction for
the AKD sized papers started at the AKD coverage normally found in paper produced for low
water absorption. As expected, PCC filler increased paper-to-paper friction.
iii
Sammanfattning Papper och kartong hydrofoberas (limmas) för att kontrollera spridning och absorption av
vattenbaserade tryckfärger och hindra kantinträngning av vattenbaserade vätskor i
vätskekartong. Alkylketendimer (AKD) och alkylbärnstensyra anhydrid (ASA) är syntetiska
hydrofoberingsmedel som allmänt används under neutrala eller något alkaliska förhållanden
vid papperstillverkning.
Arbetets övergripande målsättning var att förbättra förståelse för mäldhydrofobering av
papper och kartong med AKD och ASA genom att upprätta ett samband mellan
hydrofoberingsmekanism på ena sida och ytkemi hos hydrofoberat papper och dess
vätningsförmåga och friktion på den andra sidan. Grundläggande studier parallellt med mer
tillämpade undersökningar på laboratorie- och pilotpapper har utförts. En betydande strävan
har använts för att studera spridning av AKD. Huvudinstrumentet för att karakterisera kemisk
sammansättning av ytan av pilotpapper hydrofoberade med AKD och ASA var röntgen
fotoelektron spektroskopi (XPS). Genom att kombinera det med sekundär jon
masspektrometri (ToF-SIMS) har lateral fördelning och kemiskt tillstånd av AKD och ASA
vid en yta av papper kunnat bestämmas. Kombinerat med mätningar av kontaktvinkel med
vätskor med olika ytspänning och andra metoder för att analysera halten av
hydrofoberinsgmedel i papper har gjort det möjligt att erhålla djupare kunskap om
hydrofoberingsmekanismer av AKD och ASA.
Resultaten indikerar en klar koppling mellan omfördelning av AKD på ytan av pilotpapper
och torkningsprofil vid papperstillverkning. Emellertid, spridningen var inte fullständig, vilket
var fallet även på modellytor. Fortsatt spridning av AKD visades ske som ytdiffusion i form
av ett autofobiskt monoskikt (precursor film). Spridningshastigheten ökade linjärt med
temperatur och visade omvänd proportionalitet med avseende på AKD:s smältpunkt. Denna
monoskiktspridning är relativt långsam, diffusionshastighet är i storleksordningen 10-11 m2/s.
Hydrolyserat AKD (keton) hindrade inte AKD:s spridning utan spred även den. Dessutom
spred AKD på ytan av kristallina kalciumkarbonat. I laboratoriepapper är extraktivämnen
närvarande på ytor av CTMP fiber och tycktes ha förbättrat AKD:s spridning när fibrerna var
under vatten.
iv
Trots något lägre retention täckte ASA ytan av icke-fyllda och PCC-fyllda papper till
signifikant högre grad än AKD. ASA-papperen visade dock något lägre motstånd mot vatten.
Detta var hänvisat till bildning av hydrolyserade ASA-produkter. De uppnådda resultaten
bekräftar den föreslagna hydrofoberingsmekanismen för ASA, där hydrolyserat ASA spelar
en avgörande roll. Hydrofoberingsgraden av papper lagrade inlindade i aluminiumfolie vid
23 °C och 50 RH var nästan oändrad över den förlängda lagringstiden. Som motsats
genomgick papperen som exponerats mot omgivande atmosfärsförhållanden genomgick en
minskning av hydrofoberingsgraden, troligen på grund av hydrolys och migrering. Minskning
av hydrofoberingsgraden var större för icke-fyllda papper av AKD än av ASA. PCC:s
katalytiska effekt bidrog till hydrolys av AKD i PCC-fyllda papper.
Som förväntat reducerade användning av hydrofoberingsmedel reducerade ytenergin av
papper. Ju högre hydrofoberingsgrad desto lägre var ytenergin och därmed desto högre
motstånd mot vätning. Detta sågs särskilt i kontaktvinklar med etylenglykol som har lägre
ytspänning än vatten.
AKD minskade signifikant friktionen mellan icke-fyllda papper, medan ASA inte hade
inverkan. Denna skillnad hänfördes till skillnad i ytsammansättning. Minskning av friktion för
AKD-hydrofoberade papperen påbörjades vid en sådan yttäckning av AKD som är normalt
för papper tillverkat för låg vattenabsorption. Som förväntat ökade PCC-fyllmedel friktionen
mellan papperen.
v
List of papers The thesis consists of an introductory summary and six papers listed below. The papers are
referred to in the text by their roman numerals I-VI.
Paper I Mechanisms of internal sizing by alkyl ketene dimers (AKD): The role of
the spreading monolayer precursor and autophobicity
Rauni Seppänen, Fredrik Tiberg and Marie-Pierre Valignat
Nord. Pulp Paper Res. J., 2000, 15(5), 452.
Paper II On the monolayer spreading of AKD: Influence of relative humidity,
hydrolysed AKD, spreading promotor and the type of spreading surface
Rauni Seppänen
Manuscript.
Paper III The role of pulp type and drying profile in AKD sizing: Sizing degree of
papers determined by XPS and gas chromatography
Rauni Seppänen and Göran Ström
Manuscript. To be submitted to Nord. Pulp Pap. Res. J.
Paper IV Surface Energy Characterization of AKD-Sized Papers
R. Seppänen, M. von Bahr, F. Tiberg and B. Zhmud
J. Paper Pulp Sci., 2004, 30, 3, 70.
Paper V AKD and ASA sizing of pilot papers – a comparative study. Surface
chemical composition in relation to wetting and friction properties
Rauni Seppänen
Manuscript. To be submitted to Nord. Pulp Pap. Res. J.
Paper VI Durability of the sizing degree of AKD and ASA sized pilot papers during
storage investigated by contact angle measurements and ToF-SIMS
Rauni Seppänen
Manuscript.
vi
Author’s contribution to the papers Paper I Planned and conducted the major part of experimental work, evaluated the
major part of the results and contributed to the writing. Paper II Planned and carried out the major part of experimental work, evaluated the
major part of the results and wrote the paper. Paper III Planned and conducted the major part of the experimental work, evaluated the
results and was the main contributor to the writing. Paper IV Planned and conducted the major part of the experimental work, evaluated the
results and was the main contributor to the writing. Papers V-VI Planned and conducted the part of the experimental work, evaluated the results
and wrote the paper. Papers not included in this thesis The following relevant papers dealing with internal sizing of paper are not included in this thesis: Seppänen, R. and Tiberg, F. Mechanism of paper sizing by alkyl ketene dimer (AKD). The role of the spreading monolayer precursor and autophobicity, Proc. Scientific&Technical Advances in the Internal&Surface Sizing of Paper&Board, Florence, Italy, Pira International, Surrey, UK, 1999, Paper 2. Zhmud, B., Seppänen, R. and Tiberg, F., Hydrolysis and sizing action of cellulose-reactive sizes, Proc. Scientific&Technical Advances in the Internal&Surface Sizing of Paper&Board, Prague, the Czech republic, Pira International, Surrey, UK, 2001, Paper 6. Seppänen, R. and Zhmud, B., Wettability of AKD and ASA sized papers, Proc. Scientific&Technical Advances in the Internal&Surface Sizing of Paper&Board, Graz, Austria, Pira International, Surrey, UK, 2003, Paper 11. von Bahr, M., Seppänen, R., Tiberg, F. and Zhmud, B., Dynamic Wetting of AKD-Sized Papers J. Pulp Pap. Sci. 2004, 30 (3), 70. Seppänen, R., Efficient sizing of paper and board – How does sizing affect paper properties? Proc., Intertech. Pira Sizing 2006 Conference, Stockholm, Sweden, Pira International, Surrey, UK, 2006, Workshop 4.
Seppänen, R., Surface Characterization of AKD-sized papers, In: Characterization of Lignocellulosic Materials, Chapter 5, Wiley InterScience, Hu, T. (ed.), 2007, In Print.
vii
Table of Contents 1 INTRODUCTION.........................................................................................1 2 BACKGROUND...........................................................................................3 2.1 Internal sizing of paper ....................................................................................................3 2.1.1 General features ...........................................................................................................3 2.1.2 AKD sizing ...................................................................................................................4 2.1.3 ASA sizing.....................................................................................................................7 2.2 Wettability and absorbency of paper ................................................................................9 2.3 Paper-to-paper friction...................................................................................................11
3 MATERIALS. .............................................................................................14 3.1 Model surfaces ..............................................................................................................14 3.2 Pulps .............................................................................................................................14 3.3 AKD..............................................................................................................................14 3.4 ASA ..............................................................................................................................14 3.5 PCC filler ......................................................................................................................14 3.6 Spreading promotors......................................................................................................14 3.7 Other material................................................................................................................15
4 EXPERIMENTAL TECHNIQUES .............................................................16 4.1 Surface chemical composition and roughness ................................................................16 4.2 Spreading studies...........................................................................................................19 4.3 Contact angle and water-resistance ................................................................................19 4.4 Paper-to-paper friction...................................................................................................20 4.5 Other tests .....................................................................................................................20 4.6 Paper preparation...........................................................................................................21
5 SUMMARY OF RESULTS AND DISCUSSION .......................................23 5.1 Redistribution of AKD ..................................................................................................23 5.1.1 The role of drying profile ............................................................................................23 5.1.2 Surface diffusion ........................................................................................................26 5.1.3 Vapor phase diffusion .................................................................................................32
5.2 Mechanisms of AKD and ASA sizing of papers related to their surface chemistry.........34 5.2.1 Spreading models .......................................................................................................34 5.2.2 Unfilled pilot papers ...................................................................................................36 5.2.3 PCC-filled pilot papers ...............................................................................................40 5.2.4 Durability of sizing degree..........................................................................................42 5.3 Wettability and friction of sized papers related to their surface chemistry .......................47 5.3.1 Wettability vs. size coverage ........................................................................................47 5.3.2 Friction vs. size coverage.............................................................................................49
6 CONCLUDING REMARKS. ......................................................................52 ACKNOWLEDGEMENTS..............................................................................54 REFERENCES.................................................................................................56
1
1. Introduction Hydrophobizing (sizing) of paper and board is an interesting and challenging subject, which
often deals with interdisciplinary issues, much like colloid and interface science in general.
The latter can often explain phenomena occurring in the papermaking process. A sheet of
paper composed only of cellulosic fibers is water-absorbent. However, most paper and board
grades need to be resistant to wetting and absorption of polar liquids such as water, aqueous
solutions and suspensions. Therefore papermaking components, such as fibers are rendered
hydrophobic by introducing chemical additives to the papermaking suspension (internal
sizing) or to the paper web (surface sizing). Common sizing additives are rosin, alkyl ketene
dimer (AKD) and alkenyl succinic anhydrides (ASA). Internal sizing aims at establishing a
barrier for the penetration and spreading of the liquids through the pore system of paper and
board. In addition to improving the edge penetration properties of liquid packaging, sizing
also improves paper printability by controlling ink spreading and absorption, and by reducing
the effect of aqueous fountain solution on paper strength loss.
AKD and ASA are synthetic compounds and are generally applied under neutral or slightly
alkaline papermaking conditions. Internal sizing by AKD and ASA include sub-processes
such as emulsion preparation, retention, molecular spreading and anchoring, as well as sizing
reversion (for comprehensive reviews, see Roberts 1996, Neimo 1999 or Hubbe 2006). The
main difference between AKD and ASA is the higher reactivity of ASA (Wasser 1987) that
makes it possible to achieve most of the sizing effect already on the paper machine. AKD
sizing, on the other hand, develops more slowly and may continue for several days in paper
rolls (Roberts 1996). The difference between AKD and ASA is also seen in their reactivity
with water.
It is widely accepted, although still debated that the sizing mechanism of both AKD and ASA
involves their reaction with the hydroxyl groups on cellulose. Spreading of sizing agent on
fiber and filler surfaces is particularly important when considering sizing efficiency of paper.
It enables the size to be anchored to a higher extent to the fiber and filler, and thus minimizes
the amount of size needed to achieve the desired sizing level.
At the time when this work was started there was only limited information about the
spreading mechanism of AKD. Additionally, the mechanism of ASA sizing was debated. The
overall objective of this thesis is to improve understanding of the internal sizing of paper and
2
board by AKD and ASA. Special attention has been paid to understand sizing mechanisms by
relating the surface chemical composition of sized papers to their wettability and friction
behavior. The main part of the work focused on pilot made papers having characteristics
relatively close to those of commercial papers.
The results in this work have been obtained in several projects that YKI carried out in
acknowledged co-operation with chemical suppliers, pulp and paper and packaging
companies. Therefore, my hope is that this work will help in improving the understanding of
sizing behavior and that this knowledge will be useful, especially in practical applications.
3
2. Background 2.1 Internal sizing of paper 2.1.1 General features
Size molecules are amphiphilic comprising a hydrophilic head and a hydrocarbon part, which
gives the desired hydrophobic properties. The molecular structure of alkyl ketene dimer
(AKD) and alkenyl succinic anhydride (ASA) is shown in Figure 1. The hydrophobic part of
the AKD originates from fatty acids. The more widely used AKDs are based on mixtures of
palmitic (16 C-atoms) and stearic acid (18 C-atoms) or on pure stearic acid. The commercial
ASA products are synthesized from internal straight-chained olefins with a carbon chain
length of C16-18.
OR
R
O
R= C14H29 to C18H37
OO O
R
R
R= CH3 to C16H33
Figure 1. Molecular structure of AKD (left) and ASA (right).
Size is added to the papermaking furnish as an emulsion, often stabilized by cationic starch in
order to enhance its retention on the negatively charged fibers. The high reactivity of ASA
requires emulsification on-site at the mill.
Upon sizing, polar hydroxyl groups of cellulose react or are shielded by absorbing size
moieties, which reduces paper wettability. Thus, the improved water-resistance of sized paper
can be explained by the formation of low-energy regions on fiber surfaces that inhibit
spreading and capillary penetration of water into paper. Naturally, the build-up of
hydrophobicity with the size depends on interactions between the size and different
papermaking components. Good coverage of fiber surfaces by the size and good anchorage of
the size are the most evident requirements for efficient sizing. This is essential, since
unanchored size molecules are mobile and can turn their hydrophobic part when exposed to
air, and in contact with water the unanchored molecules may turn over and expose a more
polar part.
4
It is widely accepted that sizing of paper depends essentially on the following sub-processes: • Retention of size particles at fiber and filler surfaces
• Spreading and redistribution of sizing over fiber and filler surfaces
• Anchoring of sizing agents on the surface groups on fibers and fillers
2.1.2 AKD sizing
Sizing efficiency of paper is highly dependent upon retention of the sizing agent on
papermaking components (e.g. Lindström and Söderberg 1986b, Isogai 1997, Champ and Ettl
2004, Johansson and Lindström 2004a,b, Ravnjak et al. 2007). Most of the size retained is
associated with fines and fillers due to their relatively high surface area compared to fiber. In
the case of unfilled paper furnishs, the AKD retention depends almost totally on the fines
retention. Zeno et al. (2005) showed that surface-active substances present in deinked pulp
are detrimental to AKD sizing. While showing differences due to their structures and
properties, these surface-active species have a common feature of decreasing the amount of
size able to react, rather than the amount retained. In a recent publication Sjöström et al.
(2006) evaluated the effects on AKD sizing of released organic substances in white water.
Material originating from hardwood kraft pulp was, for example found to be more detrimental
to AKD sizing than the material from the softwood kraft pulp. The possible reason was the
influence of xylan on the levels of AKD retention.
After retention on fiber and filler surfaces, a size will naturally only cover a limited area
determined by the number of emulsified particles or droplets. In a recent article Ravjnak et al.
(2007) discussed the kinetics of colloidal alkyl ketene dimer particle deposition on pulp
fibers. They calculated that the maximum amount of AKD size particles that can attach to
bleached sulfate softwood fibers for maximum surface coverage as 11.02 mg AKD /g fiber.
Spreading of sizing agent on fiber and filler surfaces is particularly important when
considering sizing efficiency of paper. It enables the size to be anchored to a higher extent to
the fiber and filler, and thus minimizes the amount of size needed to achieve the desired
sizing level. Potentially, there are at least three possible routes for redistribution of size on
fiber and filler surfaces:
• Wetting flow when the size drop spreads to an equilibrium contact angle between 180°
and 0° by surface forces.
5
• Surface diffusion of a monolayer from the foot of the macroscopic drop. The size in the
macroscopic drop does not spread on the monolayer due to surface ordering effects
referred to as autophobicity.
• Gas phase transfer of vaporized size followed by re-adsorption on paper surfaces.
A number of studies deal with the spreading of AKD. It has been proposed that the
mechanism is complete spreading leading to a monomolecular layer of AKD on the fiber
surface (Lindström and Söderberg 1986a). However, contradictory observations have been
reported. Ström et al. (1992) suggested that spreading of molten AKD on the paper surface
stops at a certain film thickness and does not continue to a monomolecular layer. The
spreading of AKD wax and a commercial AKD emulsion on model and cellulose surfaces has
been investigated and a complete spreading of AKD to a monolayer was not observed
(Garnier et al. 1998, 1999, Garnier and Godbout 2000). Seppänen and Tiberg (1999, 2000)
concluded that AKD spread over hydrophilic cellulose and silica surfaces, but that the
spreading was not complete. Further spreading was shown to occur by surface diffusion in the
form of an autophobic monolayer precursor, which grows from the foot of the AKD drop or
particle. The monolayer diffusion process was observed to be quite slow, the apparent surface
diffusion coefficient being of the order 10-11 m2/s at 45 °C. Shchukarev et al. (2003) obtained
a diffusion coefficient of the same order at 80 °C. Lidén and Tollander (2004) concluded that
extractives, particularly saturated fatty acids, impaired the sizing of AKD and ASA by
reducing their spreading during drying. Migration of AKD occurred at 80 and 105°C through
untreated paper and through paper sheets impregnated with extractives such as betulinol,
oleate or tripalmitine (Mattsson et al. 2003). The authors assumed the migration to occur via
molecular diffusion process, either as a surface diffusion or via the vapor phase.
Attention has also been focused on the redistribution of size through desorption and vapor
phase transfer leading to readsorption and thus to sizing (Davis et al. 1956, Akpabio and
Roberts 1987, Yu and Garnier 1997, Shen et al. 2001, Yu and Garnier 2002). This mechanism
may play a role at temperatures above 80 °C, higher than usual storage temperatures in the
paper rolls. However, desorption and vapor phase transfer occur, but only to a marginal extent
that does not result in sizing of paper sheets by AKD (Hutton and Shen 2005, Zhang et al.
2007). Also, capillary wicking has been proposed for the sizing mechanism of AKD (Garnier
and Godbout 2000, Shen et al. 2001).
6
AKD may undergo different reactions, such as esterification, hydrolysis and oligomerization
illustrated in Figure 2, as well as anhydride formation (Davis et al. 1956, Cates et al. 1989,
Bottorff 1994).
C
H2C
OHCR R
COO Cellulose
C
H2C
OH2CR R
CCH
HCR R
CO
O
C
O
C
CH
C
O
CH O
R
R
C
CH
R
CH
COOH
R
CH
C
O
R
O
CH2
R
Figure 2. The main reaction of AKD; with cellulose, water, and with itself to form oligomers.
The formation of a β-ketoester between AKD and cellulose hydroxyl groups is proposed to be
the primary mechanism for AKD sizing (Davis et al. 1956, Roberts and Garner 1985,
Lindström and O’Brian 1986b), see a comprehensive review by Hubbe (2006) for further
references. Non-extractable AKD by solvent extraction is considered to be covalently bound
to fibers. However, Ödberg et al. (1987) demonstrated formation of β-ketoester bonding, with
and without the cationic stabilizer present using infrared spectroscopy. It has been shown that
relatively small portions from 0.006 to 0.07 percent of bound AKD are needed to obtain
hydrophobic paper (Roberts and Garner 1985, Lindström and Söderberg 1986a, Dumas and
Evans 1986). Recently, Laitinen (2006) reports on amounts of 0.005 percent as non-
extractable in extraction with tetrahydrofuran at high temperature and pressure. The reaction
ratio and sizing degree was observed to increase with relative humidity up to 80 percent due
to the favorable exposure of AKD lactone rings for reaction (Lee and Luner 2005).
In contrast, no covalent bonds were detected (Rohringer et al. 1985) or that formation of ester
bonds in a real sizing system does not occur to any practicable extent (Isogai et al. 1992,
Taniguchi et al. 1993, Isogai 1999). The strong bond/weak bond theory (Gess and Lund 1990)
to explain development AKD sizing suggests a number of possible reactions to occur between
AKD and the elements of the sizing system, the ester bonding being a minor side reaction at
best.
7
The competing hydrolysis reaction between water and AKD is most probable in the aqueous
environment during papermaking (Marton 1990, Ravnjak et al. 2006). Increasing bicarbonate
alkalinity also promotes the sizing reaction of AKD (Lindström and Söderberg 1986d, Jiang
and Deng 2000) and the hydrolysis reaction (Seo and Cho 2005, Lindström and Glad-
Nordmark 2007).
Quantification of AKD oligomers in paper extracts has shown that about half of the retained
AKD was oligomers in commercial paper (Hardell and Woodbury 2002). Material that reacts
with AKD, such as compounds that contain primary or secondary amine groups (e.g. retention
agents and wet strength agents), organic acids and alcohols may affect AKD sizing, positively
or negatively (Laitinen 2006). Additionally, size reversion or sizing loss may occur in the AKD sized papers. Increasing
time or temperature of heat treatment of paper sheets caused a decrease in the amount of
retained AKD, although the water-resistance of the sheets was unchanged (Roberts and
Garner 1985). The authors explained the phenomenon by diffusion and migration of AKD
caused by physical redistribution. Desorption and vapor phase transfer of AKD have been
shown to occur, but only to a marginal extent that did not result in sizing of paper sheets by
AKD (Hutton and Shen 2005, Zhang et al. 2007). The evaporated components from AKD
were mainly fatty acids and hydrolysed AKD (ketone). In a previous paper by Seppänen and
Tiberg (2000) the authors suggested that AKD diffusion due to evaporation of AKD occurs to
a minor extent parallel to or after surface diffusion, thus explaining partially a desizing
phenomenon, i.e. decreased hydrophobicity of paper after long-term storage. The study was
performed at elevated temperatures of 50 °C and 90 °C. Sizing loss of AKD in the PCC-filled
alkaline papers has been explained by transformation over time of AKD to its hydrolyzed
ketone form and calcium carbonate reaction products (Bottorff 1994), which in the latter case
will decompose to ketone.
Sizing loss can also be experienced through surfactant-induced wetting enhancement (von
Bahr 2004). The surfactant would in this case enhance the paper/liquid contact area, thus
promote faster hydrolysis of ester bonds, and solubilize the hydrolyzed AKD.
2.1.3 ASA sizing
For ASA, the sizing develops more rapidly than with AKD, usually already in the paper
machine (Roberts 1996). The reaction with cellulose is the widely accepted sizing
mechanism. McCarty and Stratton (1987) showed that ester bonds could form under
8
papermaking conditions. The cellulose-ASA ester bonds are stable to the alkaline solutions
where CaCO3 filler is present (Sato and Isogai 2003). Figure 3 shows the assumed reactions
of ASA. As an anhydride, ASA reacts easily with water (Wasser 1987). The hydrolysis of
ASA accelerates with pH, time, and temperature. The hydrolysis of ASA occurs at the
oil/water boundary since ASA itself is insoluble in water. Due to its amphiphilic nature, the
ASA-acid can lower the surface tension of polar liquids, such as water, and thus also decrease
the sizing efficiency (McCarthy and Stratton 1987, Roberts and Daud 1988, Novak and Rende
1993). On the other hand, when the ASA-acid forms a complex with calcium ions, a
hydrophobic product is formed (Wasser and Brinen 1998). Addition of alum tended to
increase sizing efficiency (Scalfarotto 1985).
Figure 3. Reaction of ASA with water and formation of dicarboxylic acid (alkenyl succinic acid) and salt formation.
It is known that the conditioning of ASA sized papers under relatively high humidity helps
the sizing effect to develop. However, this should also render part of ASA towards a
hydrolysis reaction route. Later on, Nishiyama and coworkers (1996a,b, Isogai et al. 1996)
presented proof that most retained ASA in ASA-sized paper sheets actually is ASA acid, i.e.
fully hydrolyzed. At the same time, ASA acid itself, if added to the stock at the wet end, does
not produce any sizing. Thus, it was concluded that the reactive form of ASA is needed to
enable uniform distribution of ASA in the stock. These results may be explained by spreading
of ASA-acid adsorbed at the air/liquid interface (Zhmud et al. 2001) under humid conditions.
Formation of a thin film of water at the surface and adsorption of ASA-acid facilitates rapid
film expansion. The high affinity between ASA-acid and calcium ions is evident from the fact
that increasing ASA addition in pilot papers increased the amount of calcium in the outermost
9
layer of the papers (Zhmud et al. 2001). There are earlier results that support the idea that
ASA sizing occurs in quite a similar way to rosin sizing (Hatanaka et al. 1991). Thus,
anchoring of the surface active ASA acid is achieved by alum under acidic conditions
(Hatanaka et al. 1991). In a recent paper by Laitinen (2006), the results indicated that
electrostatic bonding (ionic bonding and complex formation) is the main anchoring
mechanism for ASA. No rapid hydrolysis was assumed to occur during acidic acetone water
extraction.
Lindfors et al. (2005) have studied spreading kinetics and adhesion of ASA on hydrophilic
and hydrophobic SiO2 surfaces. They observed that ASA spread on a hydrophobic but not on
a hydrophilic surface immersed in water. At high pH, effects on the surface tensions due to
reactions of ASA (hydrolysis) were observed. Moreover, Ca2+ ions affected clearly the
spreading of ASA on the hydrophobic surface, where non-wetting behavior was observed.
It has been shown that ASA sizing can occur via vapor phase diffusion (Yu and Garnier 2002,
Zhang et al. 2007).
ASA sized paper sheets may loose their sizing efficiency via auto-oxidation at α-carbons of
double bonds in ASA molecules followed by introduction of some hydrophilic groups in the
molecules by oxygen atoms (Sato et al. 2000). Additionally, size reversion through hydrolysis
may occur (McCarty and Stratton 1987).
2.2 Wettability and absorbency of paper Capillary penetration by a liquid is retarded by increasing the hydrophobicity of the paper.
Negative capillary pressures arise for contact angles larger than 90°, which means an effective
resistance against liquid penetration according to the Laplace equation for ideal systems (i.e.
parallel cylindrical pores)
!
"P =2# cos$
r
where ΔP is the pressure difference across the liquid meniscus, the γ is the surface tension of
the liquid, θ the contact angle, and r the pore radius. Figure 4 illustrates the capillary
penetration into unsized and sized paper.
10
Figure 4. Schematic illustration of capillary penetration into unsized paper (left) and sized paper (right).
Thus, the advancing contact angle of well-sized paper should therefore be larger than 90°. As
is illustrated in Figure 5, this is sufficient for hindering spontaneous uptake of water by the
paper. This was also seen when comparing contact angle measurements with penetration rates
of water into AKD sized laboratory papers in our laboratory (von Bahr et al. 2004).
Figure 5. Time dependence of the dynamic contact angle, volume and base line of water droplets determined for an unfilled paper sheet sized by AKD.
Contact angle of water is a common measure of the hydrophobicity of a surface as illustrated
in Figure 6. However, paper substrates are not compliant with the assumptions of Young’s
equation. Paper surfaces are porous and chemically and physically heterogeneous. A liquid
drop placed on a porous paper does not only spread on the surface but also penetrates into the
paper, thus modifying its wetting properties.
Figure 6. A sessile drop applied to the surface of hydrophobic paper.
11
To consistently measure the contact angle for papers is a significant challenge due to
spreading, absorption, dissolution and swelling which can occur on similar timescales.
Despite these obstacles there is now a relatively good qualitative understanding of the factors
affecting the determination of the equilibrium contact angle on paper and the dynamics of
wetting, for example on sized paper (Shen et al. 2000, Wågberg and Westerlind 2000,
Modaressi and Garnier 2002, von Bahr et al. 2004, Zeno et al. 2005) and unsized paper (Shen
et al. 2000, Wågberg and Westerlind 2000). During the last few years there has been a
renewed interest in wetting, especially on the wetting of rough surfaces (e.g. Werner et al.
2005, Meiron et al. 2004, Bico et al. 2002) based on theoretical foundations of the Wenzel
(1936) and Cassie and Baxter (1944) relations. These theories attempt to explain the effect of
surface roughness, which increases the surface area of a hydrophobic material, and the effect
of geometrically enhancing hydrophobicity (homogeneous wetting). On the other hand, air
can remain trapped below the liquid drop, which also leads to a superhydrophobic behavior,
since the drop rests partially on air (heterogeneous wetting).
Surface energy
Good and van Oss (1992) developed a theoretical methodology allowing the determination of
the surface energy components associated with specific molecular interactions by carrying out
experiments with several test liquids involving such interactions. The most useful is the two-
term surface energy decomposion into disperse (due to van der Waals interactions) and polar
(due to hydrogen bonds and dipole-dipole interactions) parts. A renewed interest to use three
test liquids to determine surface energies of a material is observed (Della Volpe and Siboni
2000, van Oss 2002, Della Volpe et al. 2003, Zenkiewicz 2007).
As applied to cellulosic materials and paper in particular, the surface energy is very sensitive
to changes in surface speciation surface treatments or modifications, for instance sizing.
2.3 Paper-to-paper friction Other paper properties such as adhesion and friction in addition to wetting may be changed
when the surface chemistry of paper is altered with a sizing agent. Friction plays a major role
in the handling, use and converting of paper, board and packing material. For example, high
friction is essential for controlling the paper roll functionality and to obtain good runnability
of the paper in a printing press (Back 1991). In the case of linerboard, low friction gives
problems like telescoping in roll handling. Copy paper performance seems to be affected in
12
quite complex ways that are related to friction coefficients, the nature of the paper surface and
fiber orientation. The reliability of copiers and printers also depends on a constant friction of
the paper, especially during paper feeding.
The primary properties of interest in connection to friction and adhesion are surface energy,
polarity, chemical homogeneity, surface roughness and topography, surface morphology,
hardness and rheology. There are several friction mechanisms, but the main factor is
adhesional properties between the interacting surfaces. Other mechanisms build on wear
particles entrapped at the interface. It should be noted that friction is always accompanied by
wear. Friction in the microscopic level has been explained to depend on the true area of
surface-to-surface contact (Israelachvili et al. 1994). In systems other than paper, surface
roughness has been seen to play a role. It has been suggested that paper friction depends on
structural irregularities that can be related to micro-surface roughness (Withiam 1991). On the
other hand, the conclusions regarding the effect of surface roughness on paper-to-paper
friction are often contradictory probably due to varying conditions during paper making
process and friction test. It has been stated that effects of surface roughness can be neglected
as long as surface strength is not changed and the testing is done on the same side of paper
(Back 2002). In uncoated paper grades, in addition to factors such as pulp type and sizing
agent, the filler also affects surface properties, such as chemistry, structure and strength.
Kaolin and other flat particles reduce friction, whereas synthetic precipitated silica and
silicates (Withiam 1991), as well as precipitated calcium carbonate (Karademir and Hoyland
2003) increase friction. Colloidal silica, for example, when present in the paper surface,
provides an interlocking between the surfaces on a microlevel (Payne 1989).
AKD is known to reduce the frictional properties of paper (Inoue et al. 1990, Hoyland and
Neill 2001, Karademir and Hoyland 2003, Karademir et al. 2004). It has been concluded that
unbound AKD is responsible for lower paper friction (Marton 1990, Karademir and Hoyland
2003). AKD molecules with longer side chains have been reported to reduce the paper friction
more than those with shorter side chains (Hoyland and Neill 2001). Inoue et al. (1990)
showed a relationship between friction and surface free energy of paper after removing sizing
agents and waxes from the surface. A higher surface free energy corresponded to a high
friction. Friction coefficients for the cellulose-AKD system were observed to be lower than
for cellulose-silica system when measured using a colloidal AFM technique (Bogdanovic et
al. 2001), a result that correlates well with those measured between two paper surfaces.
13
Moreover, the results showed a clear relation between friction and surface free energy. The
origin of friction originated from the polar character of the surface. The humidity was also
observed to play an important role. The higher the humidity the higher was the friction as also
observed to be valid for paper-to-paper friction (McDonald 1996).
It has been reported that ASA increases the friction coefficients at low addition levels
(Hoyland et al. 2000). An observed decrease in friction at the higher ASA addition levels was
attributed to the influence of increased addition of starch or polymer, and not due to ASA.
14
3. Materials The materials used and the suppliers of these were different in different studies. This section
compiles a short list. The reader is referred to the included papers for details.
3.1 Model surfaces Silica (oxidized Si-wafers), kaolin pigment and calcium carbonate; calcium carbonate
pigment slices of freshly cleaved crystalline or calcium carbonate (Paper I and II).
3.2 Pulps
Laboratory papers
ECF bleached birch/pine kraft (1:1), after disintegration, the pulp was beaten to 35 SR in an
Escher-Wyss mill (Paper I and III). Never-dried chemothermomechanical pulp CTMP (Paper
III).
Pilot papers
Bleached kraft pulp (BKP), 50 % hardwood and 50 % softwood refined to 35-39 SR (Paper
III). Bleached ECF 70% HW and 30% SW pulp refined to 27 SR (Papers V-VI).
3.3 AKD
In Papers I-II, without any additives:
AKD in liquid form and AKD waxes having melting points of approx. 50 °C and 62 °C. An emulsion of AKD(m.p. 51 °C) was used and also a hydrolyzed AKD(m.p. 51 °C), i.e. ketone. An AKD(m.p. 51 °C) emulsion was used (Paper I, III and IV).
An AKD(m.p. 55-60 °C) emulsion was used (Papers V-VI).
3.4 ASA
An ASA-C18 emulsion was used (Papers V-VI).
3.5 PCC filler High-opacity calcitic scalenohedral precipitated calcium carbonate (PCC) was used (Papers
V-VI).
3.6 Spreading promoters Four well-defined glycerides including monolaurin (m.p. 63 °C), dilaurin (56.6 °C),
monopalmitin (77 °C) and dipalmitin (72-74 °C) were tested.
15
3.7 Other material Retention aid. 0.0125 wt% cationic retention polymer in case of unfilled papers and 0.05 wt%
refined bentonite together with 0.0125 wt% cationic retention polymer in case of filled papers
(Paper V-VI).
Alum. 0.5 % alum Al2(SO4)3.14H2O, based on the ASA amount was added in the case of ASA
sized papers (Paper V-VI).
16
4. Experimental techniques 4.1 Surface chemical composition and roughness X-ray Photoelectron Spectroscopy (XPS)
XPS also known as ESCA (Electron Spectroscopy for Chemical Analysis) is a surface
sensitive analysis technique that has widely been used in many applications, for example to
characterize original fibers, fibers modifications, such as bleaching processes and sizing and
final paper product. The principle of the technique is to use X-rays to cause the emission of
electrons from the surface of the substrate of interest, and then measure the kinetic energy of
the emitted photoelectrons as illustrated in Figure 5. With this information, the energy of the
surface bound electrons can be calculated and related to chemical state of bonding.
Figure 5. Principle of XPS measurement.
Using XPS, all elements, except H and He, can be detected and in addition to quantitative
(surface concentration) also qualitative information about different oxidation states, for
instance for carbon bonded to oxygen can be obtained.
The surface chemical composition of papers was determined with a Kratos AXIS HS x-ray
photoelectron spectrometer (Kratos Analytical, UK) using a take-off angle 90° and
monochromatic Al Kα X-ray source (1486.6 eV). The mean analysis depth for paper and
polymeric material is about 10 nm. Analyses were made on an area approximately 1 mm2 at
two different locations on each sample. Surface elemental concentrations and O/C ratios were
calculated from the survey spectra. The sensitivity factors used for quantification were 0.25
for C1s, 0.66 for O 1s and 1.58 for Ca 2p.
The relative amounts of carbons with different bonds to oxygen were determined from high-
resolution carbon C 1s spectra using a curve-fitting program from the spectrometer
manufacturer. The chemical shifts relative to C-C (C1) used in the convolution were 286.8 eV
17
for C-O (C2), 288.2 eV for O-C-O or C=O (C3) and 289.4 eV for O=C-O (C4), 290.1 for
carbonate carbon (C5). Using surface sensitive x-ray photoelectron spectroscopy (XPS) we
have analysed the aliphatic C-C carbon peak of the high-resolution C1s signal. This is
possible, since an unsized paper made from bleached pulp fibers contains only small amounts
of unoxidized carbon, while this type of carbon is the major component of AKD (Ström et al.
1992). Since the pioneering work by Dorris and Gray (1978), XPS has been widely used to
study lignocellulosic surfaces. For example, unbleached kraft pulp fibers (Laine and Stenius
1994, Fardim et al. 2006), different mechanical pulp fibers (Koljonen et al. 2003, 2004),
chemically modified lignocellulosic fibers (Gellerstedt and Gatenholm 1999, Freire et al.
2006, Östenson et al. 2006), plasma-treated surfaces (Carlsson and Ström 1995, Toriz et al.
2004), and AKD-sized papers (Ström et al. 1992, Shen et al. 2000, Shchukarev et al. 2003).
Figure 6 exemplifies high-resolution XPS C1 1s signal spectra for unsized and AKD-sized
paper.
Figure 6. High-resolution XPS C1 1s signal spectra for unsized (left) and sized paper (right).
The surface sensitivity of XPS can be further enhanced varying the angle of emission of the
analyzed electrons. Figure 7 displays how the analysis depth reduces with the reducing
photoelectron emission angles, i.e. take-off angles. In Paper III, the papers were studied in
terms of C1 carbon at the take-off angles of 90°, 55°, 40° and 25°. The AKD coverage, given
as ΔC1, was also estimated (Paper III).
18
Figure 7. The XPS photoelectron emission angles
In order to increase understanding about the distribution of ASA and AKD in the outermost
surface of the sized papers we also performed the spectral background analysis of the O 1s
signal using Tougaard’s approach (Tougaard and Sigmund 1982, Tougaard and Ignatiev
1983) according to Johansson et al. (2004). Using this approach, it has been shown that lignin
(Johansson et al. 2004) covers the fiber surface as clusters, whereas extractives form a thin
film on fiber surfaces (Österberg et al. 2005).
Time-of-flight secondary ion spectrometry (ToF-SIMS)
SIMS is the most surface sensitive spectroscopic technique available today with an analysis
depth of a couple of nanometers. It uses a pulsed primary ion beam to desorb and ionize
species from a sample surface. The resulting secondary ions are accelerated into a mass
spectrometer, where they are mass analyzed by measuring their time-of flight from the sample
surface to the detector. An image is generated by rastering a finely focused beam across the
sample surface. Due to the parallel detection nature of ToF-SIMS, the entire mass spectrum is
acquired from every pixel in the image. The mass spectrum (negative and/or positive) and the
secondary ion images are then used to determine the composition and distribution of sample
surface constituents.
Surface roughness
Surface roughness was determined by a ZYGO profilometer (New View 5010 from ZYGO).
The principle is a non-contact three-dimensional metrology based on white light
interferometry. When white light is used to illuminate both the sample and an internal
reference surface, the reflected light from these two sources will produce a standing wave
pattern due to interference of the electromagnetic waves.
19
4.2 Spreading studies Laterally resolved ellipsometry
Monolayer spreading was followed by spatially resolved ellipsometry. The technique gives
precise information about the local thickness, as long as the local slope of the drop interface is
not to steep. The instrument used is a single-wavelength, polarization-modulated ellipsometer
with a focused measurement area (lateral resolution ≈ 30 mm), which operates at the Brewster
angle. The optical thickness resolution is 0.2 Å with a time constant of 100 ms for detection.
The experimental set-up is described elsewhere (Heslot et al. 1989). The AKD deposition
procedure and the storage conditions used between imaging are described below.
Breath pattern method
A small droplet or particle of the AKD-wax with a diameter of about 1 mm was applied to the
surface of the substrate. The sample was put in a small chamber and kept at a constant
temperature and humidity (50 percent relative humidity) for 1 to 30 days. The temperatures
used ranged from 23 °C to 90 °C. The hydrophobic spreading area was determined by
measurements of the hydrophobic area of surfaces exposed to a water vapor (the so-called
breath pattern method). The spreading diameter was defined as the diameter of the
hydrophobic area minus the macroscopic drop/particle diameter. The diameter of the
drop/particle remained unchanged during the experiment showing that bulk spreading did not
occur.
4.3 Contact angle and water-resistance Apparent advancing contact angle
Dynamic contact angles of test liquid droplets were determined from side-images of the
droplet profile, which was monitored as a function of time with a Dynamic Absorption Tester
Fibro 1100 DAT (Fibro Systems AB, Sweden). The instrument applies a liquid droplet to the
surface while a high-speed video camera captures images as the drop spreads and/or is
absorbed. Liquid is automatically pumped out from a syringe and the droplet is formed at a tip
of teflon tube hence avoiding liquid sticking to the tip. Images of the drop were captured
every 20 milli-seconds. After the measurement, images are evaluated by image analysis in
terms of drop volume, height, base diameter/area and contact angle. The test liquids
comprised distilled MilliQ Plus water, ethylene glycol and diiodomethane. The drop volume
of water and ethylene glycol used in the measurements was 4 ml and diiodomethane 2 ml.
Measurements were performed at five different locations on each sample in the machine
20
direction. The measurements were performed at a relative humidity of 50 ± 3% and
temperature of 25 ± 1°C.
The water-resistance of the paper sheets was determined using the Cobb60 standard test
(SCAN-P 12:64).
Surface energy
The calculations of surface energy components are based on the surface tension component
theory including Lewis acid-base components (Good and van Oss 1992). The total surface
energy, and its apolar and polar components of pilot papers were calculated using steady state
contact angles obtained by extrapolating the linear part of the contact angle curve towards
zero time. The steady state contact angles are always slightly lower than the first initial ones.
4.4 Paper-to-paper friction Paper-paper friction was measured according to the standard procedure ISO 15359 with the
horizontal friction measurement technique Amontons II from Mµ measurements, Inc.,
USA/TJT Teknik AB, Sweden. The apparatus is described in more detail elsewhere
(Johansson et al. 1998, Garoff 2002). The Amontons II friction tester is based on the
horizontal-plane principle, which means that one surface is moved relative to the other on a
horizontal plane. On each sample, three slidings were performed, yielding three static and
three kinetic coefficients of friction. However, only the third static coefficient of friction (S3)
is reported.
4.5 Other tests The melting point for AKD, hydolyzed AKD and their mixture was determined by differential
scanning calorimeter DSC821 instrument from Mettler Toledo.
Amount of size in paper
The extractable and non-extractable amounts of AKD in laboratory papers were determined
using a capillary gas chromatographic procedure (Dart and McCalley 1990). The extraction
solvent was iso-octane.
The extractable and non-extractable amounts of AKD and ASA in pilot papers using
tetrahydrofurane as a solvent and LC-MS instrumentation according to Laitinen (2006)
(Papers V-VI).
21
Roughness, thickness, density and z-strength
PPS-roughness, thickness, density and z-strength of pilot papers were determined according
to standard methods (Paper V-VI). The surface roughness was also determined using white-
light interferometry (Paper V).
4.6 Paper preparation Laboratory paper sheets
Laboratory papers sheets were formed in a standard Finnish laboratory sheet former,
according to SCAN 26:76, except that the grammage was 80 g/m2 (Papers III-IV).
Pilot papers
The pilot papers of a grammage of approximately 70 g/m2 were prepared on a pilot
Fourdrinier paper machine at a speed of 5 m/min at the Eka Chemicals former R&D center in
Maastricht, the Netherlands (Paper III). Water temperature was 20-21 °C, water hardness 4
dH and pH 7.8. The concentration in the headbox was 0.54 %. The contact time between
AKD and the pulp suspension was 15 sec. The drying section comprised six steam-heated
cylinders. The temperature of the six drying cylinders was adjusted as in Table 1.
Table 1. Drying profiles and the temperature of the drying cylinders.
Drying profile Temperature of drying cylinders
Cylinder no 1 2 3 4 5 6
Low-high 50 86 91 84 110 120
High-low 106 112 112 75 80 75
Medium-low 92 91 91 69 77 86
Pilot papers with a grammage of 80 g/m2 were prepared on a Fourdrinier pilot paper machine
at the former Hercules European Research Center, Barneveld, the Netherlands (Papers V-VI).
The machine speed was 12 m/min. Drying was performed using steam-heated drying
cylinders, at max. 105 °C. The papers were made at pH 8.0. Figure 8 displays a schematic
flow of the pilot paper machine.
22
Figure 8. Schematic flow of the pilot paper machine used to prepare paper samples studied in Papers V-VI.
23
5. Summary of results and discussion Instead of summarizing the papers sequentially I will below discuss the results divided in
different sections relating to particular aspects of the work. Firstly, the summary surveys the
work completed on redistribution of AKD, including mechanisms of how AKD may spread,
migrate or evaporate after being adsorbed onto fiber surfaces in the wet-end of a paper
machine. Secondly, mechanisms of AKD and ASA sizing of paper related to its surface
chemistry are discussed. The results are based primarily on a comparative investigation of
pilot papers sized with ASA or AKD and prepared without filler and with participated
calcium carbonate (PCC) as filler in the same conditions. The sizing level of these papers
stored either sealed in foil or exposed to ambient conditions was also evaluated over a
prolonged period of time. Finally, the summary concentrates on the relationship between the
surface chemistry of a sized paper and its wettability and friction.
5.1 Redistribution of AKD Spreading of sizing agents on fiber and filler surfaces is particularly important when
considering the sizing efficiency of paper. It enables the size to anchor to a greater extent to
surface groups on the fibers and fillers, and thus minimize the amount of size needed to
achieve the desired sizing level. Also, unwanted side effects are minimized. Redistribution
and anchoring of retained size droplets or particles occur mainly in the press and dryer
sections of a paper machine. A high drying temperature promotes the sizing efficiency of
AKD as stressed by many authors (Roberts and Garner 1985, Lindström and Söderberg 1986,
Isogai 1999). However, it is well known that AKD sizing develops slowly and may continue
for several days, whereas ASA sizing occurs more rapidly, usually already on the paper
machine (Roberts and Garner 1985).
5.1.1 Role of drying profile on spreading of AKD
In Paper III, a study involving a systematic variation of the temperature was performed. It was
concluded that the drying temperature appears to be crucial for the redistribution of AKD over
the fiber surfaces. As calculated in paper III, ΔC1 is proportional to the degree of surface
coverage of pulp fibers by AKD estimated using the model proposed by Ström et al. (1992).
Figure 9 shows ΔC1 values obtained at different take-off angles, i.e. angles used to collect
emitted photoelectrons. It is clear that AKD covers a larger surface area when sized pilot
paper was exposed to high temperature. This is strong evidence for spreading of AKD in the
24
drier section of a paper machine as well as in an oven in the laboratory. It was concluded that
AKD spread and covered a large surface area and that the AKD patches were thinner than the
sampling depth at the take-off angle of 90 degrees. The closed circles show the same sheet
collected as a wet sample from the paper web before drying and dried at room temperature. In
this case AKD is still in the form of dispersed particles, thus the projected area of
hydrocarbon-rich spots is small compared to the overall area covered by the X-ray probe, and
therefore, much smaller values of surface coverage are obtained. Bound water molecules
cover the cellulose surface at solid contents 70-75 percent (Maloney et al. 1998). At this stage
the temperature of the paper sheet is 70-80 °C, clearly above the melting point of the AKD
(approx. 51 °C) used in this study.
0
5
10
15
20
0 20 40 60 80 100
High-to-low profile, max 112°C on paper machine22°C for 22 hrs and curing at 105°C for 15 min22°C for 22 hrs
! C
1 (
%)
Take-off angle in XPS (degree) Figure 9. ΔC1 plotted as a function of take-off angle for a pilot made paper sized by AKD(51°C) and dried in different conditions; (❍) paper dried on machine using a low-to-high drying profile with a maximum temperature of 120°C, (◆) the same sample dried at 22 °C for 24 hrs and then cured in an oven at 105 °C for 15 minutes and (●) a wet sample dried at 22 °C for 24 hrs.
The main objective of the pilot paper trial study in Paper III was to examine the role of drying
profiles on AKD sizing, in particular on the spreading of AKD. The papers were dried using
high-to-low, low-to-high and medium-to-low temperature profiles. The average temperature
of each profile was above 90 °C. This means that AKD was in liquid form at all the drying
profiles applied and thus should easily have spread over fiber surfaces. Drying of papers using
the high-to-low or low-to-high profile resulted in a higher AKD coverage of the papers
compared to the papers dried at the medium-to-low drying profile, Figure 10. This was clearly
evident in the amounts of non-extractable AKD in the papers and suggests that the spreading
of AKD on paper dried on the paper machine with the high-to-low profile can be fast at a high
temperature. This does not necessarily mean that the spreading was complete. For instance,
25
further curing of two papers dried at the low-to-high profile clearly increased their size
coverage.
0
5
10
15
20
0,04 0,06 0,08 0,1 0,12 0,14
High-to-low
Low-to-high
Medium-to-low
! C
1 (
%)
Added amount of AKD (%)
a)
R = 0.999
R = 0.974
R = 0.999
0
5
10
15
20
0,04 0,06 0,08 0,1 0,12 0,14
High-to-low
Low-to-high
Medium-to-low
! C
1 (
%)
aft
er
ace
ton
e-e
xtr
actio
n
Added amount of AKD (%)
b)
R = 0.964
R = 1.000
R = 0.998
Figure 10. ΔC1 before (a) and after extraction by acetone (b) versus added amount of AKD. The take-off angle of XPS was 90 degrees. Drying profile high-to-low (max temperature 120 °C), low-to-high (max temperature 112 °C) and medium-to-low (max temperature 92 °C).
Figure 11 demonstrates that the water-resistance of the papers improves with increasing size
coverage. The highest coverage was obtained for the paper dried using the high-to-low
profile. It also shows that the surface coverage is a determining factor, since the experimental
points fall almost on the same curve regardless of drying profile.
20
40
60
80
100
0 5 10 15 20
High-to-low
Low-to-high
Medium-to-low
Co
bb 6
0 (
g/m
2)
! C1 (%) Figure 11. Influence of drying profiles on water-resistance tested by Cobb60 and plotted against the AKD coverage given as ΔC1.
26
The results above indicate that complete spreading is not necessarily achieved in the dryer
section of the paper machine. However, the common view has for many years been that AKD
after retention spreads in a wetting spreading fashion to ultimately form a thin mono-
molecular film, i.e. a continuous decrease in contact angle from 180° to 0° driven by capillary
forces. This is not the case as was shown by Garnier et al. (1998, 1999) and also is clearly
seen in Figure 5 (Paper I). It is apparent that AKD only partially wets hydrophilic solid
surfaces and that a significant amount of AKD remains in the hemi-spherical cap also after
the wetting spreading process being complete.
Figure 12. Dynamic contact angle evolution versus time for a macroscopic AKD (≈ 4 µl) drop (m.p. < 10 °C) after being put in contact with a cellulose surface. The inset shows the contact angle at the steady state conditions, cf. Figure 1 (where the title of the y-axis is not correct and the substrate is cellulose, not silica) in Paper I.
5.1.2 Surface diffusion
In Paper I it is also shown how after initial wetting, AKD spread further on silica by surface
diffusion in the form of an autophobic monolayer precursor. This spreading was visualized by
means of spatially resolved ellipsometry and the breath pattern technique. Figure 13
demonstrates how an approximately 1 nm thick monolayer grows out from the foot of a
micro-drop of liquid AKD put in contact with a silica surface. The latter was used as a model
for cellulose as the ellipsometry requires smooth surfaces. This type of spreading has proven
to be typical for other amphilic molecules too, for instance for trisiloxane surfactant (Tiberg
and Cazabat 1994a,b).
27
a
v = 0
v ≠ 0
distance / mm
0
10
20
30
-4 0-2 42
1 day2 days
12 days
Figure 13. Laterally resolved ellipsometry profile showing an AKD precursor that spreads out from the foot of a macroscopic liquid AKD drop, which has been put in contact with a silica surface. The inset shows schematically the spreading of an autophobic precursor from the foot of the AKD droplet.
The spreading distance increased proportional to the square root of time. This linear
relationship is typical for monolayer spreading, observed earlier for trisiloxane surfactant
monolayer spreading on solid surfaces. The effective monolayer diffusion coefficient is of the
order 10-11 m2/s, as calculated from the data in Figure 14. This type of spreading of AKD has
been confirmed by other researchers (Shen et al. 2001, Shchukarev et al. 2003) and given as
the reason for the migration of AKD in paper (Johansson and Lindström 2004a).
Paper II shows that ASA spread by surface diffusion too. In the case of ASA, humidity
appears to play an important role.
0
0.5
1
1.5
2
2.5
3
3.5
4
0 2 4 6 8 10 12 14
monola
yer
dia
met
er /
mm
!time / !hours Figure 14. Spreading diameter as determined by the breath-pattern technique of the autophobic AKD monolayer on silica versus the square root of the storage time. The ambient temperature was constant ≈50 °C and the AKD melting point about 51 °C. The dashed line is drawn to guide the eye.
The spreading of a droplet of commercial AKD emulsion on silica resulted in the same
spreading behavior as for the AKD(m.p. 51 °C) wax. It was therefore concluded that
28
emulsifiers, commonly an anionic surfactant used together with a cationic starch to emulsify
AKD do not affect monolayer spreading.
The above-mentioned spreading behavior was also observed on fiber surfaces. Figure 15
shows the spreading of AKD on a sheet of bleached and unbleached kraft pulp at an ambient
temperature below the AKD melting point. The low temperature was chosen to ensure that
capillary forces did not aid spreading into the porous structure. Compared to the AKD
spreading on the silica surface, the spreading diameter on the sheet of bleached kraft pulp is
clearly lower. However, taking the rougher surface of the sheet into account the real spreading
distance may in fact be approximately equivalent. In contrast, the spreading of AKD on a
sheet of unbleached pulp was somewhat delayed and very irregular. This probably reflects the
chemical heterogeneity of the unbleached paper and the fact that the surface is somewhat
hydrophobic due to extractable compounds that results in a lower surface energy than for the
bleached sheet. This changes the interfacial tension between the spreading monolayer and the
paper sheets and thus wood resin components interfere with the monolayer spreading of
AKD. It may well be that the arrangement of AKD molecules on the unbleached sheet differs
somewhat from that on the bleached sheet.
0
0,5
1
1,5
2
2,5
3
0 5 10 15
Bleached pulp
Unbleached pulp
Sp
rea
din
g a
rea
(m
m)
!Time ( !hrs)
Figure 15. Spreading diameter versus square-root of the storage time as determined by the breath-pattern technique for an AKD monolayer spreading on a laboratory paper sheet made of bleached kraft pulp. The ambient temperature was constant 45 °C and the AKD melting point about 51 °C.
Figure 16 demonstrates that the monolayer spreading of AKD depends rather strongly on both
the storage temperature and the melting point of the size. The spreading rate increased
linearly with the temperature and showed an inverse proportionality with respect to the
melting point of the AKD. The spreading was relatively slow at all temperatures, in particular
at low ambient storage temperatures.
29
0
1
2
3
4
5
6
0 20 40 60 80 100
MP<10°C
MP=50°C
MP=62°C
monola
yer
dia
met
er2
4 h
. /
mm
temperature / °C Figure 16. Influence of ambient temperature and AKD melting point on the spreading diameter of the autophobic monolayer after 24 hours spreading of different AKD’s on silica. The spreading diameter was determined by the breath-pattern technique.
Effect of hydrolyzed AKD on monolayer spreading
AKD is known to react with water forming a ketone, and although the hydrolysis rate is
relatively slow it increases with increasing temperature and pH (Akpabio and Roberts 1987,
Marton 1990, Seo and Sho 2005). Marton (1990) argued that there is always some ketone
present during papermaking as also reported by others (Seo et al. 2005, Lindström and Glad-
Nordmark 2007). It has been claimed (Shen et al. 2002) that the ketone having a higher
melting point than AKD hinders the spreading of AKD. Thus, a question arises as to whether
the ketone affects the surface diffusion of AKD. Mixtures of AKD and its hydrolysis product
were prepared and melting points determined. The differential scanning calorimeter (DSC)
curves for AKD and its hydrolysed form showed that the melting point of the ketone was
approx. 78 °C, whereas it was 51-52 °C for the AKD (Paper II). Figure 17 demonstrates that
the ketone spread in a similar fashion as AKD on silica. The spreading rates of the mixtures
are nearly identical for the ketone and the AKD.
30
0
1
2
3
4
5
20 40 60 80 100
AKD(m.p. 51 °C)
AKD(51) / ketone 4:1
AKD(51) / ketone 1:1
ketone
Sp
rea
din
g d
iam
ete
r 24 h
ours (m
m)
Temperature (°C)
Figure 17. The spreading diameter at 24 hours for AKD, hydrolysis product of AKD, i.e. ketone and their mixtures on silica as a function of temperature. The experiments were performed at 50 °C and 50 %RH using the breath pattern method. The ratio between AKD and ketone in the AKD/ketone mixture varied from 4:1 to 1:1.
Spreading of AKD on calcium carbonate
Calcium carbonate and clay are often used in papermaking as fillers. Spreading of AKD on
these surfaces is also wanted in order to obtain efficient sizing of paper. Only a minor
spreading of the AKD(m.p. 51 °C) on the both chalk and kaolin surfaces occurred at 46 °C
and 50 percent relative humidity, although these surfaces are quite hydrophilic, thus high
energy surfaces. It was concluded that it was due to physical and chemical heterogeneity of
these surfaces. However, AKD was observed to spread from the foot of an AKD wax particle
on the surface on freshly cleaved calcium carbonate single crystal. The spreading was
confirmed by the breath pattern technique and ToF-SIMS analysis that showed typical ion
mass signals for AKD and ketone at the spreading frontier (Paper II). Promoted spreading
A complete spreading of AKD is considered to be ideal in terms of sizing efficiency. In that
view a possible way to enhance the spreading of AKD was explored. A number of glycerides,
with physicochemical properties similar to those of AKD were evaluated in terms of the
spreading rate of AKD mixtures (Paper II). The mixture of AKD/monolaurin showed a
significant increase in spreading rate as observed using the simple but indirect breath pattern
technique. This effect was particularly noticeable for solid AKD having a melting point of
approx. 51 °C as shown in Figure 18. In contrast, the influence of a diglyceride, dilaurin, was
marginal.
31
0 2 4 6 8 10
Ref. AKD(50) wax
Monolaurin
Dilaurin
Monopalmitin
Dipalmitin
Spreading diameter (mm)
Figure 18. The spreading diameter at 48 hours for solid AKD and for 4:1 mixtures of solid AKD with the glycerides investigated. The spreading was studied at 50 °C and 50 %RH. Two measurements were performed with each system.
Effect of pulp type on spreading
Unbleached kraft pulp contains more hydrophobic substances (evidenced as a C1 carbon) than
bleached kraft pulp (Paper II). As seen in Figure 19 the spreading of AKD in air at 50 percent
relative humidity was observed to be slightly faster on the latter surface compared to the
former. Furthermore, the spreading results of the unbleached sheet showed high scatter,
probably due to heterogeneous surface chemistry. From the thermodynamic point of view, the
situation should be reversed in an aqueous environment. Thus, spreading wetting of AKD on
wet fiber may also be a possibility. The spreading coefficient (S) for this system is expressed
as:
S = γFW – γFA – γAW
Where γFW is the interfacial tension between fiber and water, γFA is the interfacial tension
between fiber and AKD and γAW is the interfacial tension between AKD and water. A high
spreading coefficient promotes spreading. For hydrophilic bleached pulp surfaces the
spreading coefficient will be much below zero since γFW is small while γFA and γAW are large.
The negative value is due to strong (hydrogen-) bonds forming between water and cellulose.
In comparison, for the unbleached sheet γFW is high and γFA is small. This is exemplified in
Figure 19 using laboratory paper sheets made of bleached kraft pulp (BKP) and CTMP (Paper
III). The analysis depth in XPS is typically 5-6 nm at the take-off angle 90 degrees. In this
32
study the analysis depth was decreased approximately half by reducing the take-off angle to
25 degrees. At the high take-off angle the CTMP sheets display lower ΔC1 values than the
BKP sheets, whereas they coincide at the lower take-off angle, indicating a similar degree of
coverage. This suggests that the AKD patches are thinner on CTMP than on BKP sheet
surface. Furthermore, it is noteworthy that the non-extractable amounts calculated on the
retained amounts of AKD were slightly higher in the CTMP sheets, the average being approx.
61 percent, whereas it is approx. 53 percent for the sheets of BKP. Obviously, the extractives
on the surfaces of the CTMP pulp fibers have enhanced AKD’s spreading on the CTMP
surfaces.
0
5
10
15
20
25
30
35
0 0,01 0,02 0,03 0,04 0,05 0,06 0,07
BKP
CTMP
! C
1 (
%)
Retained amount of AKD in paper determined by GC (%)
Take-off angle 90°
a)0
5
10
15
20
25
30
35
40
0 0,01 0,02 0,03 0,04 0,05 0,06 0,07
BKP
CTMP
! C
1 (
%)
Retained amount of AKD in paper determined by GC (%)
b)
Take-off angle 25°
Figure 19. ΔC1 at take-off angle 90 ° and 25 ° for the laboratory paper sheet sheets prepared from BKP and CTMP fibers versus AKD retained in the paper determined by gas chromatography.
5.1.3 Vapor-phase diffusion
Desorption and vapor phase transfer leading to re-adsorption and thus to sizing has been
claimed to be behind the sizing mechanism of AKD (Yu and Garnier 1999, Shen et al. 2001).
In Paper I we suggested that AKD diffusion due to evaporation of AKD occurs to a minor
extent parallel to or after surface diffusion, thus explaining partially a desizing phenomenon,
i.e. decreased hydrophobicity of paper after long-term storage. The study was performed at
elevated temperatures of 50 °C and 90 °C. The decay was related to desorption and transfer to
the gas phase of AKD not bonded to fiber surfaces, a process which becomes dominant when
the AKD reservoirs (in the form of retained particles/droplets) for surface diffusive spreading
are emptied. However, the fact that the hydrophobicity significantly decreased with time
suggests that either the original and/or ketone migrated, or there was a rupture of AKD bonds.
33
It may well be the case that water vapor diffusion (or transmission) through the fiber matrix in
the papers over time has caused hydrolysis of AKD, as claimed earlier by Akpabio and
Roberts (1987). One might also speculate that the humidity has loosened bonds between
fibers in the papers stored as separate sheets resulting in larger interfiber pore radii. This in
turn would influence the wettability. The reason for the sizing reduction will be discussed in
more detail below.
The results in Table 1 indicate that sheets stored exposed to open ambient conditions have lost
some AKD within five days (Paper IV). On the other hand, no size loss occurred in the sized
paper sheet that was sealed in aluminum foil. The non-extractable fraction of AKD was
termed ‘bonded’ although we had not studied the actual mechanism of bonding. Table 1. Total amount retained and non-extractable (by isotoluene) AKD in laboratory paper sheets determined by gas chromatography.
Sample preconditioning Retained AKD (%)
Non-extractable AKD (%)
Reference sheet dried at 90°C for 20 min 0.115 0.0060
Stored exposed to ambient atmosphere at 90 °C and 50 %RH for 5 days
0.071 0.0045
Stored sealed in aluminium foil at 90 °C and 50 %RH for 5 days
0.126 0.0059
Stored exposed to ambient atmosphere at 23 °C and 50 %RH for 5 days
0.087 0.0048
In Paper II a simple test to see if vapor phase diffusion from AKD wax occurs was to place a
clean, hydrophilic filter paper at a distance of five centimeters above a beaker containing
AKD(m.p. 51) wax (Paper II). The test was carried out at 50 °C and 90 °C. The
hydrophobicity of the paper was determined using advancing contact angle. The contact angle
of the filter paper increased gradually over time. Figure 20 shows that the filter paper was
clearly hydrophobic after being exposed to AKD vapors at 90 °C for three days. The test
carried out at 50 °C did not result in hydrophobic paper, i.e. θ < 90° after eight days. It is
evident that vapor phase diffusion is not the primary sizing mechanism of AKD due to its
long development time and due to the fact that the reservoir of AKD in an AKD sized paper is
limited compared to that used in this test. It may, however may be the reason for sizing loss.
34
0
20
40
60
80
100
120
0 5 10 15 20 25 30
Front sideBack side
Co
nta
ct
an
gle
s o
f w
ate
r (d
eg
)
Time (sec)
Solid AKD at 90 °C for 3 days
Figure 20. Apparent advancing contact angles as a function of time for a filter papers exposed to vapors of AKD(51) wax at 90 °C for three days.
Evaporated compounds from AKD have been shown (Zhang et al. 2007) to be mainly fatty
acids, i.e. initial materials from the AKD synthesis and hydrolyzed ketone. Their conclusion
was that cellulose hydrophobization via vapor phase is not practicable for AKD.
5.2 Mechanisms of AKD and ASA sizing of papers related to their surface chemistry 5.2.1 Spreading models
In this section the various sizing mechanisms of AKD and ASA will be discussed. Figure 21
shows a schematic illustration for three different models relating to AKD spreading on a
cellulose surface. It has been proposed that AKD spreading leads to a monolayer film
(Lindström and Söderberg 1986a) or to isolated patches with a thickness of about 3 nm
(Ström et al. 1992). In Paper V it is suggested that AKD spreads from isolated patches via
surface diffusion (see previous chapter).
35
Figure 21. Schematic illustration of cellulose surface covered by AKD either as a monolayer, as patches or as a combination of both.
ASA sizing occurs significantly faster than for AKD. ASA sizing is generally achieved on a
paper machine at a solids content of 55-65 % (Laitinen 2006), indicating that the presence of
water is important for the spreading of ASA. Nishiyama and coworkers (1996a,b) found that
most of the ASA retained in sized paper is in a hydrolysed form (ASA-acid) and also found
that reactive ASA is necessary for efficient sizing. ASA-acid may spread on the cellulose
surface at either the air/solid, air/liquid or liquid/solid interface. It is suggested here that
Nishiyama’s results (1996) may be explained by spreading of ASA-acid adsorbed at the
air/liquid interface (Zhmud et al. 2001) under humid conditions. Formation of a thin film of
water at the surface and adsorption of ASA-acid facilitates rapid film expansion. Alternatively
it is also possible that partly hydrolyzed ASA droplets spread at the solid/liquid interface as
shown in Figure 22, although the spreading under these conditions should lead to slower
spreading. Here the driving force is a lowering of the interfacial tension between fiber and
reactive ASA and between fiber and water by the ASA-acid. Both of these mechanisms are
capable of explaining why ASA sizing occurs more rapidly than AKD sizing. The formed
ASA acid reacts in turn with sodium, calcium and aluminium ions (Wasser 1987). Figure 22
presents possible spreading mechanisms for ASA.
Figure 22. Possible spreading mechanisms for ASA; Spreading of ASA-acid at the air/water interface (Zhmud et al. 2001) and/or spreading of partly hydrolyzed ASA at the liquid/solid interface of cellulose.
36
5.2.2 Unfilled pilot papers
Using XPS, the concentrations of C1 carbon, mainly originating from ASA or AKD in the
outermost surface of paper, can be determined with a high accuracy. The C2 carbon originates
mainly from cellulose. As a result, an increase of AKD and ASA on the fiber/paper surface
can be detected as an increase in the ratio of C1 and C2 carbons as seen in Figure 23. It is
noteworthy that ASA covered the paper surface to a significantly higher degree than AKD,
although its retention was slightly lower. The retention for the ASA was 55 percent on
average, whereas it was 62 percent for the AKD sized papers (Paper VI). Note that possible
side reactions, such as oligomers and amines for AKD are not included. The retention may
increase by 10 percent if they are included (Laitinen 2006). The higher coverage of ASA is an
interesting finding, since in view of the molecular structure AKD should size paper with a
higher hydrophobic character. The content of aliphatic C1 carbons in AKD and ASA
molecule are 34 and 16, respectively.
ASA and AKD used in this study differ from each other with respect to viscosity; ASA being
oil at room temperature and AKD a solid having a melting point of 56-60 °C. Consequently,
liquid ASA would spread over fiber surfaces at a lower temperature than the solid AKD wax
which should result in higher coverage. On the other hand, spreading of ASA, in particular the
ASA-acid on fibers covered by a thin water film, as proposed above could also be the reason
for the larger coverage. The linear relationship in Figure 23 suggests that both AKD and ASA
are located on the paper surfaces as patches or patches with thin fringes. If the sizes formed a
thin film that grew in area with increased size addition, the relationship would be exponential
(Briggs 1982). The peak to background ratio (D) for the O1s signal was calculated for each
paper in order to evaluate further the surface distribution of AKD and ASA. Figure 25 shows
that the ratios for both ASA and AKD decrease in the same manner with increasing amounts
of size in paper. This means that the oxygen atoms mainly originating from cellulose are
being covered with ASA and AKD. The background analysis suggests that a thin overlayer
film was not present on the paper surface except at high addition levels of ASA or AKD.
Additionally, ASA formed a relatively complete film at a lower size amount than AKD.
However, the D values for ASA and AKD are slightly higher than the value of 18 eV
(Österberg 2005) obtained for a thin extractive film covering the total fiber surface .
37
0
0,5
1
1,5
0 0,5 1 1,5 2 2,5 3 3,5
ASAAKD
C1
/ C
2 r
atio
Added amount of size (mg/g fiber)
R= 0.994
R= 0.988a)
Unfilled pilot papers
0
0,5
1
1,5
0 0,5 1 1,5 2
ASAAKD
C1
/ C
2 r
atio
Retained amount of size in paper (mg/g fiber)
R = 0.999
R = 0.991
Unfilled pilot papers
b)
Figure 23. C1/C2 versus added amount of size (a) and retained amount of size in paper determined by LC-MS instrumentation (b) for the unfilled pilot papers sized with ASA or AKD.
18
19
20
21
22
23
24
25
0 0,5 1 1,5 2
ASAAKD
Pe
ak a
rea
/ I
ncre
ase
in
ba
ckg
rou
nd
Amount of size in paper (mg/g)
R= 0.983
R= 0.965
Figure 24. Background evaluation of the oxygen O 1s signal for the unfilled papers sized by ASA or AKD.
To gain a deeper understanding, the chemical nature of the sized papers was also evaluated by
measuring apparent dynamic advancing contact angles using two test liquids having different
surface tensions. It should be kept in mind that according to the Wenzel relation, the apparent
contact angle on hydrophobic rough surfaces is larger than the contact angle measured for the
smooth surface with the same surface energy (Wenzel 1936). Additionally, only a small
fraction of the surface needs to be hydrophobic in order to yield a high advancing contact
angle (Johnson and Dettre 1964). Ethylene glycol has a lower surface tension than water and
thus allows differentiation between papers with high degrees of sizing as seen in Figure 25. It
38
is worth noting that the contact angles achieve their highest value only at the largest size
coverage indicating that no plateau has been reached. Also, the highest contact angle does not
necessarily mean that the surface is completely covered by the size. It is difficult in practise to
prepare surfaces for which the contact angle exceeds 120 degrees (de Gennes et al. 2004).
However, combined with the results from the background analysis above, this lends weight to
the argument that a thin, almost continuous size film is not present on the surface until very
high addition levels of ASA or AKD. Furthermore, the water contact angles indicate that the
ASA sized papers wet more rapidly in spite of their higher size coverage than the
corresponding AKD papers, which means a presence of hydrolyzed ASA at the paper surface.
0
20
40
60
80
100
120
140
10 20 30 40 50 60
AKD; waterAKD; ethylene glycol ASA; water ASA; ethylene glycol
Dyn
am
ic c
on
tact
an
gle
s a
t 0
.1 s
ec (
de
g)
C1 carbon (% of Ctot
) Figure 25. Apparent dynamic contact angles of water and ethylene glycol taken at 0.1 sec versus C1 carbon for unfilled pilot papers internally sized by AKD or ASA.
The presence of partly hydrolyzed ASA on fiber surfaces would explain the slightly lower
contact angles against water. However, if ASA-acid has adsorbed to the fiber surfaces and
formed hydrogen bonds with the hydroxyl groups of the fibers, the hydrophobic tail would be
expected to point out from the S/V surface. In that case, the ASA-acid would contribute to
sizing. On the other hand, if ASA has hydrolyzed and not bonded to fibers, its hydrolyzed
product, ASA-acid, would easily react further with inorganic ions generally present in the
process water. Indeed, both calcium and aluminium were found in the surface of the ASA
sized papers as illustrated in Figure 26. With increased C1 carbon from ASA, i.e. increased
ASA content, accumulation of Ca and Al components in the surface layer clearly increases.
This is most likely due to the fact that ASA-acid has formed complexes with Ca2+ ions from
the raw water and with Al3+ and other positively charged Al ions from the alum added to the
39
stock. In comparison, merely a trace of calcium and no aluminium were found in the surface
of AKD sized papers (Paper VI).
0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
10 20 30 40 50 60
CalciumAluminium
Su
rfa
ce
co
nce
ntr
atio
ns o
f C
a a
nd
Al (a
t.%
)
C1 carbon (% of Ctot
)
Unfilled pilot papers sized with ASA
Figure 26. Surface concentrations of calcium and aluminium versus C1 carbon detected for the unfilled papers sized with ASA.
The results so far implies that ASA has to some extent hydrolyzed and the hydrolyzed ASA
formed complexes with calcium and aluminium. These, in particular ASA-acid, ‘bi-products’
may increase the paper wettability if they are not bound to fibers. This was seen in Figure 25
where the contact angles for the ASA-sized papers were slightly lower than for the AKD
sized papers. However, it has also been postulated that calcium salt of ASA-acid is
hydrophobic, and so, it can contribute to the sizing effect (Wasser and Brinen 1998). Also,
aluminium salts have been shown to improve the sizing efficiency of ASA (Scalfarotto 1985).
In view of the seemingly complex chemical nature, especially of the ASA sized papers it was
also important to examine the papers further using another surface sensitive technique. Time-
of-flight secondary ion mass spectrometry (ToF-SIMS) was used to determine surface
chemical composition and surface distribution of the ASA and AKD sized papers stored
sealed in Al foil (Paper VI). Figure 27 displays images taken in both negative and positive
mode for the ASA paper according to Wasser and Brinen (1998). A typical feature was that
only marginal amount of active, non-bonded ASA was found at the paper surface. Instead
clear signals for hydrolysed ASA and in particular for calcium and aluminium were observed.
This suggests that ASA was present at the paper surface mainly in the ASA-acid form and as
salts with calcium and aluminium. In comparison, a typical AKD sized paper showed mainly
original (active), non-bonded AKD (533.5) and hydrolysed AKD, i.e. ketone (507.5), the
40
latter being the main portion as exemplified in Figure 28. The signals in the mass ion spectra
for AKD were found at the same positions as proposed in the literature (Zimmerman et al.
1995, Kleen 2004). Additionally, a small signal for AKD oligomer was detected (Paper VI).
Figure 27. ToF-SIMS (top) negative mode images and (bottom) positive mode images for unfilled, 0.24 wt% ASA sized paper. Negative mode images show fragmented ASA (M-CO-H), ASA (M-H), hydrolysed ASA and total ions. Positive mode images show aluminium, calcium and ASA (M+H) and total ions. The resolution was 128 x 128 pixels.
Figure 28. ToF-SIMS positive mode images of hydrolysed AKD (507.5), original, non-bonded AKD (533.5) and total ions for the unfilled 0.24 wt% AKD sized paper. The resolution was 128 x 128 pixels.
5.2.3 PCC-filled pilot papers
Figure 29 shows that ASA also covers the surface of the PCC-filled papers to a higher extent
than AKD, although the difference was smaller than in the case of the unfilled papers. The
size coverage was significantly lower for the PCC-filled than unfilled papers in Figure 23.
The reason seems to be the different surface area between these papers since the PCC filled
papers have higher surface areas (Paper V). In the case of AKD, the retention of AKD was
41
better for the PCC-filled (from 74 to 82 percent) than for unfilled (62 percent on average)
papers. Higher retention of AKD to PCC surfaces is in line with other work that showed
higher adsorption of AKD on calcium carbonate than filler surfaces (Bottorff 1994,
Voutilainen 1996). Also, AKD spreads on the model surface for calcium carbonate as stated
above (page 29 and Paper II).
5
10
15
20
25
30
35
40
0 0,5 1 1,5 2 2,5 3 3,5
ASA
AKD
Re
lative
am
ou
nt
of
C1
ca
rbo
n (
at.
%)
Added amount of size (mg/g)
R= 0.956
R = 0.984
Figure 29. Relative amount of C1 carbon normalized with respect to Ctotal versus added amount of AKD for PCC-filled pilot papers.
The results from the background analysis of O1s signal presented in Figure 30 support this
conclusion. The higher the size dosage, the lower the ratio between peak area and increase in
background value meaning higher AKD overlayer film on the paper surface. The lowest value
is about 21 eV, which is only slightly lower than the ratio value 25 eV given for the
homogeneous polymeric materials (Zeggane and Delamar 1987). Thus, it is clear that AKD
covers only a fraction of the surface. In the case of ASA, the background analysis showed
large scatter. However, it may indicate the fashion how the ASA acid interacts with the
surface of the PCC, bonding through calcium bridges is possible. For a deeper understanding,
these papers need to be analyzed with respect to the retained amounts of ASA.
42
18
19
20
21
22
23
24
25
0 0,05 0,1 0,15 0,2 0,25 0,3 0,35
ASA
AKD
Peak a
rea /
Incre
ase i
n b
ackgro
und
Added amount of size (mg/g)
<- 2 points
Figure 30. Background evaluation of the oxygen O 1s signal for the PCC-filled papers sized by ASA or AKD.
5.2.4 Durability of sizing degree
Under 5.1.3 the durability of a sized laboratory sheet in terms of sizing degree was evaluated
after five days’ storage. The results indicated sizing loss. In this study, the aim was to
investigate possible sizing loss in more detail using pilot papers sized by ASA or AKD and
stored at 23 °C and 50 % RH either wrapped in aluminium foil or as separate sheets exposed
to ambient conditions over prolonged time. Figure 31 and 32 show that the unfilled papers
stored sealed in aluminium foil since papermaking indicated only a slight decrease in the
sizing degree over prolonged storage time compared to the original ones, whereas the PCC
filled papers stored in the same way had lost some of their sizing. The latter occurred to a
higher extent in the case of AKD than ASA sized papers. The ToF-SIMS analysis results
showing the presence of AKD mainly in the ketone form on the surface of the PCC filled,
AKD sized papers thus confirming the hydrolysis (Paper VI) catalyzed by PCC. In the case
of AKD, this was attributed to the formation of ketone (Bottorff 1994), catalyzed by PCC.
The slight reduction in the sizing level for ASA may depend on further hydrolysis and salt
formation. The ToF-SIMS data displayed approximately the same surface chemical
composition for unfilled and PCC filled papers sized by ASA (Paper VI).
43
0
20
40
60
80
100
120
0 10 20 30 40 50 60 70
ASA0.03 wt%
ASA0.06
ASA0.09
ASA0.12
ASA0.24
ASA0.30
Co
nta
cta
an
gle
s f
or
wa
ter
at
10
se
c (
de
g)
Storage time in weeks
Stored in pile and wrapped in Al foil
a)
0
20
40
60
80
100
120
0 10 20 30 40 50 60 70
AKD0.03 wt%
AKD0.06
AKD0.09
AKD0.12
AKD0.24
AKD0.30
Co
nta
ct
an
gle
s o
f w
ate
r a
t 1
0 s
ec (
de
g)
Storage time in weeks
Stored in pile and wrapped in Al foil
b)
Figure 31. Effect of storage on the hydrophobicity of unfilled pilot papers sized with ASA (a) or AKD (b). The apparent contact angles of water obtained at 10 sec are average values of five different measurements. The papers were stored in a stack and sealed in aluminium foil at 23 °C and 50 %RH.
0
20
40
60
80
100
120
0 10 20 30 40 50 60 70
ASA0.09 wt%
ASA0.12
ASA0.15
ASA0.18
ASA0.24
ASA0.30
Co
nta
ct
an
gle
s f
or
wa
ter
at
10
se
c (
de
g)
Storage time in weeks
a)
0
20
40
60
80
100
120
0 10 20 30 40 50 60 70
AKD0.09 wt%
AKD0.12
AKD0.15
AKD0.18
AKD0.24
AKD0.30
Co
nta
ct
an
gle
s f
or
wa
ter
at
10
se
c (
de
g)
Storage time in weeks
b)
Figure 32. Effect of storage on the hydrophobicity of PCC-filled pilot papers sized with ASA (a) and AKD (b) and stored sealed in Al foil at 23 °C and 50 % RH. The apparent contact angles of water obtained at 10 sec are average values of five different measurements.
The high durability of the sizing degree of the unfilled papers was also seen in their water-
resistance values. Amongst the AKD sized papers, only the paper having the lowest dosage
suffered from reduced water-resistance; the Cobb60-value changed from 34 to 79 g/m2, Figure
33a. After extraction, the results in Figure 33b indicate that sufficient sizing was obtained at a
surprisingly small amount of non-extractable AKD, i.e. 0.005 %. This amount is slightly
44
lower than those reported earlier (e.g. Davis et al. 1956, Roberts and Garner 1985, Lindström
and Söderberg 1986a). For a comprehensive review and references there, see Hubbe (2006).
However, it is in line with previous results obtained using the same extraction procedure
(Laitinen 2006) where extraction is carried out at high temperature and pressure. This small
non-extractable fraction, removable only by alkaline hydrolysis supports the existence of
covalently bonded AKD argued in the above-mentioned publications.
0
10
20
30
40
50
60
70
80
0 0,05 0,1 0,15 0,2 0,25 0,3 0,35
2 days
68 weeks
Co
bb 6
0 (
g/m
2)
Added amount of AKD (wt.%)
a)
0
10
20
30
40
50
60
70
80
0 0,002 0,004 0,006 0,008 0,01 0,012
68 weeks
Co
bb 6
0 (
g/m
2)
Non-extractable AKD in paper (%)
b)
Figure 33. Cobb60 values vs. added (a) and non-extractable (b) amount of AKD for unfilled pilot papers sized with AKD. Note that the Cobb test was performed on the papers before extraction.
Figure 34 shows that ASA sizing follows the same trend as AKD in terms of storage, namely
that the long-term storage in the conditions used does not negatively affect the water-
resistance.
0
10
20
30
40
50
60
70
80
0 0,05 0,1 0,15 0,2 0,25 0,3 0,35
2 days
68 weeks
Co
bb 6
0 (
g/m
2)
Added amount of ASA (wt.%) Figure 34. Cobb60 values versus added amount of ASA for unfilled pilot papers sized with ASA.
45
In comparison, the sizing degree of the papers stored as separate sheets exposed to ambient
conditions was significantly reduced after only a few weeks’ storage as illustrated in Figure
35. This occurred in spite of the fact that the papers were stored at a relatively low
temperature of 23 °C. The XPS-analysis showed reduced C1 carbon amounts at the outermost
surface of the papers thus confirming the sizing reversion or sizing loss (Paper VI). It is clear
from Figure 36 that at the end of the storage the AKD papers had greatly reduced sizing
degree compared to the ASA papers of the same loading.
0
20
40
60
80
100
120
0 10 20 30 40 50 60 70
ASA0.03 wt%ASA0.06ASA0.09ASA0.12ASA0.24ASA0.30
Co
nta
ct
an
gle
s f
or
wa
ter
at
10
se
c (
de
g)
Storage time in weeks
Stored as separate sheets
a)
0
20
40
60
80
100
120
0 10 20 30 40 50 60 70
AKD0.03 wt%AKD0.06AKD0.09AKD0.12AKD0.24AKD0.30
Co
nta
ct
an
gle
s f
or
wa
ter
at
10
se
c (
de
g)
Storage time in weeks
Stored as separate sheets
b)
Figure 35. Apparent contact angles of water at 10 sec for unfilled pilot papers sized with ASA (a) or AKD (b), and stored as separate sheets exposed to ambient conditions (23 °C and 50 %RH).
The decrease was believed to partly depend on the slow humidity diffusion into the paper
matrix thus affecting the pore structure, hydrolysis of the size and possibly the covalently
bonded AKD. However, the ToF-SIMS analysis revealed that the paper stored exposed to
ambient conditions showed a strongly decreased signal for original, non-bonded AKD (533.5)
compared to the same paper stored in aluminium foil. The signal for hydrolyzed AKD
(507.5) had also decreased, as well as the other typical signals for AKD. The results indicate a
loss of AKD from the unprotected paper. The mass loss of AKD was confirmed by the LC-
MS analysis (Paper VI). A question arises whether desorption and vapor phase diffusion of
AKD and ketone would the reason for the sizing loss although it can be questioned due to
their relatively high molecular mass (532 for AKD-C18). Vapor phase diffusion would occur
from a monolayer of AKD or ketone or a mixture of them that have spread on the fiber
surfaces through surface diffusion. The contact angles in Figure 35 may reflect the rate of the
slow surface diffusion. It has been shown by Zhang et al. (2007) that ketone and fatty acids
46
can evaporate at the typical papermaking temperature. Migration through desorption and
vapor phase diffusion at the low storage temperature used in this work seems not possible due
to the relatively high molecular mass of AKD and ketone (e.g. 532 for AKD-C18). Migration
occurs most probably through surface diffusion, from a monolayer of ketone.
507.65
533.61
479.60
507.62
533.58
mas s / u470 480 490 500 510 520 530 540
3
x10
0. 5
1. 0
1. 5
2. 0
Inte
nsit
y
3
x10
0. 5
1. 0
1. 5
2. 0
Inte
nsit
y
AKD pape r s to r ed in a luminium fo il
AKD pape r s to r ed ex ps ed to a mbie nt c o nditions
1001.52
1012.24
1022.17
mas s / u1005 1015 1025
1
x10
1. 0
2. 0
3. 0
4. 0
5. 0
6. 0
Inte
nsit
y
1000.65
1013.38
1022.19
mas s / u1005 1015 1025
1
x10
1. 0
2. 0
3. 0
4. 0
5. 0
6. 0
Inte
nsit
y
Figure 36. ToF-SIMS positive ion mass signals of 533.6 (original, non-bonded AKD) and 507.6 (hydrolysed AKD) for the 0.24 wt% unfilled AKD sized paper stored in aluminium foil (top) and as a separate sheet exposed to ambient conditions (bottom).
The PCC filled papers showed increased wettability only after a few weeks’ storage, most
probably due to a combined effect of further hydrolysis of AKD and humidity (50 %RH).
Moreover, the signal for hydrolyzed AKD was significantly lower than for the protected paper
thus indicating sizing loss. The results demonstrated that ASA sized papers also suffered from
sizing loss, though to a remarkable lower extent. In addition to the effect of humidity, possible
vapor phase diffusion (Zhang et al. 2007), a kind a rearrangement also seemed to have
47
occurred. This, since the amount of calcium detected by ToF-SIMS at the surface of the PCC
filled paper, sized by ASA and stored exposed to ambient conditions, was clearly lower
compared to the paper stored in Al foil. A surface-active compound may have migrated to the
air/solid interface thus covering the compound containing calcium or the ASA-Ca complex
have decomposed during storage. The fact that aluminium signal in the same analysis was
unchanged supports the latter hypothesis.
5.3 Wettability and friction properties of sized papers related to their surface chemistry 5.3.1 Wettability vs. size coverage
As shown above, using the XPS technique we have determined the concentration of the C1
carbon originating mainly from ASA or AKD in the outermost surface of paper. Figure 18
above displayed initial contact angles for the sized papers. Figure 37 shows dynamic contact
angles obtained 10 sec after the application of the liquid droplets. At this stage, possible liquid
absorption in paper may have occurred resulting in lower contact angles. As seen, the contact
angles increase continuously with increasing size coverage. A maximum resistance against
liquid absorption is achieved when the fiber/paper surface is covered by about 50 atomic
percent of C1 carbon. This finding is perfectly consistent with the results of the XPS
background analysis (Paper VI) demonstrating the fiber/paper surface not being completely
covered by the size below that value. It is also important to note that it appears that only a
certain part of the paper surface needs to be covered by sizing agent to result in contact angles
higher than 90 degrees for water. In that case, however, the sizing effect is not very stable due
to non-uniform size coverage and large wetting hysteresis (von Bahr et al. 2004). For
example, water vapor starts to condense on hydrophilic spots on the pore walls at pressures
below the equilibrium liquid vapor pressure. Consequently, the contact angle is shifted
gradually from advancing to receding and the sustained pressure head is reduced, thus
enabling water uptake. This is important, particularly for liquid packaging, which demands a
long-term resistance to liquid absorption.
Optimization of liquid penetration is necessary in many applications, such as in water-based
printing. Given the same sizing degree in a paper, liquid penetration time can be affected by
choosing a liquid with a lower surface tension than water. In packaging board, resistance to
penetration to liquids with lower surface tension than water is required. The influence of the
surface tension is exemplified in Figure 37b where the above-mentioned papers were also
48
studied using ethylene glycol, which has a lower surface tension (48.0 mN/m) and higher
viscosity than water (Paper VI).
Most importantly, it is clear that the ASA produces papers with a lower resistance to the
sorption of polar liquids. This is ascribed to the presence of hydrolyzed ASA products in the
paper surface.
0
20
40
60
80
100
120
140
10 20 30 40 50 60
AKD
ASA
Dyn
am
ic c
on
tact
an
gle
s a
t 1
0 s
ec (
de
g)
C1 carbon (% of Ctot
)
a)
0
20
40
60
80
100
120
140
10 20 30 40 50 60
AKD
ASA
Dyn
am
ic c
on
tact
an
gle
s a
t 1
0 s
ec (
de
g)
C1 carbon (% of Ctot
)
b)
Figure 37. Apparent contact angles of water (water) and ethylene glycol (b) at 10 sec versus C1 carbon at the outermost surface of unfilled ASA and AKD sized pilot papers. The results are extracted from figures 7, 13 and 14 in Paper VI.
The contact angles shown above were from a dynamic wetting study where the time-
dependent wetting of each ASA and AKD sized paper by water and ethylene glycol was
followed, as illustrated in Figure 38 for ethylene glycol. The wetting and absorption by
ethylene glycol is clearly dependant on the sizing degree of the paper surface. Three different
wetting regimes are observed, seen most clearly for the papers with low sizing degree. For
example, the PCC filled papers show fastest wetting up to 0.1 second followed by a slower
wetting regime. At low addition levels, the contact angles decrease after 10s due to absorption
of the liquid droplet into paper structure. The long initial wetting is most likely due to its high
viscosity. The viscosity retards sub-surface penetration more than surface spreading
(Ranabothu et. al 2005). It has to be noted that there is most probably a difference in the z-
structure of the filled paper due to the disruption caused by PCC particles, which may affect
how a liquid absorbs into paper.
49
0
20
40
60
80
100
120
140
0,1 1 10 100
ASA0.09, no fillerASA0.09+PCCASA0.12, no fillerASA0.12+PCCASA0.24, no fillerASA0.24+PCC
Dyn
am
ic c
on
tact
an
gle
s o
f e
thyle
ne
gly
co
l (d
eg
)
Time (s)
a)
0
20
40
60
80
100
120
140
0,1 1 10 100
AKD0.09, no fillerAKD0.09+PCCAKD0.12, no fillerAKD0.12+PCCAKD0.24, no fillerAKD0.24+PCC
Dyn
am
ic c
on
tact
an
gle
s o
f e
thyle
ne
gly
co
l (d
eg
)
Time (s)
b)
Figure 38. Time-dependence of dynamic apparent contact angles of ethylene glycol for unfilled and PCC filled pilot papers sized by ASA (a) and AKD (b).
In Paper IV, to evaluate surface energies of AKD sized laboratory paper sheet sheets, contact
angle measurements with three test liquids including water, ethylene glycol and
diiodmethane, were carried out according to the Good-van Oss method (1992). As expected,
the surface energy of AKD-sized sheets reduced with the increasing AKD content, the polar
component of the surface energy being the most influenced.
5.3.2 Friction vs. size coverage
A strong correlation was observed between friction and AKD coverage for the unfilled paper.
In order to compare PCC-filled papers with the unfilled ones, the C1 carbon mainly
originating from ASA or AKD is given as a relative amount, i.e. it has been normalized with
respect to the total amount of carbon since carbon and oxygen signals originate from fibers as
well as from carbonate. Figure 39 shows that AKD clearly decreases the paper-to-paper
friction as also reported in other studies (Marton 1990, Hoyland et al. 2001 and Karademir et
al. 2004).
50
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
5 10 15 20 25 30 35 40
AKD; Unfilled paper
ASA; Unfilled paper
AKD; PCC filled paper
ASA; PCC filled paper
Coeffic
ient of th
ird s
tatic friction, S
3
Relative amount of C1 carbon (atomic %)
Figure 39. Coefficients of third static friction, S3 vs. the relative amount of C1 carbon for unfilled and PCC-filled pilot papers.
In the case of ASA, the friction values are surprisingly independent of coverage of ASA.
However, the reason is not clear. ASA molecules appear to be in the form of insoluble salts at
the paper surface, and thus formation of bilayers is possible. A fraction of ASA-Ca and ASA-
Al salt compounds deposit at the fiber/water interface with carbon chains upwards and ASA-
acid molecules adsorb with upside down orientation over the first layer. This model would
explain the higher wettability of ASA sized papers in spite of their higher surface coverage.
As seen in Figure 39, the use of PCC as filler increases the paper-to-paper friction markedly
compared to the unfilled papers. This is in good agreement with other work (Hoyland and
Neill 2001, Garoff et al. 2002, Karademir et al. 2004). In the PCC series of papers, there is no
decrease in friction caused by AKD. It seems to be merely due to the fact that the AKD
coverage values for the PCC filled papers are below the critical value that resulted in lower
friction for the unfilled papers.
Figure 40 shows that the reduction in friction starts when the AKD coverage, given as C1
carbon is above 28 atomic percent. At this critical point AKD seems to start acting like a
lubricant. This can perhaps be explained by the existence of a certain degree of coverage of
AKD patches at the surface, also on the elevated areas, which are contact regions between
two rough paper surfaces. Paper with AKD coverage at this level showed sufficient resistant
against wetting and absorption by water; the contact angles were clearly over 90 degrees (see
Figure 37). After extraction of the AKD by tetrahydrofuran the friction was the same as for
51
the reference paper without AKD. The extraction removed non-bonded AKD composing both
original AKD and ketone, which resulted in higher friction. This is in line with the results
reported by Marton (1990) and Karademir and Hoyland (2003).
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
15 20 25 30 35 40 45 50
S3K3
Co
eff
icie
nts
of
fric
tio
n
C1 carbon (of Ctot
)
Unfilled papers sized with AKD
Figure 40. Coefficients of third static, S3 and kinetic, K3 friction versus C1 carbon (mainly originating from AKD) for unfilled pilot papers sized with AKD.
Storage of pilot papers unprotected from ambient atmosphere for four months resulted in
lower paper-to-paper friction values for AKD sized pilot papers than those of the original
ones (Seppänen et al. 2001). It was concluded that AKD had spread/migrated and hydrolyzed
further during storage and thus increased the AKD coverage and the amount of ketone at the
paper surface. Additionally, the higher drying temperature (from 23 °C to 90 °C) and time of
prepared laboratory sheets the lower friction.
52
6. Concluding remarks Throughout the work effort has concentrated on improved understanding of sizing
mechanisms of ASA and AKD. When starting this work, knowledge concerning the spreading
of AKD was relatively limited. The sizing mechanism of ASA was still under debate. It was
believed that the primary sizing mechanism for ASA and AKD was the formation of ester
bonds between the size and the hydroxyl group of cellulose. The difference in the rate in
which the sizing developed was thought to be dependent on the higher reactivity of ASA than
AKD.
The results of several studies in this work argue the importance of AKD spreading. The
results strongly suggested that AKD exists at the paper surface mainly as thin patches. The
role of drying, in particular the importance of high temperature in the early stage of drying
appears to be crucial. Moreover, it is shown that the water-resistance improved with
increasing size coverage. On the other hand, only a surprisingly small non-extractable fraction
is needed to obtain sufficient sizing degree, thus supporting the previous findings done by
other researchers on covalently bonded AKD. Therefore it would be interesting in the future
to study the spreading of AKD contra as opposed to its anchoring in a systematic and
controlled way to see as to whether improved spreading increases the amount of non-
extractable/bonded AKD.
The spreading of monolayer precursor from the root of an AKD droplet/particle through
surface diffusion was shown to occur. This explains the way in which AKD migrates in the
paper matrix and between paper sheets. The monolayer diffusion process is quite slow, the
apparent surface diffusion coefficient being of the order 10-11 m2/s. There is a linear increase
with temperature and an inverse proportionality with respect to the melting point of AKD. It
is reasonable to believe that the migration of AKD occurs to a lower extent if the initial
spreading of molten AKD in the dryer section of the paper machine is as complete as
possible. Thus small AKD particles are preferred instead of large agglomerates. In that case
the reservoirs of AKD will be emptied more rapidly and the migration will stop. It is also
clear that both AKD and hydrolyzed AKD (ketone) migrate. It was concluded that sizing loss
occurred from papers stored exposed as separate sheets to ambient conditions was mainly due
to migration through surface diffusion. This was based on the fact that the amount of AKD
and hydrolyzed AKD decreased according to the analysis performed by XPS, ToF-SIMS and
53
liquid chromatography mass spectrometry (LC-MS). Migration through desorption and vapor
phase diffusion at the low storage temperature is not possible due to the relatively high
molecular mass of AKD and ketone (e.g. 532 for AKD-C18).
It would be also interesting to investigate in more detail the effect of promoters on the sizing
efficiency of AKD. It is known that a number of promoters improve AKD sizing by
improving the retention of AKD. AKD sizing may be improved by promoting its spreading
capability.
A comparison study between ASA and AKD sizing using pilot papers made under the same
conditions resulted in many interesting findings. It is by no means an exaggeration to state
that the results provide strong evidence for the ASA sizing mechanism where hydrolyzed
ASA has the key role. The proposed spreading mechanism of ASA explains its higher
coverage of cellulose compared to AKD. ASA appears to be manifested mainly as thin
patches. Still the question of the existence of the ester bond is unanswered.
This work also studied the durability of AKD and ASA sizing over prolonged time in a
systematic way. It was shown that the sizing level of the unfilled papers sized by AKD or
ASA and stored protected against the environment was surprisingly stable. However, most
probably there will be changes if papers are stored under elevated temperature and/or
humidity. The results confirm findings by other researchers concerning the catalytic effect of
PCC. It is shown that the means of storage is crucial. Clearly, AKD sized papers suffered
from sizing loss more than those sized by ASA. There was an indication that the ASA-Ca
compared to the ASA-Al complex was not as durable over prolonged storage when it was
exposed to ambient conditions.
XPS gives quantitative information on the content of size on the paper surface and, combined
with ToF-SIMS, valuable knowledge concerning the chemical state of the size can be
obtained. Clear relations between wettability and friction of paper and the surface chemical
composition have been demonstrated.
54
Acknowledgements First I want to thank the industries (AssiDomän, Eka Chemicals, Hercules, Korsnäs, Kemira,
M-real, Stora Enso, Tetra Pak, UPM) supporting financially the projects on sizing of paper
that produced the results summarized in this thesis. The studies were conducted at YKI,
Institute for Surface Chemistry.
KTH and YKI are gratefully thanked for the opportunity to write this thesis.
I would like to express my deepest gratitude to Prof. Mark Rutland, Prof. Bruce Lyne and
Assoc. Prof. Agne Swerin for scientific discussions of the four latest manuscripts and thesis,
and moreover, for your enthusiasm and support.
Prof. Göran Ström, my former colleague is gratefully thanked for introducing me to the world
of surface chemistry, especially paper sizing, and moreover for co-operation in Paper III. I
learned a lot about papermaking during the week in Maastricht, many years ago. I want
expand my special thanks to Prof. Fredrik Tiberg, my former colleague for introducing me in
the world of spreading and for co-operation in Paper I. I think that research combining model
surfaces and laboratory paper sheets resulted in some quite interesting findings. Maria von
Bahr and Boris Zhmud, my former colleagues are thanked for co-operation in Paper IV.
I want to express special thanks to Karin Hallstensson and Mikael Sundin for skilful
laboratory assistance. Karin is also thanked for help with figures. Anne-Marie Härdin, now
deceased, is remembered for her assistance with spreading studies. I want thank Marie
Ernstsson for valuable discussions and help with XPS analyses in Paper V and Ulrika
Johansson at SP for help and valuable discussions concerning ToF-SIMS analyses. Rodrigo
Robinson is thanked for help with DSC. Moreover, Rodrigo and Annika Dahlman are
acknowledged for helping me to check some experimental points during the course of writing.
I thank Jerzy Kizling, my former colleague for scientific discussions concerning ASA sizing
and friction and moreover, for commenting manuscript V.
Special thanks to Britt, Peder, Ingvar, Marjatta, my former colleague and Susanne. There
would not be successful projects without your help.
A list of the present and former colleagues at YKI who I would like to thank for making my
work and life enjoyable is long, so I say THANK YOU / KIITOS! to all of you without
mentioning any names.
55
My dear mom, sister, brothers and their children are thanked for the encouragement. I am
privileged to have many relatives and friends in different countries. Thanks for being in my
life.
Finally, my deepest gratitude and love go to Anders and Jennifer for making every day of my
life enjoyable.
56
References Akpabio, U. D. and Roberts, J. C. (1987), Alkylketene Dimer Sizing: Vapor-Phase Deposition of Tetradecylketene Dimer on Paper, Tappi J. 70(12), 127. Back, E. L., (1991), Paper-to paper and paper-to-metal friction, In: Tappi International Paper Physics Conference, Kona, Hawaii, Tappi, pp. 49-55. Back, E. L. (2002), Friction of Paper, In: Handbook of Physical Testing of Paper, Volume 2, edided by Borch, J., M., Lyne, M. B., Mark, R. E. and Habeger Jr, C. C., Marcel Dekker Inc. p. 451-476. Bico, J., Thiele, U. and Quere, D. (2002), Wetting of textured surfaces, Colloids Surf. A, 206, 41. Bogdanovic, G., Tiberg, F. and Rutland, M. W. (2001), Sliding friction between cellulose and silica surfaces, Langmuir, 17(19), 5911. Bottorff, K.J. (1994), AKD Sizing Mechanism: More Definitive Description, Tappi J., 77(4), 105. Briggs, D. (1982), XPS Studies Of Polymer Surface Modifications and Adhesion Mechanisms, Adhesion 13(3-4), 287. Carlsson, G. and Ström, G. (1995), Water sorption and surface composition of untreated or oxygen plasma-treated chemical pulps, Nord. Pulp Pap. Res. J., 1, 17. Cassie, A.B.D. and Baxter, S. (1944), Wettability of porous surfaces, Trans. Faraday Soc., 40, 546. Cates, R. E., Dumas, D. H. and Evans, D. B., (1989), Alkyl ketene dimer sizes, In: The sizing of paper, Reynolds, W. F. (ed.), Tappi Press, Atlanta, pp. 33-50. Champ, S. and Ettl, R. (2004), The dynamics of alkylketene dimer (AKD) retention, J. Pulp Paper Sci. 30(12), 322. Dart, P. J., and McCalley, D. V. (1990), Determination of alkyl ketene dimer sizing agent products in paper by capillary gas chromatography. Analyst, 115(1), 13. Davis, J. W., Robertson, W. H. and Weisgerger, C. A. A new sizing agent for paper – alkyl ketene dimers, Tappi J., 39(1), 21. deGennes, P. G., Brochard-Wyart, F. and Quéré, D. (2004), Capillarity and wetting phenomena : Drops, bubbles, pearls, waves, Springer-Verlag New York, Inc, New York, pp.1-291. Della Volpe, C. and Siboni, S. (2000), Acid-base surface free energies of solids and the definition of scales in the Good-van Oss-Chaundry theory, J. Adhes. Sci. Technol., 14(2), 235. Della Volpe, C., Maniglio, D., Siboni, S. and Morra, M. (2003), Recent theoretical and experimental advances in the application of van Oss-Chaudbury-Good acid-base theory to analysis of polymer surfaces. I. General aspects, J. Adhes. Sci. Technol., 17(11), 1477. Dettre, R. H. and Johnson, Jr., R. E. (1964), Contact Angles, Wettability and Adhesion, Adv. in Chem. Ser. 43, p. 136, Gould, R. F. (ed.), Am. Chem. Soc, Washington, D.C. Dorris, G.M. and Gray, D.G. (1978), The surface analysis of paper and wood fibers by ESCA (Electron Spectroscopy for Chemical Analysis). I. Application to cellulose and lignin. Cellul. Chem. Technol., 12(1), 9.
57
Dumas, D. H. and Evans, D. B. (1986), AKD-cellulose reactivity in papermaking systems, Proc. TAPPI Papermakers Conf., Atlanta, TAPPI Press, 31-35. Extrand, C. V. (2004), Contact Angles and Their Hysteresis as a Measure of Liquid-Solid Adhesion, Langmuir, 20, 4017. Fardim, P., Hulten, A. Hejnesson, Boisvert, J.-P., Johansson, L.-S., Ernstsson, M., Campbell, J.M., Lejeune, A., Holmbom, B., Laine, J. and Gray, D. (2006), Critical comparison of methods for surface coverage by extractives and lignin in pulps by X-ray photoelectron spectroscopy (XPS), Holzforschung, 60(2), 149. Freire, C. S. R., Silvestre, A. J. D., Pascoal Neto, C., Gandini, A., Fardim, P. and Holmbom, B. (2006), Surface characterization by XPS, contact angle measurements and ToF-SIMS of cellulose fibers partially esterified with fatty acids, J. Colloid Interface Sci., 301(1), 205. Garoff N, Nilvebrant, N-O and Fellers C, J. (2002), The friction of paper based on recycled fibre, J. Appl. Polym. Sci., 85(7), 1511.
Garnier, G., Wright, J., Godbout, L. and Yu, L. (1998), Wetting mechanism of alkyl ketene dimers on cellulose films, Colloids Surf. A, 145(1-3), 165.
Garnier, G., Bertin, M. and Smrckova, M. (1999), Wetting dynamics of alkyl ketene dimer on cellulosic model surfaces, Langmuir, 15(22), 7863.
Garnier, G. and Yu, L. (1999), Wetting mechanism of a starch-stabilized alkylketene dimer emulsion: A study by Atomic Force Microscopy, J. Pulp Paper Sci., 25, (7), 235.
Garnier, G. and Godbout, L. (2000), Wetting behaviour of alkyl ketene dimer on cellulose and model surfaces, J. Pulp Paper Sci., 26(5), 194. Gellersted, F. and Gatenholm, P. (1999), Surface properties of lignocellulosic fibers bearing carboxylic groups, Cellulose, 6, 103. Gess, J. M. and Lund, R. C. (1991), Strong-Bond/Weak-Bond Theory of Sizing. (2). Alkyl Ketene Dimer and the Strong-Bond/Weak-Bond Theory, Tappi J., 74(1), 111. Good, R. J. and van Oss, C. J. (1992), The Modern Theory of Contact Angles and the Hydrogen Bond Components of Surface Energies, In: Modern Approaches to Wettability, Schrader, M. E. and Loeb, G. I. (Eds.), Plenum, New York, p. 1. Hardell, H. - L. and Woodbury, S. E. (2002), A new method for the analysis of AKD oligomers in papermaking systems, Nord. Pulp Pap. Res. J., 17(3), 340. Hatanaka, S., Takahashi, Y. and Roberts, J. C. (1991), Sizing with saponified alkenyl succinic acid, Tappi J., 74(2), 177. Heslot, F., Cazabat, A. M., Levinson, P. (1989), Dynamics of Wetting of Tiny Drops: Ellipsometric Study of the Late Stages of Spreading. Phys. Rev. Lett. 62:1286. Hoyland, R. W., Neill, M. P. and Keenan, T. (2000), Factors affecting the frictional properties of paper - the effect of ASA neutral size, J. Pap. Techn. 41(10), 50. Hoyland, R. W. and Neill, M. (2001), Factors affecting the frictional properties of paper - The effect of AKD neutral size, J. Pap. Techn., 42(3), 45. Hubbe, M. A. (2006), Paper's resistance to wetting - A review of internal sizing chemicals and their effects, BioResources, 2(1), 106. Hutton, B. and Shen, W. (2005), Sizing effects via AKD vaporisation, Appita J. 58(5), 367.
58
Inoue, M., Gurbagul, N. and Aroca, P. (1990), Static friction properties of linerboard, Tappi J., 73(12), 81. Isogai, A. Taniguchi, R., Onabe, F. And Usuda, M. (1992), Sizing mechanism of alkylketene dimers, Part 1. Possibility for AKD to form β-ketoesters in paper sheets. Nord. Pulp Pap. Res. J., 7(4), 193. Isogai, A., Nishiyama, M. and Onabe, F. (1996), Mechanism of Paper Sizing with ASA, 3: Mechanism of Retention of Alkenyl Succinic Anhydride (ASA) on Pulp Fibers at the Wet End of Papermaking, Sen’i Gaggaishi 52(4), 201. Isogai, A. (1997), Effect of Cationic Polymer Addition on retention of Alkylketene Dimer, J. Pulp Paper Sci., 23(6), 276. Isogai, A. (1999), Mechanism of paper sizing by alkylketene dimers, J. Pulp Paper Sci. 25(7), 251. Israelachvili, J. N., Chen, Y. L. and Yoshizawa, H. (1994), Relationship Between Adhesion And Friction Forces, J. Adhes. Sci. Techn., 8(11), 1231. Jiang, H. and Deng, Y. (2000), The effects of inorganic salts and precipitated calcium carbonate filler on the hydrolysis kinetics of alkylketene dimer, J. Pulp Paper Sci. 26(6), 208. Johansson, L.-S., Campbell, J.M., Koljonen, K. and Kleen, M. (2004), On surface distributions in natural cellulosic fibers, Surf. Interface Anal., 36, 706. Johansson, A., Fellers, C., Gunderson, D. and Haugen, U. (1998), Paper Friction - Influence of Measurement Conditions, Tappi J. 81(5), 175. Johansson, J. and Lindström, T. (2004a), A study on AKD-size retention, reaction and sizing efficiency Part 1: The effects of pulp bleaching on AKD-sizing, Nord. Pulp Pap. Res. J., 19(3), 330. Johansson, J. and Lindström, T. (2004b), A study on AKD-size retention, reaction and sizing efficiency Part 2: The effects of electrolytes, retention aids, shear forces and mode of addition on AKD-sizing using anionic and cationic AKD-dispersions, Nord. Pulp Pap. Res. J. 19(3), 336. Johnson, Jr., R. E. and Dettre, R. H. (1964), Contact Angles, Wettability and Adhesion, Adv. In Chem. Ser. 43, p. 112, Gould, R. F. (ed.), Am. Chem. Soc. Washington, D.C. Karademir, A. and Hoyland, R. W. (2003), The sizing mechanism of AKD and its effect on paper friction, Appita J., 56(5), 380. Karademir, A., Hoyland, D., Wiseman, N. and Xiao, H. (2004), A study of the effects of alkyl ketene dimer and ketone on paper sizing and friction properties, Appita J., 57(2), 116. Koljonen, K., Österberg, M., Kleen, M., Fuhrmann, A. and Stenius, P. (2004), Precipitation of lignin and extractives on kraft pulp: Effect on surface chemistry, surface morphology and paper strength, Cellulose, 11(2), 209. Koljonen, K., Österberg, M., Johansson, L.-S. and Stenius, P. (2003), Surface chemistry and morphology of different mechanical pulps determined by ESCA and AFM, Colloids Surf. A: Physicochem. Eng. Aspects, 228, 143. Laine, J. and Stenius, P. (1994), Surface characterization of unbleached kraft pulps by means of ESCA, Cellulose, 1, 145.
59
Laitinen, Risto (2007), Development of LC-MS and extraction methods for the analyses of AKD, ASA, and rosin sizes in paper products, Doctoral Thesis, ISBN 952-214-328-6, Lappeenranta University of Technology, Finland. Lee, H. L. and Luner, P. (2005), Effect of relative humidity and unreacted AKD on AKD sizing, Nord. Pulp Pap. Res. J., 20(2), 227. Lindfors, J., Ahola, S., Kallio, T., Laine, J., Stenius, P. and Danielsson, M. (2005), Spreading and adhesion of ASA on different surfaces present in paper machines, Nord. Pulp Pap. Res. J., 20(4), 453. Lindström, T. and Glad-Nordmark, G. (2007), A study of AKD-size retention, reaction and sizing efficiency. Part 3: The effects of fibre charge density and electrolyte concentration on size retention, Nord. Pulp Pap. Res. J., 22(2), 161. Lindström, T. and Söderberg, G. (1986a), On the Mechanism of Sizing with Alkylketene Dimers. Part 1. Studies on the Amount of Alkylketene Dimer Required for Sizing Different Pulps, Nord. Pulp Pap. Res. J., 1(1), 26. Lindström, T. and O’Brian, H. (1986), On the mechanism of sizing with alkylketene dimers, Part 2. The kinetics of reaction between alkylketene dimer required for sizing different pulps, Nord. Pulp Pap. Res. J., 1(1), 34. Lindström, T. and Söderberg, G. (1986b), On the Mechanism of Sizing with Alkylketene Dimers. Part 3. The Role of pH, Electrolytes, Retention Aids, Extractives, Ca-Lignosulphonates and Mode of Addition on Alkylketene Dimer Retention, Nord. Pulp Pap. Res. J., 1(2), 31. Lindström, T. and Söderberg, G. (1986c), On the Mechanism of Sizing with Alkylketene Dimers. Part 4. The Effects of HCO3
- Ions and Polymeric Reaction Accelerators on the Rate of Reaction between Alkylketene Dimers and Cellulose, Nord. Pulp Pap. Res. J., 1(2), 39. Maloney, T. C., Johansson, T. and Paulapuro, H. (1998), Removal of water from cell wall during drying, Paper Techn. 39(6), 39. Marton, J. (1990), Practical Aspects of Alkaline Sizing on Kinetics of Alkyl Ketene Dimer Reactions: Hydrolysis of Alkyl Ketene Dimer, Tappi J., 73(11), 139.
Mattson, R., Lindström, D., Sterte, J. And Ödberg, L. (2003), Influence of Abietic Acid, Betulinol, Sodium Oleate and Tripalmitine on the Migration of AKD in Paper, J. Pulp Paper Sci., 29(8), 281.
McCarthy, W. R. and Stratton, R. A. (1987), Effects of Drying on Alkenyl Succinic Anhydride Esterification and Sizing, Tappi J., 70(12), 117. McDonald, J. D. (1996), Measuring Paper Friction: Testing Conditions and Papermaking Factors, TAPPI Finishing&Converting Conf., 6-9 Oct, Bal Harbor, Florida, USA, pp. 163-171. Meiron, T.S., Marmur, A. and Saguy, I.S. (2004), Contact angle measurement on rough surfaces, Colloid Interface Sci., 274, 637. Modaressi, H. and Garnier, G. (2002), Mechanism of wetting and absorption of water drops on sized paper: effects of chemical and physical heterogeneity, Langmuir, 18(3), 642. Neimo, L. (1999) Internal sizing of paper, In: Papermaking Science and Technology, Book 4, Chapter 7, Papermaking Chemistry, Neimo, L. (Ed.), Fapet Oy, ISBN 952-5216-04-7, pp. 150-203.
60
Nishiyama, M, Isogai, A. and Onabe, F. (1996a): Mechanism of paper sizing with ASA; Structures of alkenyl succinic anhydride (ASA) components in ASA-sized paper sheet, 1: Sen’i Gaggaishi 52(4), 180. Nishiyama, M., Isogai, A. and Onabe, F. (1996b), Mechanism of Paper Sizing with ASA, 2: Roles of Reactive Alkenyl Succinic Anhydride (ASA) in Paper Sizing, Sen’i Gaggaishi 52(4), 189. Novak, R. W. and Rende, D. S. (1993), Size Reversion in Alkaline Papermaking, Tappi J., 76(8), Ödberg, L., Lindström, T., Liedberg, B. and Gustavsson, J. (1987), Evidence for Beta-Keto Ester Formation During the Sizing of Paper with Alkylketene Dimers, Tappi J., 70(4), 135. Östenson, M., Järund, H., Toriz, G. and Gatenholm, P. (2006), Determination of surface functional groups in lignocellulosic materials by chemical derivatization and ESCA analysis, Cellulose, 13, 157. Österberg, M., Koljonen, K. and Johansson, L.-S. (2005), Detecting wood extractives on pulp fiber surfaces using AFM and ESCA, Appita J., 2, 69. Payne, C. C. (1989), Coating Paper with Colloidal Silicas Helps Stop Slippage of Containers, J. Pulp Paper 63(8), 91. Ranabothu, S.R., Cassandra, K. and Dai, L.L. (2005), Dynamic wetting: Hydrodynamic or molecular-kinetic? J. Colloid Interface Sci., 288, 213. Ravnjak, D., Moze, A. and Plazl, I. (2006), Hydrolysis of micron alkylketene dimer particles under alkaline conditions, Chem. Biochem. Eng. Quart. 20(4), 463. Ravnjak, D., Plazl, I. and Moze, A. (2007), Kinetics of colloidal alkylketene dimer particle deposition on pulp fibers, Colloid Polym. Sci., 285, 907. Roberts, J. C. and Garner, D. N. (1985), The mechanism of alkyl ketene dimer sizing of paper. Part 1, Tappi J., 68(4), 118. Roberts, J. C. (1996), Neural and alkaline sizing. In: Paper Chemistry. Roberts J. C. (ed.), Glasgow, Blackie Academic & Professional, ISBN 0 7514 0236 2, pp. 140-160. Rohringer, P., Bernheim, M. and Werthemann, D. P. (1985), Are So-Called Reactive Sizes Really Cellulose-Reactive?, Tappi J., 68(1), 83. Sato, T., Isogai, A. and Onabe, F. (2000), Autoxidation: Predominant mechanism of size reversion for ASA-sized paper, Nord. Pulp Pap. Res. J., 15(3), 176. Scalfarotto, R. E. (1985), Remedies for press picking boost efficiency of ASA synthetic sizing, J. Pulp Paper Sci., 59(4), 126. Seo, W.-S. and Cho, N.-S. (2005), Effect of water content on cellulose/AKD reaction, Appita J. 58(2), 122. Seppänen, R. and Tiberg, F. (1999), Mechanism of paper sizing by alkyl ketene dimer (AKD). The role of the spreading monolayer precursor and autophobicity, Proc. Scientific&Technical Advances in the Internal&Surface Sizing of Paper&Board, Florence, Italy, Pira International, Surrey, UK, Paper 2.
Seppänen, R., Tiberg, F. and Valignat, M. P. (2000), Mechanism of internal sizing by alkyl ketene dimers (AKD): The role of the spreading monolayer precursor and autophobicity, Nord. Pulp Pap. Res. J., 15(5), 452.
61
Seppänen, R., Kizling, J. and Tiberg, F. (2001), Influence of internal sizing on surface chemistry, adhesion and friction properties of paper, YKI Status report 3. Seppänen, R. and Zhmud, B. (2003), Wettability of AKD and ASA sized papers, Proc. Scientific&Technical Advances in the Internal&Surface Sizing of Paper&Board, Graz, Austria, Pira International, Surrey, UK, Paper 11. Seppänen, R., von Bahr, M., Tiberg, F. and Zhmud, B. (2004), Surface energy characterization of AKD-sized papers, J. Pulp Paper Sci. 30(3), 70. Shchukarev, A.V., Mattsson, R. and Ödberg, L. (2003), XPS imaging of surface diffusion of alkylketene dimer on paper surfaces, Colloids Surf. A, 219, 35. Shen, W., Filonanko, Y., Truong, Y., Parker, I. H., Brack, N., Pigram, P. and Liesegang, J. (2000), Contact angle measurement and surface energetics of sized and unsided paper, Colloids Surf. A, 173(1), 117. Shen, W., Brack, N., Ly, H., Parker, I.H., Pigram, P.J. and Liesegang, J. (2001), Mechanism of AKD migration studied on glass surfaces, Colloids Surf. A, 176(2-3), 129. Shen, W., Parker, I. H., Brack, N. and Pigram, P.J. (2001b), A simplified approach to understanding the mechanism of AKD sizing, Appita J. 54(4), 352. Shen, W., Xu, F. and Parker, I. H. (2002), The effect of the melting of AKD and its corresponding ketone on spreading behaviour, Appita J., 55(5), 375. Shen, W. and Parker, I. H. (2003), A study of the non-solid behaviour of AKD wax, Appita J. 56(6), 442.
Shen, W., Xu, F. and Parker, I. H. (2003), An experimental investigation of the redistribution behaviour of alkyl ketene dimers and their corresponding ketones, Colloid Surf. A, 212, 197.
Sjöström, L., Jacobs, A., Rådeström, R. and Nordlund, M. (2006), Effects of released organic substances on sizing efficiency - Influence of origin, composition and molecular properties of the organic material, Nord. Pulp Pap. Res. J., 21(5), 575. Ström, G., Carlsson, G. and Kiaer, M. (1992), Alkylketene dimer distribution in handsheets as determined by ESCA, Wochenblatt für Papierfabrikation, 120(15), 606. Taniguchi, R, Isogai, A., Onabe, F. and Usuda, M. (1993), Sizing mechanism of alkylketene dimers, Part 3. Appearance of sizing features of AKD-sized sheets, Nord. Pulp Pap. Res. J., 8(4), 352. Tiberg, F. and Cazabat, A. M. (1994a), Spreading Of Thin-Films Of Ordered Nonionic Surfactants - Origin Of The Stepped Shape Of The Spreading Precursor, Langmuir 10(7), 2301. Tiberg, F. and Cazabat, A. M. (1994b), Self-Assembly And Spreading Of Nonionic Trisiloxane Surfactants, Europhys. Lett. 25(3), 205. Toriz, G., Ramos, J and Young, R.A. (2004), Lignin-polypropylene composites Part II. Plasma modification of kraft lignin and particulate polypropylene, J. Appl. Polym. Sci., 91: 1920. Tougaard, S. and Sigmund, P. (1982), Influence Of Elastic And Inelastic-Scattering On Energy-Spectra Of Electrons Emitted From Solids, Phys. Rev. B 25(7), 4452. Tougaard, S. and Ignatiev, A. (1983), Concentration depth profiles by XPS: A new approach, Surf. Sci., 129, 355.
62
van Oss, C. J. (2002), Use of the combined Lifshitz-van der Waals and Lewis acid-base approaches in determining the apolar and polar contributions to surface and interfacial tensions and free energies, J. Adhes. Sci. Techn., 16(6), 669. von Bahr, M., Seppänen, R., Tiberg, F. and Zhmud, B. (2004), Dynamic wetting of AKD-sized papers, J. Pulp Paper Sci., 30(3), 74. Voutilainen, P. (1996), Competitive Adsorption of Alkyl Ketene Dimer (AKD) on Pulp Fibers and CaCO3 Fillers, Proc. Intl. Paper and Coating Chemistry Symposium, Pointe Claire, Quebec, CPPA, 195. Wågberg, L. and Westerlind, C. (2000), Spreading of drops of different liquids on specially structured papers, Nord. Pulp Pap. Res. J., 15(5), 598. Wasser, R. B. (1987), Reactivity of Alkenyl Succinic Anhydride: Its Pertinence with Respect to Alkaline Sizing, J. Pulp Paper Sci., 13(1), 29. Wasser, R. B. and Brinen, J. S. (1998), Effect of hydrolyzed ASA on sizing in calcium carbonate filled paper, Tappi J., 81(7), 139. Wentzel, R. N. (1936), Resistance of solid surfaces to wetting by water, Ind. Eng. Chem., 28, 988. Werner, O., Wågberg, L. and Lindström, T. (2005), Wetting of structured hydrophobic surfaces by water droplets, Langmuir, 21, 12235. Withiam, M. C. (1991), Effect of Fillers on Paper Friction Properties, Tappi J., 74(4), 256. Yu, L. P. and Garnier, G. (1997), Mechanisms of internal sizing with alkyl ketene dimers: The role of vapour deposition, The fundamentals of papermaking materials, vol. 2, Baker, C.F. (ed.), Pira International, Surrey, pp. 1021-1046. Yu, L. P. and Garnier, G. (2002), The Role of Vapour Deposition During Internal Sizing: A Comparative Study between AKD and ASA, J. Pulp Paper Sci., 28(10), 327. Zeggane, S. and Delamar, M. (1987), Composition Depth Information In Organic Materials By Measurement Of Xps Background Signal, Appl. Surface Sci., 29(4), 411. Zenkiewicz, M. (2007), Comparative study on the surface free energy of a solid calculated by different methods, Polymer Testing, 26, 14. Zeno, E., Carre, B. and Mauret, E. (2005), Influence of surface active substances on AKD sizing, Nord. Pulp Pap. Res. J., 20(2), 253. Zhang, H., Kannangara, D., Hilder, M., Ettl, R. and Shen, W. (2007), The role of vapour deposition in the hydrophobization treatment of cellulose fibres using alkyl ketene dimers and alkenyl succinic acid anhydrides, Colloids Surf. A, 297(1-3), 203. Zhmud, B., Seppänen, R. and Tiberg, F. (2001), Hydrolysis and sizing action of cellulose-reactive sizes, Proc. Scientific&Technical Advances in the Internal&Surface Sizing of Paper&Board, Prague, the Czech republic, Pira International, Surrey, UK, Paper 6. Zimmerman, P.A., Hercules, D. M. and Loos, W. O., (1995) Direct quantitation of alkyl ketene dimers using time-of-flight secondary ion mass spectrometry, Anal. Chem. 67, 2901.