Utilisation of tube flow fractionation in fibre and ...
Transcript of Utilisation of tube flow fractionation in fibre and ...
ABCDEFG
UNIVERS ITY OF OULU P.O.B . 7500 F I -90014 UNIVERS ITY OF OULU F INLAND
A C T A U N I V E R S I T A T I S O U L U E N S I S
S E R I E S E D I T O R S
SCIENTIAE RERUM NATURALIUM
HUMANIORA
TECHNICA
MEDICA
SCIENTIAE RERUM SOCIALIUM
SCRIPTA ACADEMICA
OECONOMICA
EDITOR IN CHIEF
PUBLICATIONS EDITOR
Senior Assistant Jorma Arhippainen
Lecturer Santeri Palviainen
Professor Hannu Heusala
Professor Olli Vuolteenaho
Senior Researcher Eila Estola
Director Sinikka Eskelinen
Professor Jari Juga
Professor Olli Vuolteenaho
Publications Editor Kirsti Nurkkala
ISBN 978-951-42-9448-8 (Paperback)ISBN 978-951-42-9449-5 (PDF)ISSN 0355-3213 (Print)ISSN 1796-2226 (Online)
U N I V E R S I TAT I S O U L U E N S I SACTAC
TECHNICA
U N I V E R S I TAT I S O U L U E N S I SACTAC
TECHNICA
OULU 2011
C 382
Ossi Laitinen
UTILISATION OF TUBE FLOW FRACTIONATION IN FIBRE AND PARTICLE ANALYSIS
UNIVERSITY OF OULU,FACULTY OF TECHNOLOGY,DEPARTMENT OF PROCESS AND ENVIRONMENTAL ENGINEERING
C 382
ACTA
Ossi Laitinen
C382etukansi.kesken.fm Page 1 Thursday, May 19, 2011 1:40 PM
A C T A U N I V E R S I T A T I S O U L U E N S I SC Te c h n i c a 3 8 2
OSSI LAITINEN
UTILISATION OF TUBE FLOW FRACTIONATION IN FIBRE AND PARTICLE ANALYSIS
Academic dissertation to be presented with the assent ofthe Faculty of Technology of the University of Oulu forpublic defence in Auditorium TA105, Linnanmaa, on 17June 2011, at 12 noon
UNIVERSITY OF OULU, OULU 2011
Copyright © 2011Acta Univ. Oul. C 382, 2011
Supervised byProfessor Jouko NiinimäkiDoctor Tuomas Stoor
Reviewed byProfessor Samuel SchabelDoctor Jaana Sjöström
ISBN 978-951-42-9448-8 (Paperback)ISBN 978-951-42-9449-5 (PDF)http://herkules.oulu.fi/isbn9789514294495/ISSN 0355-3213 (Printed)ISSN 1796-2226 (Online)http://herkules.oulu.fi/issn03553213/
Cover DesignRaimo Ahonen
JUVENES PRINTTAMPERE 2011
Laitinen, Ossi, Utilisation of tube flow fractionation in fibre and particle analysis. University of Oulu, Faculty of Technology, Department of Process and EnvironmentalEngineering, P.O. Box 4300, FI-90014 University of Oulu, FinlandActa Univ. Oul. C 382, 2011Oulu, Finland
Abstract
The tube flow fractionation method used in this thesis is one of the field flow separation processes,which can be used to fractionate particles over the entire size range of pulp particles (1–5000 µm)with high selectivity. The classification of particles in different size categories enables thefractional analysis of samples. According to this thesis, particle length is the most significantindividual shape factor in particle classification for different size categories using tube flowfractionation.
The fractional approach investigated in this thesis offers an interesting tool for investigatingdifferent pulp types more thoroughly than many other conventional laboratory methods such asthe Bauer-McNett classification of pulp or the hyperwashing of DIP samples. The tube flowfractionator can be used to replace the Bauer-McNett device to determine the mass percentages ofpulp sample fractions. Furthermore, pulp fractionation can be done quickly with the tube flowfractionation method and the results achieved are also comparable to the Bauer-McNettclassification.
The utilisation of recycled raw material in paper production is likely to continue to increase inthe near future. This fact leads to even more heterogeneous raw material becoming one of themajor challenges in deinked pulp production. Changes in the quality of raw material such as ageor paper type have a notable effect on the deinkability of pulp. This creates a need to develop newmethods for measuring DIP properties; in particular, information about the ink-releasing rate andink particle size would be useful in controlling the deinking process. If DIP is fractionated withtube flow fractionation, it is possible to study the ink size distribution and ink content inside eachfraction, thus obtaining fractional information on the ink release rate and ink particle size.
A comparison of different DIP raw materials and process solutions can be performed with thefractional analysis presented in this thesis. It was noticed that aging had a negative effect on theslushing rate of LWC and SC raw material, but not on ONP. The ink detachment of LWC and SCwas poorer when furnishes aged. However, the aging of LWC and SC did not result in a significantfragmentation of ink particles. In contrast, strong ink fragmentation is typical of aged ONPfurnishes. The fractional approach gives new insight into deinking mechanisms, and could beessential for finding optimal deinking process conditions, for example in a pulping process orflotation environment. The method would also enable enhanced control of different DIP processstages and would contribute to creating a more efficient process with better waste management andimproved final product quality.
Keywords: Aging, Deinked pulp, fractional analysis, ink detachment, ink fragmentation,ink particles, pulping, slushing, tube flow fractionation
Laitinen, Ossi, Putkivirtausfraktioinnin hyödyntäminen kuitu- ja partikkelianalytiikassa. Oulun yliopisto, Teknillinen tiedekunta, Prosessi- ja ympäristötekniikan osasto, PL 4300, 90014Oulun yliopistoActa Univ. Oul. C 382, 2011Oulu
Tiivistelmä
Tässä työssä käytetyllä putkivirtausfraktioinnilla, joka on eräs kenttävirtausfraktioinnin sovellus,voidaan luokitella hyvin selektiivisesti erikokoisia (1–5000 µm) partikkeleita. Partikkeleidenluokittelu partikkelikooltaan erikokoisiin jakeisiin mahdollistaa näytteiden fraktionaalisenanalyysin. Tässä työssä todettiin, että partikkelin pituus on tärkein yksittäinen muototekijä, jokavaikuttaa partikkelien fraktioitumiseen eri kokoluokkiin putkivirtausfraktioinnissa.
Työssä tutkittu fraktionaalinen menetelmä mahdollistaa paljon syvällisemmänpaperisulppujen analyysin kuin monet perinteiset laboratorioanalyysit, kuten Bauer McNett-luokittelu tai siistausmassojen hyperpesu. Putkivirtausfraktionaattorilla voidaan korvata BauerMcNett -luokittelu massasulppujen eri fraktioiden massaosuuksien määrityksessä. Lisäksimassojen fraktiointi putkivirtausfraktionaattorilla on nopea ja luotettava menetelmä, ja saaduttulokset ovat vertailukelpoisia Bauer McNett -luokittelulla saatujen tulosten kanssa.
Kierrätyspaperiraaka-aineen kasvava hyödyntäminen paperin tuotannossa jatkuneelähivuosina. Tämä tarkoittaa, että siistausprosessissa hyödynnettävän raaka-aineen laatuvaihtelutaiheuttavat entistä suurempia haasteita. Muutokset raaka-aineen laadussa, kuten paperin ikä taikäytetty paperityyppi vaikuttavat huomattavasti massan siistattavuuteen. Näistä tekijöistä johtuenon tarvetta kehittää uusia mittausmenetelmiä määrittää siistausmassojen ominaisuuksia.Erityisesti musteen irtoaminen ja mustepartikkeleiden kokojen määritys voisi auttaamusteenpoistoprosessin ohjauksessa. Jos siistausmassa fraktioidaan putkivirtausfraktionaattorilla,on mahdollista tutkia mustekokojakaumia ja mustepitoisuuksia eri fraktioissa ja näin saadaantietoa mustepartikkeleiden irtoamisesta ja mustepartikkelien koosta.
Fraktionaalisella analyysilla on tässä työssä tutkittu eri siistausmassojen raaka-aineita japrosessiratkaisuja. Työssä huomattiin, että LWC- ja SC-paperin ikääntymisellä on negatiivinenvaikutus paperin sulputtumisnopeuteen pulpperissa, mutta sanomalehtipaperilla vastaavaa ilmiötäei huomattu. Musteen irtoaminen oli huonompaa LWC- ja SC-paperin ikääntyessä. Kuitenkaanikääntyminen ei vaikuttanut LWC- ja SC-paperin mustepartikkeleiden pilkkoutumiseenmerkittävästi. Sen sijaan sanomalehtipaperilla havaittiin voimakasta mustepartikkeleidenpilkkoutumista raaka-aineen ikääntyessä.
Fraktionaalinen tutkiminen antaa uuden näkökulman musteenpoistomekanismeihin ja voisiolla hyödyllinen, kun etsitään optimaalisia musteenpoisto-olosuhteita esimerkiksi pulpperointiintai flotaatioon. Menetelmä voisi mahdollistaa myös eri siistausprosessin osaprosessiensäätämisen. Lisäsi menetelmän avulla voitaisiin saada tehostettua prosesseja ja parannettua laatuasekä saada prosesseista ympäristöystävällisempiä.
Asiasanat: fraktionaalinen analyysi, ikääntyminen, kuituuntuminen, muste partikkelit,musteen irtoaminen, musteen pilkkoutuminen, pulpperointi, putkivirtausfraktiointi,siistausmassa
7
Acknowledgements
The research work reported in this thesis was carried out at the Fibre and Particle
Engineering Laboratory, University of Oulu, during the period 2005–2010. The
work was financed by Metso Automation, Metso Paper, Kemira and the Finnish
Funding Agency for Technology and Innovation, all of which are gratefully
acknowledged.
First of all, I would like to express my gratitude to my supervisor Prof. Jouko
Niinimäki for his guidance during this work and for the opportunity to work in his
group. I am also grateful to Dr. Tuomas Stoor for his valuable advice. Special
thanks also go to the personnel of Metso Automation for their technical support
and advice during the thesis work.
I am most grateful to the reviewers of this thesis, Prof. Samuel Schabel from
Technical University of Darmstad and Dr. Jaana Sjöström from Metso Paper. I
would like to thank Sue Pearson and Mike Jones for proofreading the English
language of this manuscript. I also wish to thank the co-authors of the articles.
I would like to thank all my colleagues I have worked with during these years
at the Fibre and Particle Engineering Laboratory. In particular, I would like to
thank Mr. Jarno Karvonen and Mr. Jani Österlund for their help in the
experimental work in the laboratory and pilot plant.
Finally, I would like to express my warmest thanks to my family and friends,
who have encouraged and supported me during the time I have been working on
this thesis.
Oulu, May 2011 Ossi Laitinen
8
9
List of abbreviations and symbols
BMN Bauer-McNett
cc correlation coefficient of a model
CCD camera Charge-coupled device camera
CSF Canadian standard freeness
DIP Deinked pulp
EFGP Ecole Française de Papeterie et des Industries Graphiques
FiberLab Kajaani FiberLab fibre length analyser
F or FR Fraction
FSA Kajaani FSA fibre length analyser
GW Groundwood
HW Hyperwashed pulp
INGEDE International Association of the Deinking Industry
KCL Finnish Pulp and Paper Research Institute
LWC Light weight coated paper
MLR-model Multilinear regression model
ONP Old newspaper
Paprican Pulp and Paper Research Institute of Canada
PI-flow control Proportional plus integral flow control
PM Paper machine
Q² Predictability of a model
R² Correlation coefficient of a model
R Refiner
Re-number Reynolds number
RI Residual ink
RR TMP reject refiner
SC Supercalendered paper
STFI Swedish Pulp and Paper Research Institute
TAPPI Technical Association of the Pulp and Paper Industry
TMP Thermomechanical pulp
ave(err) average error [%] D diameter [m]
°dH water hardness [°dH], German degree of hardness
ISO-% brightness of pulp [ISO-%]
pcs pieces [pcs]
10
ppm parts per million [ppm]
Q volume flow [m³/s]
ρ density of fluid [kg/m³]
RI residual ink of pulp [ppm]
µ dynamic fluid viscosity [kg/ms]
11
List of original publications
This thesis is based on the following publications, which are referred to in the text
by their Roman numerals:
I Laitinen O, Kemppainen K, Stoor T & Niinimäki J (2011) Fractionation of pulp and paper particles selectively by size. BioResources 6(1): 672–685.
II Laitinen O, Löytynoja L & Niinimäki J (2006) Tube flow fractionator − A simple method for laboratory fractionation. Paperi ja Puu - Paper and Timber. 88(6): 351–355.
III Laitinen O, Körkkö M & Niinimäki J (2008) The effects of aging in different raw material furnishes. Progress in Paper Recycling 18(1): 9–15.
IV Laitinen O, Kemppainen K, Körkkö M, Stoor T & Niinimäki J (2009) The effect of pre-wetting of recycled raw material before pulping in a deinking process. Progress in Paper Recycling 18(4): 24–32.
All the manuscripts for the publications were written by the primary author of this
thesis, whose main responsibilities were the experimental design, data analysis
and reporting of results.
12
13
Contents
Abstract
Tiivistelmä
Acknowledgements 7
List of abbreviations and symbols 9
List of original publications 11
Contents 13
1 Introduction 15
1.1 Background ............................................................................................. 15
1.2 Aim of the study ...................................................................................... 16
1.3 Initial assumptions .................................................................................. 16
1.4 Research environment ............................................................................. 17
1.5 Outline of thesis ...................................................................................... 17
2 Tube flow fractionation mechanism – present understanding 19
2.1 From phenomenon to measurement device ............................................. 19
2.2 Principle of fractionation of fibre suspensions ........................................ 20
3 Materials and methods 23
3.1 Tube flow fractionation equipment ......................................................... 23
3.1.1 Characterisation of different fractions .......................................... 25
3.2 Microscope and image analysis of fractionated samples ........................ 25
3.3 Brightness and residual ink determination of samples ............................ 26
3.4 Materials used in fractionation time studies ............................................ 26
3.5 Materials used in calculation of mass proportions .................................. 27
3.6 Materials used in fractional analysis studies ........................................... 28
3.6.1 One-loop pilot process sample preparation .................................. 29
3.6.2 Procedure for pre-wetting and laboratory pulping ........................ 30
3.6.3 Chemicals used during sample preparation procedures ................ 31
4 Results 33
4.1 Fractionation time of different particles .................................................. 33
4.1.1 MLR model and statistical analysis .............................................. 33
4.1.2 Verification of the MLR model .................................................... 36
4.2 Calculation of mass proportions in different fractions ............................ 37
4.2.1 Modelling the BMN classifier with optical signals ...................... 38
4.3 Fractional analysis of different DIP fractions ......................................... 40
4.3.1 Slushing of different raw materials .............................................. 40
4.3.2 Ink detachment and fragmentation ............................................... 41
14
4.3.3 Ink particle distributions in different fractions ............................. 43
5 Discussion 47
5.1 Fractionation time of different particles in tube flow
fractionation ............................................................................................ 47
5.2 Calculation of mass proportions of different fractions ............................ 48
5.3 Fractional analysis of different DIP fractions .......................................... 49
5.3.1 Effect of aging on DIP furnishes .................................................. 49
5.3.2 Slushing of furnishes with a pilot-scale pulper ............................. 50
5.3.3 Ink detachment and fragmentation of inks ................................... 51
5.3.4 Ink particle size distributions in different fractions ...................... 52
6 Conclusions and proposed applications for future research 55
References 57
Original publications 61
15
1 Introduction
1.1 Background
The tube flow fractionation method can be used to fractionate particles over the
entire size range of pulp particles (1–5000 µm). The flow rate is higher in the
centre of the tube than near the tube wall. It has been assumed that a slight
turbulence in the transient flow conditions in the fractionation system moves the
particles randomly inside the flow. Some particles drift close to the tube walls,
whereas others stay in the middle of the tube. Of the particles that are close to the
tube wall, the larger ones have a higher probability of being captured by the faster
moving middle flow than the smaller ones, due to their larger dimensions. Thus
the more large dimensions a particle has, the higher is the probability of its being
captured by the faster moving middle flow and finally exiting the tube faster than
fine particles (Olgård 1970, Pascal & Silvy 1991, Pascal & Silvy 1993, Krogerus
et al. 2003, Paper I). In practice this means that the method enables the
fractionation of particles to different size categories, mostly according to their
largest dimension (Paper I), which is typically length when classifying pulp
particles. After fractionation, each fraction can be also collected for further
analysis using microscopes, spectrophotometers or fibre analysers.
Screen fractionation, with equipment such as the Bauer-McNett apparatus,
has been used traditionally in pulp analysis. The major weaknesses of screen
fractionations are the laborious nature of measurement and the poor repeatability
between individual fractionation devices (Bos 1966, Levlin 1982, Clark 1985,
Gooding & Olson 2001). These weaknesses could even be reduced by the use of
screen fractionators in research and mill laboratories. Pulp fractionation can be
done quickly with the tube flow fractionation method and results are comparable
(Paper II). The fractional approach with the tube flow fractionation method also
offers an interesting tool to investigate different pulp types in a more thorough
way than many other conventional methods.
Utilisation of recycled raw material in paper production will continue to
increase in the near future (Leberle 2008). This will lead to even more
heterogeneous raw material becoming one of the major challenges in deinked
pulp production. Changes in the quality of raw material such as age or paper type
have a notable effect on the deinkability of pulp (Dorris 1999, Dorris & Page
2000a, Haynes 2000, Ben & Dorris 2006). These facts create a need to develop
16
new methods for measuring DIP properties; in particular, information about the
ink-releasing rate and ink particle size would be the most useful in controlling the
deinking process. Measuring these properties directly from the pulp is challenging
with traditional methods, but this new fractional method offers the possibility of a
fractional study of DIP properties. If DIP is fractionated according to particle size,
it is possible to study ink size distribution and ink content inside each fraction,
and thus fractional information on the ink release rate and ink particle size is
obtained (Laitinen et al. 2006, Laitinen et al. 2007a). In addition, information on
for example how much ink exists in each fibre length class and the attached ink
size distribution would be essential for investigating deinked pulp processes. The
fractional approach gives new insight into deinking mechanisms, and would be
essential for finding optimal deinking process conditions.
1.2 Aim of the study
The aim of this thesis was:
1. to obtain new information about the fundamental phenomena governing the
fractionation time of particles with various dimensions in a tube flow
fractionator (Paper I),
2. to find out whether the tube flow fractionator could be utilised in the mass
percentage determination of fibre fractions of pulp samples (Paper II),
3. to find out whether the tube flow fractionator could be utilised in the
fractional analysis of different DIP raw materials and DIP process solutions
(Papers III-IV).
1.3 Initial assumptions
Due to the complex nature of tube flow fractionation, the phenomena existing in
pulp suspensions and modelling the theoretical flow of the phenomena is very
problematic. Hence this thesis is based on an empirical modelling of tube flow
fractionation. The prototype fractionation device used in this thesis was
manufactured by Metso Automation Corp. The essential parameters during
fractionation were as follows: flow velocity, temperature, sample volume and
consistency, which were all maintained precisely at a constant level, as each
parameter has a direct effect on the fractionation phenomenon inside the tube.
17
Flake-like or undefiberised particles exit the tube faster than smaller
individual long fibres. As a result, it is possible to determine the flake content as a
mass percentage after fractionation. In the case of DIP, it was assumed that the
long and short fibre fractions contained all the attached ink particles and that all
the detached ink particles had drifted to the finest fraction. Due to the particle size
fractionated in this thesis (typically over 1 µm), it can be assumed that the
Brownian motion effect was minimal, and likewise the particle attraction forces,
because of the fairly turbulent flow during fractionation and also the low
consistency of the classified sample.
The raw materials used in this thesis were assumed to represent typical
furnishes in the pulp and paper industry. Trials and sampling were done during
normal process operating conditions and it was assumed that the measured values
of consistency, temperature and process variables describe the common situation
of researched processes well.
1.4 Research environment
The author’s work on tube flow fractionation consists of modelling (Papers I-II)
and application parts (Papers III-IV). In the modelling part the effect was
investigated of different particle dimensions on the fractionation time during tube
flow fractionation (Paper I) and whether tube fractionation could replace screen
fractionation techniques such as the Bauer-McNett apparatus (Paper II).
In the application part, the utilisation of the tube flow fractionation method
for analyses of different DIP raw materials and DIP process solution was studied.
The effects of aging on different DIP furnishes was investigated in Paper III.
Furthermore, in Paper IV the potential to improve ink detachment by using a pre-
wetting stage before the actual pulping process was clarified.
1.5 Outline of thesis
Chapter 1 of this thesis presents a short introduction to the study, the aim of the
studies, the research environment and an outline of the thesis. Chapter 2
summarises the present understanding of the tube flow fractionation mechanism
while Chapter 3 presents the materials and methods used in this study. Chapter 4
summarises the main results achieved during the work. The interpretation and
discussions are presented in Chapter 5. Finally, conclusions and suggestions for
future work are presented in Chapter 6.
18
19
2 Tube flow fractionation mechanism – present understanding
2.1 From phenomenon to measurement device
The separation phenomenon in flowing pulp suspension was noticed as early as
the 1950s by Mason at Paprican (1950) and Johansson & Kubat at STFI (1956).
In the seventies, the first prototype of a tube flow fractionation device was
developed at EFGP in Grenoble (Silvy & Pascal 1968). The fractionation method
used successfully by the French was based on the utilisation of continuous water
flow in a long tube. The potential of using a fractionation device like this in pulp
analysis was clearly pointed out in their research. At the same time, the
fractionation of pulp with liquid flow was also reported by the Axel Johnson
Institute for Industrial Research in Sweden (Olgård 1970), where two different
types of fractionation prototypes were studied. These were called a hose and a
helical disk apparatus, which did not use continuous water flows, but several
separated independent pulp plugs. Research made by Olgård presented that pulps
could be divided into two fractions with different properties than those of the
original pulp sample. Even though the results of the studies made in the seventies
were very encouraging, it was still at that time unclear, whether these above-
mentioned inventions would lead to any practical applications.
Nevertheless, fractionation investigations continued more or less intensively
at EFGP in Grenoble in the following decades and researchers became more
concerned with specific fractionation conditions and the tube flow fractionation
technique (Pascal & Silvy 1991, Pascal & Silvy 1993). As a result of this research,
they developed a practical and modernised version of the fractionation device at
EFGP (Silvy & Pascal 1992).
Metso Automation Corporation became interested in the fractionation method
invented at EFGP and bought the licence at the end of the nineties, entitling them
to utilise the tube flow fractionation technique. Metso Automation then
constructed their first prototype of the fractionation device called a tube flow
fractionator utilising the fractionation conditions suggested by the French
researchers. The first version of tube flow fractionator was already equipped with
accurate control systems and high-resolution imaging technology (Laitinen 2002).
This new prototype immediately arose interest in pulp and paper research
20
institutes, because the potential of the device was clear and it was also easy, fast
and reliable.
Krogerus et al. at KCL (2003) were interested in utilising the new tube flow
fractionator to characterise pulp particles. They also speculated that the
fractionation device could be an interesting tool for fibre research and process
control.
At the same time, one prototype was installed in the Fibre and Particle
Engineering Laboratory at University of Oulu. From that moment, the tube flow
fractionator was adopted for use in the research work pertaining to this thesis and
the main results of this thesis work were performed using this prototype device.
Further research with regard to putting on the market a new analyser, based on the
work on fractionation technology described in this thesis is still going on
(Tiikkaja 2007, Niskanen 2010). During the past ten years, a promising laboratory
device has finally come very close to completion as a modern analyser for
laboratory- and mill-scale analyses.
2.2 Principle of fractionation of fibre suspensions
Tube flow fractionation method is a field flow separation process, which can be
used to fractionate particles over the entire size range of pulp (1–5000 µm) with
high selectivity (Papers I and II). In the tube flow technique the particles are
separated axially so that the largest ones accumulate at the front end of the flow
plug and the smallest ones at the rear end (Mason 1950, Johansson and Kubat
1956, Olgård 1970, Johansson et al. 1970).
21
Fig. 1. Simplified schematic description of tube flow fractionation (modified from
Paper I, © 2011 BioResources).
When the sample is injected into a tube full of eluant water, the particles are
distributed randomly (see Fig. 1a). As the flow begins, particles are carried with
the flow and a slight turbulence in the transition flow regime begins to move the
particles randomly inside the flow. It is assumed that small particles that drift
close to the tube wall may not easily be captured again by the faster middle flow
and will thus stay in the zone where the flow velocity is lower. Large particles
may also drift close to the tube wall but may be recaptured by the faster middle
flow due to their greater dimensions. The probability of being captured by the
faster middle flow is higher for particles with a single long dimension (e.g. fibres)
or multiple long dimensions (e.g. paper flakes) than for particles with short
dimensions (e.g. fines). Thus, the large particles tend to concentrate at the front
end of the flow (Fig. 1b) and finally come out from the tube first (Fig. 1c).
It has been observed that the tube flow fractionation method to fractionate
pulp particles selectively mostly according their length (Paper I). For efficient
particle separation, the flow conditions should differ from the laminar flow profile.
Reynolds numbers within the range 1,000 to 10,000 for water are considered to be
the most efficient for the separation of pulp particles (Pascal & Silvy 1993). The
total consistency of the sample must be below 0.5% (depending slightly on the
pulp type), as no selective separation occurs above this level, since the fibres form
a coherent network (Kerekes and Schell 1992, Krogerus et al. 2003). In the case
of fibre suspensions, fibres will also interact with each other, so the diameter of
…
…
…
…
…
…
…
…
…
…
…
……
…
…
…
…
…
……
…
…
…
…
…
…
…
…
…
…a)
b)
c)
Flow direction
Vmin
Vmin
Vmax
Flow direction
22
the tube (as well as the concentration of particles) has an effect on the efficiency
of the separation (Hemström et al. 1976, Lee & Duffy 1976, Jäsberg et al. 2008).
The fractionation phenomenon used in tube flow fractionation differs from
that in hydrodynamic chromatography (Tijssen et al. 1986, Provder 1997,
Schimpf et al. 2000, Lin & Jen 2002). In tube flow fractionation, hydrodynamic
forces, which in this case means the forces holding up the particles, play a more
important role than they do in hydrodynamic chromatography. This means in
practice that separable particles try to keep away from the tube wall. Particles that
settle against the bottom wall of the tube do not progress through the tube as
easily as the particles drifting with the water flow.
Because the pulp particle components separated by this method are quite
large (over 1 µm), the Brownian motion effect and particles attraction forces can
be assumed to be minimal (Schimpf et al. 2000). Nevertheless, the densities of
fractionated particles must be quite close to the value for water to ensure that lift
forces are the most dominant force to keep particles away from the so-called
lubrication layer, which is a thin layer of pure water created next to the tube wall
(Jäsberg et al. 2008).
23
3 Materials and methods
The tube flow fractionation equipment is presented in Chapter 3.1. The materials
used in the modelling sections are presented in Chapters 3.2 and 3.3 and finally,
the materials used in DIP fractional analysis are presented in Chapter 3.4.
3.1 Tube flow fractionation equipment
The tube flow fractionator (Metso Automation, Finland) used in the experiments
employs a method in which a defined amount of pulp or sample (volume of 50 ml
with a consistency of 0.3%, which is the equivalent of 150 mg absolute dry
sample) is injected into the water flow through a 100-metre-long Teflon tube
wound onto a wheel (Laitinen et al. 2006, Laitinen et al. 2007b). The sample can
be injected into the tube, when a three-way ball valve is opened. After the sample
is injected into the tube, the three-way ball valve is turned so that the continuous
water flow can transport the sample through the tube (see Fig. 2).
The pulp sample is fractionated into different size categories as it flows
through the tube and when the pulp-water mixture exits the long tube, it is divided
into size fractions according to the setup of the device. The fibre fractions can
then be analysed with other equipment such as a fibre analyser or microscope.
The mass of the pulp in the fractions can be measured by an optical
consistency transmitter, including depolarisation and absorption signals. Signal
processing of the depolarisation and absorption signals is presented as one part of
the results in Chapter 4.2 (Paper II). The CCD camera resolution used in the tube
flow fractionator is 9 µm. A schematic illustration of the fractionation apparatus
and measurement devices is presented in Fig. 2.
Fig. 2. Schematic illustration of the tube flow fractionator. (Paper I, © 2011
BioResources).
FR1
Sample injection
Plastic tube
CCD camera
Signal processing
Temperature and flow control
Water inlet
FR2 FR3 FR4
Fraction collectors Water outlet
FR1
Sample injection
Plastic tube
CCD camera
Signal processing
Temperature and flow control
Water inlet
FR2 FR3FR2 FR3 FR4
Fraction collectors Water outlet
24
The parameters of importance during fractionation, including flow velocity,
temperature, sample volume, and consistency, were all maintained precisely at a
constant level, as each parameter has a direct effect on fractionation. The
temperature of the eluant water used was controlled by mixing hot and cold tap
water and the flow velocity was controlled with a PI flow controller. The essential
variables and parameters during fractionation are presented in Table 1.
Table 1. Parameters and variables of the tube flow fractionation device. (Paper I, ©
2011 BioResources).
Parameter or variable Value
Flow velocity 5.7 l/min
Volume flow 9.5 · 10-5 m3/s
Water temperature 25 ºC
Water viscosity [T = 25 ºC]1 8.9040 · 10-4 kg/ms
Water density [T = 25 ºC]1 997.05 kg/m3
Length of fractionation tube 100 m
Diameter of fractionation tube 16 mm
Reynolds number [in described conditions]2 about 8500 1Values checked from CRC Handbook of Chemistry and Physics. 58th edition 1977-1978. 2Determined
with e.g. (1).
It is well known that when the flow velocity and temperature are both constant,
certain fibre sizes always exit the flow tube at the same time. Besides the
fractionation time, it is also possible to regard the amount of water that has
flowed through the fractionation hose as a relevant parameter. Thus the
fractionation results can be expressed as a function of either fractionation time
(Paper I) or the volume of water that has flowed through the tube (Papers II-IV).
The particles were fractionated at a flow rate of 5.7 l/min with a Reynolds
number of around 8,500 (Liukkonen 2006, Paper I). This flow rate enables all of
the particles to be separated by size during the fractionation process. The
Reynolds number can be determined using Equation 1:
,Re
VD
(1)
where V is the flow velocity and ρ is the density of fluid (water). In addition, D is
the diameter of the fractionation tube and μ is the dynamic fluid viscosity (water).
25
3.1.1 Characterisation of different fractions
The fractionated samples are typically divided into four bins where the first
contains the biggest particles and the last contains the smallest (Paper II). In the
case of DIP, fibre fractions from the first to the fourth bin can be characterised to
contain flakes (FR1), long fibres (FR2), short fibres (FR3) and fines/fillers (FR4),
respectively. On the other hand, ink particles from the first to the last fraction can
be described as dense printing on the flakes (FR1), ink attached to long fibres
(FR2), ink attached to short fibres (FR3) and free ink or ink particles attached to
fines (FR4) (Papers III-IV). Table 2 presents the tube flow fractionator fractions
and average particle size of typical DIP fractions according to FiberLab (Metso
Automation, Finland). When the fractionation process is started, the flow meter
starts to measure the total flow of water through the fractionation tube and after
complete fractionation in the tube, a certain volume of water represents a certain
size of particles. Due to this fact, the boundaries for each fraction, known as litre
ranges, can be fixed (Table 2).
Table 2. The litre ranges for tube flow fractionator fractions and average fibre length of
each fraction. (Modified from Paper IV, © 2009 Tappi).
Tube flow fractionator Tube flow fractionator
litre ranges
Bauer-McNett
classifier
Average fibre length [mm]
1st fraction, flakes (FR1) 14.80-16.30 R12 Unable to analyse
2nd fraction, long fibres (FR2) 16.30-17.50 12-48 1.85
3rd fraction, short fibres (FR3) 17.50-18.30 48-200 0.60
4th fraction, fines (FR4) 18.30-20.30 P200 0.18
3.2 Microscope and image analysis of fractionated samples
The fractionated samples were filtered out onto membrane filter foils (the
diameter and pore size were 50 mm and 0.45 μm, respectively) in order to
investigate the samples with a microscope. After drying each sample was
microphotographed (Leica DFC320) using a constant picture ratio. The images of
the samples were analysed using image analysis software (Matrox Inspector) that
calculated the number, length, width and area of dark particles (i.e. ink particles).
The threshold value was kept constant. Particles in the different size categories
were counted on the basis of the image analysis. The smallest particle size that
could be analysed with the method described above was 3 μm (9 μm²).
26
3.3 Brightness and residual ink determination of samples
When analysing the pulp samples, pulp pads (225 g/m²) were prepared according
to INGEDE Method 1. The pads were analysed with a Lorenzen & Wettre
Elrepho 070 spectrophotometer using a 457 nm illumination for brightness and
700 nm for residual ink (RI). RI values were measured using a constant scattering
coefficient (s = 55 m²/kg). Residual ink was determined according to Equation 2
(Körkkö et al. 2011):
,713.247.0849.657.08919.0 2 kkRI (2)
where k is the absorption coefficient with a constant scattering coefficient (when s
= 55 m²/kg). Some of the pulp samples were also hyperwashed (HW) with a 150-
mesh wire screen in order to measure the attached ink.
3.4 Materials used in fractionation time studies
The materials used in this section were various pulp, paper and peat particles (see
Table 3). Samples of peat particles were taken from a peat swamp and they were
selected for this test because their shape was close to rectangular. Pulp samples
were obtained from different pulp manufacturing processes: Softwood kraft pulp
from Scots pine, hardwood kraft pulp from Eucalyptus species,
Thermomechanical pulp (TMP), Groundwood (GW) and Deinked pulp from
newsprint production (DIP). In addition, the fractionation times of different sized
pieces of copy paper were also investigated.
27
Table 3. Particles and their dimension ranges. (Paper I, © 2011 BioResources).
Sample Shape Length range [μm]1 Width range [μm]2 Thickness range
[μm]3
Peat particles rectangular 25-932 9-203 8-170
Softwood kraft
(pine)
stick 140-3410 15-34 15-34
Hardwood kraft
(eucalyptus)
stick 200-1280 14-18 14-18
TMP stick 180-3500 17-32 17-32
GW stick 110-3230 16-41 16-41
DIP stick 170-3410 15-37 15-37
Paper disc round disc 2047-4530 2047-4530 100
Paper piece rectangular 1080-4310 840-1020 100 1The largest dimension of the particle. 2The shorter side of the plane in which the length was measured.
The length and width dimensions of the paper discs are represented by the diameter of the disc. 3Perpendicular to the plane in which the length and width were measured. The thickness of the pulp
particles was assumed to be the same as their width.
Since the peat samples included some very large particles, the sample supplier
had ground the samples in order to obtain smaller sizes (mostly under 1000 µm).
After pre-treatment, the ground peat samples were then divided into size
categories by means of sieve shakers (Retsch) of 32 µm, 45 µm, 63 µm, 90 µm,
125 µm, 250 µm and 500 µm in order to achieve narrow particle size distributions.
Copy paper (grammage 80 g/m²) was used in order to obtain samples with a
uniform shape. This was achieved by cutting rectangular pieces using a paper
guillotine and small, round discs of various size categories (diameter from 2 mm
to 4.5 mm) using perforator blades.
3.5 Materials used in calculation of mass proportions
The virgin pulps used in the calculation of mass proportions were sampled at
several sampling points in order to obtain a large variety of different pulps (see
Table 4). One sample per week was taken over the test period of 3 months from
each sampling point. Each sample was analysed in order to obtain the CSF and
fibre length results. The mean values of these analyses of the sampling points are
listed in Table 4.
28
Table 4. Pulp types used in calculation of mass proportions. (Paper II, © 2006 Paperi ja
Puu).
Pulp type Description of the test
point
Freeness [ml]1 Length weighted fibre
length [mm]2
TMP 1 Thermo-mechanical pulp
line 1
47 1.56
TMP 2 Thermo-mechanical pulp
line 2
35 1.53
GW 1 Ground wood line 1 32 0.75
PM 1 Paper machine 1, mixing
chest
54 1.17
PM 2 Paper machine 2, mixing
chest
31 1.09
TMP 1 R11 TMP 1 line 1 Refiner 11 373 1.97
TMP 1 R12 TMP 1 line 1 Refiner 12 100 1.70
TMP 1 R21 TMP 1 line 1 Refiner 21 370 1.58
TMP 1 R22 TMP 1 line 1 Refiner 22 98 1.53
TMP 1 RR1 TMP 1 line 1 Reject
refiner 1
136 1.75
TMP 1 RR2 TMP 1 line 1 Reject
refiner 2
76 1.63
1Mean values of Canadian standard freeness during sampling period, 2Mean values of fibre length results
measured with Kajaani FSA fibre length analyser.
3.6 Materials used in fractional analysis studies
Pulp samples for fractional analysis were collected from a one-loop deinking pilot
plant at the University of Oulu (Paper III) or from laboratory pulper (Paper IV).
Sample preparation procedures are described in the following chapters. The
studies were performed with a batch of exactly the same issues of LWC, SC and
ONP or sort of mixture with these issues (see Table 5).
Natural aged material (LWC 60 weeks, SC 60 weeks and ONP 30 weeks),
was compared to the same kind of batches, which were first aged naturally and
then additionally subjected to accelerated aging according to INGEDE (72h at
60°C, INGEDE method 11). After heat treatment these accelerated aged furnishes
(LWC, SC and ONP) were used. This means in practice that every type of raw
material was compared to each other using natural and thermal aging of naturally
aged material (see Table 5).
29
Table 5. Furnishes used. (Based on Paper III and IV, © 2008 & 2009 Tappi).
Furnish Printing
method
Grammage
[g/m2]
Ash [%] Furnish age
[weeks]
Accelerated
aging
Share of batch
[%]
LWC1 offset heatset 80 40 60 no 100
LWC1
(accelerated
aging)
offset heatset 80 40 60 72h, 60°C 100
SC1 offset heatset 58 35 60 no 100
SC1
(accelerated
aging)
offset heatset 58 35 60 72h, 60°C 100
ONP1 offset coldset 50 18 30 no 100
ONP1
(accelerated
aging)
offset coldset 50 18 30 72h, 60°C 100
ONP2 offset coldset 44 16 20 no 50
SC2 offset heatset 62 39 10 25
LWC2 offset heatset 94 30 10 25 1Paper III used pure raw material furnishes, 2Paper IV used a raw material mixture (ONP-SC-LWC,
proportion 50%-25%-25%)
3.6.1 One-loop pilot process sample preparation
A schematic description of the one-loop pilot process at the University of Oulu is
illustrated in Fig. 3 (Laitinen et al. 2007a). Slushing ability of furnishes were
studied in the pilot drum pulper which has a diameter of 2 m. Pulping time was
set to 30 min (pulping samples were also taken after 10 and 20 min of drum
pulping). The pulping was performed at 16% consistency with alkaline soap
chemistry (see Table 6). After the drum pulper, the pulp was diluted first to a
consistency of 2–3% in a pulper accept tank (5 m³) and then the pulp was
circulated for 15 minutes in a pressure-screening loop to achieve a homogenous
batch of pulp after dilution (see Fig. 3). The screening accepts and rejects were
circulated back to the pulper accept tank. After mixing, samples were taken from
the pulper accept tank and analysed in the laboratory.
30
Fig. 3. A schematic flow chart of the one-loop pilot process. (Laitinen et al. 2007a, ©
2007 Tappi).
3.6.2 Procedure for pre-wetting and laboratory pulping
Three different pre-wetting times (1 hour, 4 hours and 24 hours) were compared
to a reference sample (i.e. no pre-wetting before actual pulping). Two process
concepts are introduced in Fig. 4. During the pre-wetting stage, pulps were kept at
16% consistency. Batches of the raw material were kept inside plastic bags during
pre-wetting in a water bath at 45°C and all pulping chemicals were already added
in the plastic bags. No mixing or any other mechanical forces were used during
pre-wetting. The temperatures in pulping and wet disintegration were also kept
constant at 45°C. After 10 minutes of laboratory pulping at 16% consistency
(Kitchen Aid, planetary mixer, KCM Catering Equipment), the samples were
diluted with deionised water and wet disintegrated at 2% consistency for 60,000
revolutions in order to make all the batches homogeneous.
Drum pulper
Pulper accept tank
Screening
Flotation Cell
Wire press (accept)
Final pulp
Wire press (reject)
Reject
Process water tank
Chemicals
Wire press filtrate
Wire press filtrate
Watertower
Overflow
= valve
= pump
31
Fig. 4. Different process concepts after laboratory pulping. (Paper IV, © 2009 Tappi).
3.6.3 Chemicals used during sample preparation procedures
Chemicals used in pilot pulping (Paper III) and laboratory pulping (Paper IV) are
presented in Table 6. In essence, two different chemical mixtures were used:
alkaline soap chemistry (sodium hydroxide 10 kg/t) and reduced alkaline soap
chemistry (no sodium hydroxide). For these various charges of sodium silicate
(water glass), hydrogen peroxide and commercial fatty acid soap (Serfax MT90)
were applied to batches depending on the research tasks (see Table 6).
Table 6. Pulping conditions of furnishes. (Based on Papers III and IV, © 2008 & 2009
Tappi)
Properties Pilot pulping performances1 Laboratory pulping performances2
Raw material German furnishes (LWC, SC
and ONP)
German furnishes (LWC, SC and
ONP)
Pulping consistency 16% 16%
Wet disintegration - 2%
Pulping temperature 50°C 45°C
Pulping batch 35 kg (air dry) 200 g (air dry)
NaOH 10 [kg/t] 10 [kg/t] or 0 [kg/t]
Sodium silicate (as product) 18.6 [kg/t] 18.6 [kg/t]
Commercial fatty acid soap (Serfax
MT90)
6 [kg/t] 2 [kg/t]
H2O2 10 [kg/t] 4 [kg/t]
Hardness [°dH] adj. CaCl2 20 [°dH] not analysed 1In Paper III pilot drum pulper was used, 2In Paper IV laboratory pulper was used.
Sampling
Reference sample
Sample
LaboratoryPulping
Pre-wetting
Dwell 1h, 4h or 24h
Wet disintegration
SamplingLaboratoryPulping
Wet disintegration
32
33
4 Results
The results of the tube flow fractionation experiments are divided into two
separated modelling sections (Chapters 4.1 and 4.2) and fractional analysis
section (Chapter 4.3). Chapter 4.1 describes modelling the fractionation time
during tube flow fractionation with particles of various dimensions. Chapter 4.2
presents how a tube flow fractionator could be used to determine mass
proportions of separated fractions. Further, Chapter 4.3 presents how the tube
flow fractionation method could be used in the specified analysis of DIP fractions,
including slushing of pulp furnishes, detachment of ink particles from fibre
matrices and size distributions of ink particles among fibre fractions.
4.1 Fractionation time of different particles
A study was made of the effect of different particle dimensions on the
fractionation time (elution time) during tube flow fractionation (Paper I). A
Multilinear Regression (MLR) model was developed and verified to estimate the
fractionation time of pulp, paper and rectangular peat particles of various
dimensions (see Table 3). The validity of the model was then verified with a large
number of different pulp, paper and peat samples.
4.1.1 MLR model and statistical analysis
The fractionation experiments were conducted on pulp, paper and peat particles
on which FiberLab measurements (of pulp fibre dimensions) and microscopic
analyses (of peat particles and pieces of paper) had been performed (see Table 3).
An MLR model created for estimating fractionation times for particles of
different lengths, widths and thicknesses in tube flow fractionation was evaluated
by means of statistical analysis. The MLR model for tube flow fractionation time
as determined by MODDE 7.0 is presented in Equation 3:
).(1049.1)(051708.0
)(001473.0)(011352.0861.197sec),.(26 mlengthmthickness
mwidthmlengthtimepredY
(3)
The significance of each simplified MLR model parameter is presented in Fig. 5.
The coefficient plot displays the regression MLR coefficients with their
confidence intervals. All selected parameters are significant (different from the
noise) as the confidence interval does not cross zero.
34
Fig. 5. Significance of each simplified model parameter. (Paper I, © 2011
BioResources).
The correlation between the predicted fractionation time and the observed
fractionation (R2 = 0.976) time after the data collecting period is presented in Fig.
6. Since the predicted fractionation time versus the observed fractionation time is
linear (see Fig. 6), the MLR model (created using 69 sample points representing
pulp, paper and peat particles of different sizes) is valid and can be used to predict
the fractionation times of various pulp, paper and peat particles.
Fig. 6. Correlation between predicted and observed fractionation times (69 sample
points). (Paper I, © 2011 BioResources).
160
170
180
190
200
160 170 180 190 200
Ob
serv
ed [
s]
Predicted [s]
Time with Experiment Number labels
1
3
4
56
7
8
9
10
11
1213
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
6566
67
6869
R2=0.9755
-10
-5
0
5
10
Len
Wid Th
i
Len
*Len
s
Scaled & Centered Coefficients for Time
35
A normal probability plot for the calculated residuals (differences between the
observed and predicted fractionation times) is presented in Fig. 7. This figure
indicates whether the data include outliers that point to defects in the model.
Since the residuals form a straight line and all the points are between -4 and +4 on
the x-axis (studentized deviation) in the normal probability plot, it can be
concluded that there were no outliers and the model fits well with the measured
data.
Fig. 7. Normal probability plot of the measured residuals. (Paper I, © 2011
BioResources).
The basic characteristics R² (correlation coefficient of the created model), Q²
(predictability of the created model) and the reproducibility of the MLR model
are presented in Table 7. The R² is the percentage of the variation and it denoted a
measure of fit, i.e. how well the given model fits the measured data. The Q² is the
percentage of variation in the predictions given by the model according to cross-
validation, and indicates how well the model will predict new data. A good model
should have both high a R² and Q². Finally, reproducibility is the variation in the
response under the same conditions (pure error), as compared with the total
variation in the response. The maximum value in all of the previously described
characteristics is 1.0.
0.0050.010.02
0.05
0.1
0.2
0.30.40.50.60.7
0.8
0.9
0.95
0.980.99
0.995
-4 -3 -2 -1 0 1 2 3 4
N-P
rob
abil
ity
Deleted Studentized Residuals
Time with Experiment Number labels
61
5657
28475855 1069 4125334826346239402944525419596335 41115684265 36182449216053045321253272031466675043517166681338 179
1437222331
64
2
36
Table 7. Certain characteristics of the current MLR model. (Paper I, © 2011
BioResources).
Characteristic Value
Correlation coefficient of the model, R² 0.976
Predictability of the model, Q² 0.971
Reproducibility 0.982
The effects of different particle dimensions in the studied range of dimensions
(see Table 3) on the fractionation time (middlemost) are presented in Fig. 8. The
graphs are plotted so that one dimension is plotted at a time and the other two
dimensions are kept at the average values for the investigated data (length = 1700
µm, width = 70 µm, and thickness = 40 µm). The graphs clearly show that the
fractionation time decreases as the particle dimensions increase. It was particle
length, however, that had the most significant effect on fractionation time in the
tube flow (see also Fig. 5). In addition, Fig. 8 includes plots of the lower and
upper confidence limits (95% confidence level plotted with dashed lines) of the
predicted effects of the various particle dimensions on a given factor.
Fig. 8. Prediction plots of the dimension variables (length, width and thickness)
including 95% confidence level plotted with dashed lines. (Paper I, © 2011
BioResources).
4.1.2 Verification of the MLR model
After formulation of the MLR model for fractionation time in a tube flow
fractionator (see Eq. 3), this model was used in the subsequent verification phase
(with 112 independent sample points not used in the modelling phase) to
175
180
185
190
195
0 1000 2000 3000 4000
Tim
e [s
]
Length [µm]
175
180
185
190
195
0 1000 2000 3000
Tim
e [s
]
Width [µm]
175
180
185
190
195
10 20 30 40 50 60 70 80 90 100
Tim
e [s
]
Thickness [µm]
Prediction Plot
Length = 1700
Width = 70
Thickness = 40
37
investigate the accuracy of the model in predicting new data points. The
correlation between the predicted and observed fractionation times is shown in
Fig. 9 for the various pulp size fractions, pieces of paper (i.e. round discs and
rectangular pieces) and peat samples (sieved into size categories). As noticed, the
model successfully predicted the fractionation times for all tested materials. In
general, the observed fractionation time decreased as the particle dimensions
increased.
Fig. 9. Correlation between predicted and observed fractionation times during the
verification period (112 independent sample points). (Paper I, © 2011 BioResources).
4.2 Calculation of mass proportions in different fractions
Concerning the second research topic, a traditional Bauer McNett (BMN) analysis
was simulated using the consistency transmitter signals from the tube flow
fractionator (Paper II). Basically two different signals are received from the
optical consistency transmitter of the fractionation device (see Fig. 2). These
signals were summarised, because the depolarisation signal corresponds mainly to
the longest fibres and the absorption signal to the finest material.
After the summation operation, all the material was seen in this new
summation signal. The summation signal was normalised so that its area was
R2 = 0.975
160
162
164
166
168
170
172
174
176
178
180
182
184
186
188
190
192
194
196
198
200
160 162 164 166 168 170 172 174 176 178 180 182 184 186 188 190 192 194 196 198 200
Predicted time [s]
Ob
serv
ed
tim
e [s
]
Peat particles
Softwood kraft
Euca kraft
TMP
GW
DIP
Paper disc
Paper pieces
MLR-model
Lin. (MLR-model)
38
exactly 1 unit. This enabled the calculation of the fractions as a mass percentage
directly from the summation signal area (see Fig. 10). Nevertheless, it should be
noted that the summation signal area does not necessarily represent (without
certain calibration of the signal levels) the direct mass percentage of the fractions
(depending on pulp type). However, the signals are a good indication of whether
any modifications have occurred in pulp suspensions.
Fig. 10. The formation of the summation signal. (Paper II, © 2006 Paperi ja Puu).
4.2.1 Modelling the BMN classifier with optical signals
As mentioned earlier, two different signals, known as depolarisation and
absorption, are received from the optical consistency transmitter at the end of the
fractionation tube. Typically the absorption signal overestimates the fine material
content of pulp due to the light scattering phenomenon and as a result, the
absorption signal has to be multiplied by some coefficient factor in order to obtain
the real fibre fraction distribution. During the modelling period, the real
coefficient factor was determined for the absorption signal. As a result, the
absorption signal coefficient factor should be 0.3 for the tests that used
independent samples from each of the mechanical wood-based raw materials (see
Table 4). Furthermore, the depolarisation signal coefficient was kept at exactly 1
in the mass proportion calculations.
In Fig. 11, the correlation after the subsequent verification period
(independent sample points not used in the modelling period) between the BMN
apparatus and the tube flow fractionator is shown using an absorption signal
coefficient factor of 0.3, which was valid for all fractions. It can be seen that the
normalization
Depolarization signal
Absorption signal
Summation signal
normalization
Depolarization signal
Absorption signal
Summation signal
39
model created for the tube flow fractionator successfully predicted the BMN
results (Paper II).
Fig. 11. The tube flow fractionator mass proportions results compared to BMN
apparatus. (Modified from Paper II, © 2006 Paperi ja Puu).
Repeatability of the tube flow fractionator
It is widely accepted that the inter-laboratory repeatability of different Bauer-
McNett devices is limited (Bos 1966, Levlin 1982, Clark 1985, Gooding & Olson
2001). The repeatability of an individual Bauer-McNett and a tube flow
fractionator was investigated with secondary refiner TMP samples (see Table 4,
CSF around 100 ml). Five parallel test samples for both devices were prepared
and tested using both methods. The results of the repeatability tests are presented
in Table 8. These results consisted of combined errors from sample preparations,
analysis and mass percentage determinations for both of the methods tested. It
was proved that the tube flow fractionator had better repeatability than the Bauer-
McNett procedure used in these tests. Standard deviation and the difference
between maximum and minimum values were much better with the tube flow
fractionator than with the Bauer-McNett apparatus. The results presented in Paper
40
II also indicated that the tube flow fractionator divides pulps into the same
fractions as the Bauer-McNett. This observation is based on fibre length analyses
as well as mass proportions levels.
Table 8. Results of the BMN and tube flow fractionator repeatability tests. (Modified
from Paper II, © 2006 Paperi ja Puu).
Bauer-McNett [%] tube flow fractionator [%]
R28 R48 R200 P200 FR1 FR2 FR3 FR4
Run 1 34.9 22.7 17.3 25.1 37.2 20.4 18.3 24.1
Run 2 38.3 23.1 18.5 20.1 40.0 19.3 17.5 23.2
Run 3 35.8 24.9 18.6 20.7 38.0 20.5 18.5 23.0
Run 4 36.3 25.2 19.3 19.2 38.5 20.1 18.7 22.7
Run 5 35.1 23.0 18.1 23.8 37.4 20.1 18.2 24.3
Standard deviation 1.36 1.17 0.73 2.54 1.12 0.47 0.46 0.70
Max and min value dif. 3.4 2.5 2.0 5.9 2.8 1.2 1.2 1.6
4.3 Fractional analysis of different DIP fractions
The major aim of this chapter is to present how the different DIP raw materials
can be analysed using the tube flow fractionation method among different size
fractions. Of special interest is the slushing of pulp, the detachment and
fragmentation of ink, and how ink is distributed among the fibre length classes.
Tube flow fractionation classifies pulp gently according to particle size, after
which it is possible to gather the separated fractions for microscopy study and
image analysis. Through fractional information it was possible to recognize the
amount of free and attached ink as well as the ink particle size distribution
(Papers III-IV).
4.3.1 Slushing of different raw materials
The amount of flakes (mass percentages of FR1) are described to establish
slushing after a certain pulping time. The materials tested are presented in Table 5.
The results show that deflaking was slower with accelerated aging material during
pulping in LWC and SC furnishes alike (see Fig. 12). Heat treatment (accelerated
aging) had a detrimental effect on slushing in the pulper, but during the
circulation loop treatment the differences evened out between naturally and
accelerated aging batches. Thermal aging clearly had a greater influence on the
SC than on the LWC furnish. After 30 minutes of pulping there were still a lot of
41
flake-like pieces (based on fractionation time), which had not yet disintegrated
into separate fibres and fines in the aged SC furnish.
Fig. 12. The amount of flake contents in LWC (a) and SC (b) pulps during drum pulping
times and after 15 minutes of pressure screening. (Paper III, © 2008 Tappi).
The slushing response of ONP differed greatly compared to that of the LWC and
SC furnishes. The slushing of ONP was similar with naturally and accelerated
aging raw material (see Fig. 13). Almost complete slushing was achieved in
practice after 10 minutes of drum pulping.
Fig. 13. The amount of flakes in ONP pulp in different drum pulping times and after 15
minutes of pressure screening. (Paper III, © 2008 Tappi).
4.3.2 Ink detachment and fragmentation
The detachment of ink particles from fibres and ink fragmentation in fines
fractions are presented in Figs. 14–16. Basically, the RI measures the amount of
ink present in pulp, however, the ink size also affects the RI measurement i.e. RI
is most sensitive to ink sizes below 10 µm (Walmsley and Silveri 1999, Haynes
4.6
1.10.5 0.6
10.3
4.0
2.2
0.5
0
2
4
6
8
10
12
14
16
10min 20min 30min Circulation15min
Flak
e co
nte
nt
[%]
LWC natural aging LWC accelerated aging
a)
3.9 3.5
1.5 1.7
14.1
7.1 7.2
0.8
0
2
4
6
8
10
12
14
16
10min 20min 30min Circulation15min
Flak
e co
nten
t [%
]
SC natural aging SC accelerated aging
b)
1.82.7
1.9
0.1
1.8 1.6 1.50.1
0
2
4
6
8
10
12
14
16
10min 20min 30min Circulation15min
Flak
e co
nte
nt
[%]
ONP natural aging ONP accelerated aging
42
2000). Thus, the smaller the ink size, the higher the measured RI becomes, thus
indicating fragmentation of ink, especially within the same raw material.
The values of the samples taken after 15 minutes of pressure screening are
presented in the figures. In the fractional approach, ink particles in long- and
short-fibre fractions (FR2 and FR3) can be characterised as attached ink particles.
Ink detachment from short fibres was slightly poorer in the case of LWC with
accelerated aging than natural aging (see Fig. 14a). The residual ink value was
lower in the finest fraction in the case of accelerated aging material due to lower
ink detachment from the short fibre fraction. The brightness values (Fig. 14b)
showed a logical correlation with the residual ink values. Due to the greater
amount of attached ink particles in FR3 in the case of accelerated aging material,
the brightness values were slightly lower in the short-fibre fraction.
Fig. 14. Residual ink (a) and brightness (b) of different fractions, pulp pads and
hyperwashed pulp pads in the case of LWC after 15 minutes pressure screening
circulation loop. (Paper III, © 2008 Tappi).
The same phenomenon was also noticed with the SC furnish as in LWC, but the
differences in ink detachment between natural and accelerated aging were even
bigger (see Fig. 15a). It can be seen that the amount of attached ink is high in both
the long- and short-fibre fractions after accelerated aging. This can also be seen in
the really high hyperwashed RI-value (≈ 730 ppm) after heat treatment. Due to
the attached ink particles in the fibre fractions, the brightness differences between
natural and accelerated aging were significant, at over 10 points in both fractions
(see Fig. 15b).
53
241
672
501
2850
440528
593
108
0
200
400
600
800
1000
1200
1400
1600
1800
2000
Long fibres Short fibres Fines Pulp pad HW Pulp pad
RI [
ppm
]
LWC natural aging LWC accelerated aging
Attached ink
Detached ink
a)73
58
50
59
7674
5553
58
72
30
35
40
45
50
55
60
65
70
75
80
Long fibres Short fibres Fines Pulp pad HW Pulp pad
Bri
ghtn
ess
[ISO
-%]
LWC natural aging LWC accelerated aging
b)
43
Fig. 15. Residual ink (a) and brightness (b) of different fractions, pulp pads and
hyperwashed pulp pads in the case of SC after 15 minutes pressure screening
circulation loop. (Paper III, © 2008 Tappi).
The ink detachment with accelerated aging ONP was hindered in a similar way as
noticed with LWC or SC. In the case of ONP, a great difference in the ink
fragmentation was observed and was duly noted (see Fig. 16a). The residual ink
value in the finest fraction (FR4) after accelerated aging was over double that of
the naturally aged material. This obviously leads to a brightness loss in the pulp,
which can be seen in Fig. 16b.
Fig. 16. Residual ink (a) and brightness (b) of different fractions, pulp pads and
hyperwashed pulp pads in the case of ONP after 15 minutes pressure screening
circulation loop. (Paper III, © 2008 Tappi).
4.3.3 Ink particle distributions in different fractions
The number of ink particles in long-fibre fractions according to four different ink
size categories is presented in Figs. 17a and 17b. The illustrations indicate that
pre-wetting before pulping detached the ink particles from all ink size categories.
140
540
980873
217
486
1509
818
986
729
0
200
400
600
800
1000
1200
1400
1600
1800
2000
Long fibres Short fibres Fines Pulp pad HW Pulp pad
RI [
ppm
]
SC natural aging SC accelerated aging
Attached ink
Detached ink
a)
61
5047
49
59
49
39
4947 47
30
35
40
45
50
55
60
65
70
75
80
Long fibres Short fibres Fines Pulp pad HW Pulp pad
Bri
ghtn
ess
[ISO
-%]
SC natural aging SC accelerated aging
b)
81 123
822
615
106228
451
1955
1441
276
0
200
400
600
800
1000
1200
1400
1600
1800
2000
Long fibres Short fibres Fines Pulp pad HW Pulp pad
RI [
ppm
]
ONP natural aging ONP accelerated aging
Attached ink
Detached ink
a)
62 63
47
53
64
5653
38
43
59
30
35
40
45
50
55
60
65
70
75
80
Long fibres Short fibres Fines Pulp pad HW Pulp pad
Bri
ghtn
ess
[ISO
-%]
ONP natural aging ONP accelerated aging
b)
44
With reduced alkaline soap chemistry, the same phenomenon was also noticed,
however, the amount of ink particles was clearly higher than in the case of
alkaline soap chemistry, especially in categories under 30 µm.
Fig. 17. Microscopic analysis of long-fibre fraction with different pre-wetting times
with alkaline (a) and reduced alkaline (b) soap chemistry. (Modified from Paper IV, ©
2009 Tappi).
The number of ink particles in short-fibre fractions in four different ink size
categories is presented in Figs. 18a and 18b. Pre-wetting detached ink particles
from all categories similarly to the case of long-fibre fraction. In the case of the
short-fibre fraction, it was observed that most of the ink particles were smaller
than 30 µm (see Fig. 18).
Fig. 18. Microscopic analysis of short-fibre fraction with different pre-wetting times
with alkaline (a) and reduced alkaline (b) soap chemistry. (Modified from Paper IV, ©
2009 Tappi).
In practice, when investigating ink particles but not other substances, the fines
fraction can be considered as only the free ink or ink particles attached to the
fines due to the small particle size contained in the fines fraction. The largest
number of ink particles in a fines fraction was in the size category between 3–10
µm (see Figs. 19a and 19b.).
Alkaline soap chemistry
0
50
100
150
200
250
300
350
400
450
500
3-10 μm 10-30 μm 30-50 μm over 50 μm
Am
oun
t of
ink
part
icle
s [p
cs]
REF alkaline
Alkaline 1h
Alkaline 4h
Alkaline 24h
a)Reduced alkaline soap chemistry
0
50
100
150
200
250
300
350
400
450
500
3-10 μm 10-30 μm 30-50 μm over 50 μm
Am
oun
t of
ink
part
icle
s [p
cs]
REF red. alkaline
Red. alkaline 1h
Red. alkaline 4h
Red. alkaline 24h
b)
Alkaline soap chemistry
050
100150
200250
300350
400450
500550
600650
700
3-10 μm 10-30 μm 30-50 μm over 50 μm
Am
oun
t of
ink
part
icle
s [p
cs]
REF alkaline
Alkaline 1h
Alkaline 4h
Alkaline 24h
a)Reduced alkaline soap chemistry
0
50100
150200
250300
350
400450
500550
600650
700
3-10 μm 10-30 μm 30-50 μm over 50 μm
Am
oun
t of
ink
part
icle
s [p
cs]
REF red. alkaline
Red. alkaline 1h
Red. alkaline 4h
Red. alkaline 24h
b)
45
Fig. 19. Microscopic analysis of fines fraction with different pre-wetting times with
alkaline (a) and reduced alkaline (b) soap chemistry. (Modified from Paper IV, © 2009
Tappi).
Alkaline soap chemistry
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
3-10 μm 10-30 μm 30-50 μm over 50 μm
Am
oun
t of
ink
part
icle
s [p
cs]
REF alkaline
Alkaline 1h
Alkaline 4h
Alkaline 24h
a)Reduced alkaline soap chemistry
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
3-10 μm 10-30 μm 30-50 μm over 50 μm
Am
oun
t of
ink
part
icle
s [p
cs]
REF red. alkaline
Red. alkaline 1h
Red. alkaline 4h
Red. alkaline 24h
b)
46
47
5 Discussion
This study investigated a new fractional analysis method in order to obtain more
information about different kinds of pulps. The effect of various dimensions of
particles on the fractionation time during tube flow fractionation is described in
Chapter 5.1 (Paper I). In Chapter 5.2, a method to determine the mass proportions
of separated fractions is considered (Paper II). Finally, in Chapter 5.3, a specified
analysis of DIP fractions is considered using a fractional approach (Papers III-IV).
5.1 Fractionation time of different particles in tube flow
fractionation
Although the tube flow fractionation method has been found to fractionate pulp
particles primarily according to their length (Pascal and Silvy 1993, Krogerus et al. 2003), there was still a lack of systematic knowledge regarding the effect of
the other particle dimensions (i.e. width and thickness) on the fractionation time.
The author had noticed that changes in fibre length do not exclusively explain the
variation in fractionation time observed between pulp types (e.g. in kraft pulp
fibres versus mechanical pulp fibres). The aim was to clarify the effect of the
various pulp, paper and peat particle dimensions on the fractionation time by
means of empirical modelling.
A multilinear regression (MLR) model was developed to estimate the
fractionation time. All the selected dimensions (length, width, and thickness) can
be determined independently from each other. The exceptions were image
analyses, where the thickness was assumed to be the same as the width due to the
fact that the pictures were only two-dimensional. The model developed for the
predicted fractionation time gave a linear correlation with the observed
fractionation time (see Fig. 6).
Fig. 7 shows that all the data points measured lie either on or very close to a
straight line between -4 and +4 along the x-axis (studentized deviation). This
proves that the residuals were almost completely evenly distributed and no
systematic error existed in the data. Furthermore, all the characteristic values (see
Table 7) were very high (close to 1), which indicated that the model was valid for
this fractionation analysis.
The model developed can be used to estimate the effect of pulp, paper and
peat particle dimensions on the fractionation time (see Fig. 8). It was ensured that
the particle length had the most significant effect on fractionation time in a tube
48
flow environment within the range of dimensions examined here (see Table 3, Fig.
5 and Fig. 8), but the data also suggested that the plot of particle length versus
time is not linear over the entire length range (Fig. 8). It should be noted that in
practice no pulp fibres in the over 3500 µm category were analysed. All the
particles in this size category were pieces of paper and their shape was not
elongated (like the fibres) but roundish. Overall, the fractionation time became
shorter as the mean dimensions of the particles increased.
The MLR model was verified with pulp, paper and peat particles of different
types and a wide size range (see Fig. 9). The model proved to be valid for all the
particles clear of the tube wall. Special attention should be paid to the range of
dimensions (see Table 3), since the model is not necessarily valid outside this
range.
The model formulated here can be used to calculate the specific pulp, paper
and peat particle fractions in a sample beforehand, given that the essential
dimensions of the particles (length, width and thickness) are known. Thus, it can
be utilised, for example in situations where the amount of flake-like pieces in a
pulp sample must be known, since these will exit the tube faster than smaller
individual long fibres. This makes it possible to calculate the flake content as a
mass percentage after fractionation (Paper I).
5.2 Calculation of mass proportions of different fractions
Screen fractionation methods have conventionally been used to perform
laboratory fractionation. The major weaknesses of these methods are the
laborious nature of screen fractionation, the poor repeatability between individual
devices and the impossibility of using the method in online process control (Bos
1966, Levlin 1982, Clark 1985, Gooding & Olson 2001). The aim of this part of
research was to investigate whether the tube flow fractionator could replace
traditional screen fractionation such as the BMN apparatus. If fractionation could
be done quickly and the results between devices were comparable, the
fractionation of fibres into classes and further analysis with other methods would
offer a practical method to understand final pulp properties. Fig. 10 presents a
method with which it is possible to determine the fractions as a mass percentage
directly from the optical signal area.
It is well known that various fibre size categories of pulp furnish have very
different effects on final paper properties. In the case of virgin fibre-based pulps,
increasing the amount of long fibres usually improves the strength properties of
49
the final paper. On the other hand, increasing the amount of fine particles
improves the formation of the final paper.
In the case of recycled fibre-based pulps, some ink particles are still attached
to the fibre surfaces after flotation or the washing stage. The amount of this ink is
commonly called the attached ink content. Typically, the amount of attached ink
is very difficult to measure without classifying the sample in some way. On the
other hand, detached ink is typically found in the finest fraction of pulps and is
called free ink.
A practical model for the tube flow fractionator to estimate the BMN results
was found in the tests. Correlations between the tube flow fractionator and the
BMN apparatus were excellent in the different fractions (see Fig. 11). Only the
correlation of the second fraction was quite poor. All pulps contained an equal
percentage of the Bauer McNett R48 fraction, which complicated the modelling.
Usually, if the long-fibre content decreases, a mass percentage increase will be
seen in the fine fractions but not necessarily in the Bauer McNett R48 fraction.
Based on these tests, it can be stated that the tube flow fractionator can
replace the Bauer-McNett device in mass proportion determination. The tube flow
fractionator can divide pulps into the same kind of fractions as the Bauer-McNett.
The results showed that the tube flow fractionator also has far better repeatability
than the Bauer-McNett apparatus (see Table 8). Standard deviation and the
difference between maximum and minimum values were better with the tube flow
fractionator than with the Bauer-McNett apparatus (Paper II).
5.3 Fractional analysis of different DIP fractions
This chapter focuses on a comparison of different recycled raw material furnishes
using fractional analysis; especially in respect of the slushing rate of furnishes,
attached and detached ink content of the deinked pulp and ink particle size
distributions.
5.3.1 Effect of aging on DIP furnishes
The age of recovered paper can vary considerably. It has been widely observed
that the deinkability of aged paper is generally lower than that of fresh paper
(Paper III, Dorris & Pagé 2000a, Carre et al. 2000, Ben & Dorris 2006). The
aging of raw material has negative effects in particular on the behaviour of paper
components (especially coatings and inks) in pulping and flotation (Rao & Kuys
50
1995, Rosinski 1995, Dorris & Pagé 2000b, Le Ny et al. 2004). The consequence
of aging is a reduction in final pulp brightness, due to the decreased detachment
of ink and in some cases the increased fragmentation of ink particles, which
means that ink particles are able to be broken down into very fine particles during
processing. The decreased detachment of ink is due to the oxidation of the ink,
which strongly bonds the ink to fibre material (Castro et al. 1996, Haynes 2000,
Castro et al. 2002). The oxidation of the ink increases with prolonged storage
time, especially at elevated temperatures. This is one reason why in the mill
environment, the aging or thermal aging phenomenon is usually noticed during
the summer period and is known informally as the summer effect (Haynes 1999,
Haynes 2008).
In addition to ink attachment, it has also been shown that aging and thermal
aging lead to increased ink fragmentation (Paper III, Hestner & Carre 2000,
Haynes & Hestner 2001, Merza & Haynes 2001). The detrimental influence of
strong ink fragmentation on deinking arises from the fact that it induces
redeposition of fragmented ink particles onto fibre surfaces and into the fibres
themselves (Galland et al. 2001).
Difficulties in the deinking process have been highlighted related to ink
formulation and type of paper (Carre et al. 2000, Ben & Dorris 2006).
Furthermore, offset heatset printed papers (coated and uncoated) reduce the
brightness of deinked pulp and increase speck contamination, especially with
aged prints. For uncoated papers, aging leads to poor ink detachment from fibres
and high speck contamination, especially if the papers have low filler content.
The thickness of the ink film on LWC paper has a strong impact on deinkability,
with flotation efficiency increasing with increasing thickness, but in contrast, in
the case of uncoated papers, ink film has a minimal impact according to Carre et al. (2000).
5.3.2 Slushing of furnishes with a pilot-scale pulper
Paper disintegrates when the forces in a pulper exceed the wet strength of the
paper. At the beginning of the pulping process, water penetrates the fibre network
and further into the fibre cell wall. When the fibres swell and detach from each
other, the structure of the paper weakens and the paper then disintegrates.
Depending on the intensity and duration of the mechanical action and also on the
wet strength of the paper, the paper flake content will vary. Nevertheless raw
51
material aging will change inks, and coatings will modify the rate of water
penetration and the wet strength of the paper.
The slushing ability of different raw materials was investigated with a pilot-
scale pulper and samples were analysed using the tube flow fractionation method.
Thermally aged LWC and SC were slower to slush (see Fig. 12). However, aging
does not affect ONP slushing ability (see Fig. 13). It is likely that pulping
becomes difficult when using LWC because the coating itself changes so that it is
more difficult to disintegrate. Typically, SC furnishes have a larger ink cover than
ONP furnishes and, in addition, SC and ONP are not printed with the same inks.
The SC grade used in these tests was printed with offset heatset ink which
contained unsaturated components. These components typically oxidise with time
and at an elevated temperature. After thermal aging, the ink in an SC furnishes
sticks the fibres together, hence these elements are much harder to disintegrate
than in a fresh SC furnish. In the case of ONP, the effect of oxidation on slushing
was much lower due to the use of coldset printing ink. In addition, it has to be
noted that the ONP (30 weeks old) used in these trials was less aged than the
LWC and SC (60 weeks old). It is widely believed that 6 months is a sufficient
time for the oxidation of ink. Changes do not happen at the same rate after this
point with natural aging.
5.3.3 Ink detachment and fragmentation of inks
Fractional analysis showed the long-fibre fraction to be the cleanest fraction in all
the researched furnishes, with respect to attached ink particles in fibres (see Fig.
14–16). The amount of ink particles is clearly higher in the short-fibre fraction,
while almost all the detached ink particles from fibres are in the fines fraction
(Paper III).
On thermal aging, the ink became more difficult to remove from furnishes
due to an increasing ink fibre/coating attraction. There were more attached ink
particles in fibre fractions in all the furnishes used (LWC, SC and ONP) with
thermally aged material compared to naturally aged material. However, heat
treatment had even bigger influence on SC than on LWC. The most probable
reason is that SC is not coated so there are more ink fibre binding possibilities
than in the case of LWC. Poorer ink detachment in the case of LWC and SC aged
raw material could lead to the formation of large ink particles and dirt specks.
Nevertheless, these large ink particles are not so harmful to the final brightness of
pulp as the small ones obviously are.
52
Fragmentation of ink particles is usually understood to give a negative effect
on the removal of ink particles in the flotation stage. In the case of thermally aged
ONP, the ink particle fragmentation was noticed to be a more serious problem
than ink particle detachment (see Fig 16a). The detached ink particles (i.e. free
ink particles) are carried out in the finest fraction. Due to the large amount of
small ink particles in the case of the ONP furnish (based on high residual ink
values in the finest fraction) there will be difficulties in ink removal by flotation.
It is impossible to reach the same brightness level in the bleaching stage and
therefore the final pulp brightness will be much lower than a fresh ONP furnish.
Furthermore, due to the tendency of ink particles to fragment in aged ONP, it is
suggested that a shorter pulping time be used to the extent that is practical (Paper
III).
5.3.4 Ink particle size distributions in different fractions
The potential of improving ink detachment with pre-wetting without any
mechanical forces was investigated. Rao and Stenius (1998) have suggested that
the detachment of ink is a two-step process: ink softening is followed by ink
release. The softening of ink particles affects the ink/fibre interaction and this is
the most probable reason why the pre-wetting of pulp was so effective in this
research (Paper IV). Further research results with pure raw materials (ONP, SC
and LWC) indicated that in the case of ONP and SC ink, release could be
improved considerably by pre-wetting, but not as much in the case of LWC
(Kemppainen et al. 2010, Kemppainen et al. 2011). The amount of ink particles in
different size categories is presented in Fig. 17–19. In addition, the results of
fractional analysis suggest that the same amount of mechanical forces in the
actual pulping process resulted in better ink detachment from fibres with pre-
wetting. Ink detachment levels were already notably better after one hour of pre-
wetting, which suggests the possibility of successful implementation in an
industrial environment. This also means that the actual pulping process could be
shorter and the energy consumed in dispersion lower if pulp were pre-wetted for a
sufficient time before actual pulping, due to the fact that after pre-wetting ink is
detached more easily.
When increasing the ink detachment from fibres, excessive ink fragmentation
should be avoided, because small ink particles are difficult to remove in the
flotation stage (Holik 2000, Lassus 2000, Fabry 2000, Fabry et al. 2001). Strong
ink fragmentation may also lead to the redeposition of fragmented ink particles
53
onto fibre surfaces and into the fibre lumens. Ben et al. (Ben et al. 2000a, Ben &
Dorris 2000b) have stated that ink particles smaller than 5 µm can access the
lumen because they are smaller in size than the lumen pits.
In the long- and short-fibre fractions (see Figs. 17–18), it can be seen that the
use of pre-wetting clearly detached the ink particles from all size categories with
both alkaline and reduced alkaline chemistries. In the finest fraction, it can be
observed that when the ink particles detached from the fibres, they were only
slightly fragmented into smaller ink particles if no mechanical forces were used
(see Fig. 19). This proves that the dwell before pulping is not harmful in respect
to ink fragmentation (see Fig. 19). If pre-wetting before pulping was a problem, a
greater number of small ink particles in the finest particle size category (smaller
than 10 µm) would be seen.
The results of fractional analysis show that as a consequence of using a pre-
wetting stage, the ink content of the long-fibre fraction was notably lower. Using
a fractionation stage after a pre-wetting and a pulping stage would enable the long
fibres to bypass the first flotation stage and be diverted directly to the dispersing
stage of the deinking process (Paper IV).
54
55
6 Conclusions and proposed applications for future research
This thesis provides new possibilities for the fractional approach in the analysis of
pulp and paper samples. The aim of the thesis was to obtain new information
about the fundamental phenomena governing the fractionation time of particles
with various dimensions in a tube flow fractionator. According to the study done
in Paper I, particle length has the most significant effect on fractionation time,
while particle width and thickness have a smaller, but still statistically significant
effect, at a 95% confidence level.
Another aim of the thesis was also to find out whether the tube flow
fractionator could be utilised in the mass percentage determination of fibre
fractions of pulp samples. On the basis of the study presented in Paper II, the tube
flow fractionator can be used to replace the Bauer-McNett device to determine the
mass percentages of pulp sample fractions. Furthermore, the tube flow
fractionator divides pulps into the same kind of fractions as the Bauer-McNett
apparatus.
The third aim was to find out whether the tube flow fractionator could be
utilised in the fractional analysis of different DIP raw materials and DIP process
solutions. According to the studies presented in Papers III and IV, a comparison of
different DIP raw materials and process solutions can be performed with the
fractional analysis presented in this thesis. It was noticed that thermal aging has a
negative effect on the slushing rate of LWC and SC raw material, but not on ONP.
The ink detachment of LWC and SC is poorer when furnishes aged. However, the
aging of LWC and SC did not result in significant ink fragmentation, but only
minor fragmentation. In contrast, strong ink fragmentation is typically noticed
with aged ONP furnishes.
Ink particle fragmentation is usually understood to have a negative effect on
the removal of ink particles in the flotation stage. Different process solutions,
where excessive ink particle fragmentation is prevented, would be useful for the
better functioning of flotation. One proposal for less fragmenting process
solutions could be preferential wetting before the actual pulping stage or a less
intensive pulping procedure. However, the solutions mentioned above may need
an additional fractionation stage that enables the cleaner long fibres to bypass the
first flotation stage and be diverted directly to the dispersing stage in order to
achieve sufficient quality of the final pulp and at the same time avoid unnecessary
costs. These proposed concepts could not be verified in an actual mill
56
environment in the context of this thesis, but many of the existing DIP processes
could be modified so that it is possible to test and ensure the applicability of the
proposed process concepts on mill scale. Nevertheless, it can be concluded that
ink detachment from slightly aged DIP furnishes was improved by pre-wetting the
raw materials using conventional chemistry without the additional fragmentation
of ink particles. Furthermore, ink particles are also released as larger pieces when
using a pre-wetting stage.
It is well known that different recycled fibre-based raw materials have very
different deinkability potentials. The fractional method introduced in this research
gives new insight into investigating the furnishes of different raw materials and
their deinkability potential. Typically, many traditional laboratory analyses are
laborious and slow to perform and thus new reliable measurement solutions are
needed. One feasible process environment is obviously the deinking process,
where the tube flow fractionation method would provide a lot of useful
information when developing more environmental friendly and better quality
processes. The fractional method described in this thesis enables the
determination of component removal (including various sizes of fibres, fines and
ink particles) in different process stages. Future work should focus on the
development of more progressive image processing and analysis tools for the tube
flow fractionation device. This means in practice that essential fibre properties,
the amount of ink particles and particle size distributions, could be determined
directly online from pulp suspensions. This new analytical device would enable
enhanced control of different DIP process stages and would contribute a more
efficient process with better waste management and improved final product
quality.
57
References
Ben Y, Dorris GM & Dagenais M (2000a) Irreversible ink redeposition during repulping. Part 1: Model deinking systems. J Pulp Pap Sci 26(3): 83–89.
Ben Y & Dorris GM (2000b) Irreversible ink redeposition during repulping. Part 2: ONP/OMG furnishes. J Pulp Pap Sci 26(8): 289–293.
Ben Y & Dorris G (2006) Deinkability of uncoated groundwood papers. Part 1. Supercalendered grades and soft-nip calendered vs. LWC and ONP. Prog Pap Recycling 15(2): 5–10.
Bos J (1966) A strange experience with the Bauer-McNett classifier. Tappi Journal 49(11): 508.
Carre B, Magnin L, Galland G & Vernac Y (2000) Deinking difficulties related to ink formulation; printing process and type of paper (Extended Abstract). Tappi J 83(6): 60.
Castro C, Daneault C & Dorris GM (1996) Analysis of the accelerated thermal aging of oil ink vehicles using isothermal thermogravimetry. J Pulp Pap Sci 22(10): J365-J371.
Castro C, Dorris GM & Daneault C (2002) Monitoring and characterization of ink vehicle autoxidation by inverse gas chromatography. Journal of Chromatography A 969(2002): 313–322.
Clark J (1985) Pulp technology and treatment of paper. Miller Freeman Publications Inc., USA 1985.
Dorris GM (1999) Ink detachment and flotation efficiency in ONP/OMG blends. Part I: effect of specimen preparation procedure on estimates. J Pulp Pap Sci 25(1): 1–9.
Dorris GM & Pagé M (2000a) Natural and accelerated aging of old newspapers printed with black mineral oil inks and colored vegetable oil inks. Part 1: Deinkability. Prog Pap Recycling 9(4): 14–26.
Dorris GM & Pagé M (2000b) Natural and accelerated aging of old newspapers printed with black mineral oil inks and colored vegetable oil inks. Part 2: Flotation losses. Prog Pap Recycling 9(4): 27–36.
Fabry B (2000) Pulping, a key factor for optimising deinking. Proc. 53rd ATIP annual meeting, Bordeaux, France, 17 – 19 October 2000.
Fabry B, Carré B & Cremon P (2001) Pulping optimisation: effect of pulping parameters on defibering, ink detachment and ink removal. Proc. 6th Research Forum on Recycling, Tappi 2001, Magog, Canada, 1 – 4 October 2001: 37–44.
Galland G, Vernac Y, Carré B & Rousset X (2001) Effect of pulping conditions on ink redeposition and ink removal when recycling waterbased ink printed papers. Proc. TAPPI Pulping Conference, Tappi 2001, Seattle, USA, 4 – 7 November: 587–628.
Gooding RW & Olson JA (2001) Fractionation in a Bauer-McNett Classifier. J Pulp Pap Sci 27(12): 423–427.
Haynes RD (1999) Measuring the summer effect in North American newsprint deinking mills. Proc. 5th Research Forum on Recycling, Tappi 1999, Ottawa, Canada, 28 – 30 September 1999: 25–35.
Haynes RD (2000) The impact of summer effect on ink detachment and removal. Tappi J 83(3): 56–65.
58
Haynes RD (2000) Measuring ink content from pulper to deinked pulp. Proc. The Recycling Symposium, Washington, DC, USA, 5 – 8 March 2000.
Haynes RD & Hestner A (2001) Benchmark of Deinking Performance for Europe and North America. Prog Pap Recycling 10(1): 35–42.
Haynes RD (2008) A decade of deciphering the "summer effect". Pap Age 124(3): 30–33. Hemström G, Moller K & Norman B (1976) Boundary layer studies in pulp suspensions
flow. TAPPI 59(8): 115–118. Hestner A & Carré B (2000) Benchmark of deinking performance in Europe. Proc. 9th
PTS-CTP Deinking symposium, Munich, Germany, 8 – 11 May 2000: Paper 7. Holik H (2000) Unit Operations and Equipments in Recycled Fiber Processing. Göttsching,
L. & Pakarinen, H. (eds.): Papermaking Science and Technology. Book 7. Recycled Fiber and Deinking, Fapet Ltd. p. 91–134.
Johansson B & Kubat J (1956) An axial separation effect in flowing pulp suspensions. Svensk Papperstidning 59(23): 845–846.
Johansson H, Olgård G and Jernqvist Å (1970) Radial particle migration in plug flow − a method for solid-liquid separation and fractionation, Chem. Eng. Sci. 25(3): 365–372.
Jäsberg A, Raiskinmäki P & Kataja M (2008) Rheology and flow behaviour of fibre suspensions In: Kataja M (ed.) Rheological materials in process industry. ReoMaT Final Report. Helsinki, Edita Prima Oy: 55–83.
Kemppainen K, Laitinen O, Körkkö M, Illikainen M & Niinimäki J (2010) Analysis of recycled papers pre-wetting conditions prior to pulping in deinking – ONP furnish. Prog Pap Recycling 19(3): 3–16.
Kemppainen K, Laitinen O, Körkkö M, Illikainen M & Niinimäki J (2011) Analysis of recycled papers pre-wetting conditions prior to pulping in deinking – SC and LWC furnishes. Manuscript accepted to Tappi J.
Kerekes R & Schell C (1992) Characterization of fibre flocculation regimes by a crowding-factor. J Pulp Pap Sci 18(1): J32-J38.
Krogerus B Fagerholm K & Loytynoja L (2003) Analytical fractionation of pulps by tube flow. Pap Puu 85(4): 209–213.
Körkkö M, Haapala A, Liimatainen H, Ämmälä A and Niinimäki J (2011) Challenges in residual ink measurement – Effect of fibre fines and fillers. Appita J. 64(1): 71–75.
Laitinen O (2002) Determination of mechanical pulp fibre and fines fractions with a fractionator. MSc thesis, Department of Process and Environmental Engineering, University of Oulu (Finnish).
Laitinen O, Körkkö M, Vahlroos S & Niinimäki J (2006) Ink release and particle size analysis with combined use of a tube flow fractionator and a microscope. Proc. 12th PTS-CTP Deinking symposium, Leipzig, Germany, 25 – 27 April 2006: Paper 24.
Laitinen O, Körkkö M, Ämmälä A, Saren M & Niinimäki J (2007a) Analysis of pulp component in a DIP process with tube flow fractionation. Proc. 8th Research Forum on Recycling, Tappi 2007, Niagara Falls, Canada, 23 – 26 September 2007: 272–281.
Laitinen O, Löytynoja L, Kumpulainen H and Niinimäki J (2007b) Method and measuring device for measuring recycled fibre pulp, International Patent, Publication number: WO 2007/122289.
59
Lassus A (2000) Deinking Chemistry. Göttsching, L. & Pakarinen, H. (eds.): Papermaking Science and Technology. Book 7. Recycled Fiber and Deinking, Fapet Ltd.: 240–265.
Le Ny C, Haveri M & Pakarinen H (2004) Impact of RCF raw materials quality and storage conditions on deinking performance. Pap.Technol. (Bury, U.K.) 45(7): 25–32.
Leberle U (2008) Paper Recycling – A reality to grow further. Proc. Final Conference of COST Action E46, Bordeaux, France, 22 – 23 October 2008.
Lee PFW & Duffy GG (1976) An analysis of the drag reducing regime of pulp suspensions flow. TAPPI 59(8): 119–122.
Levlin JE (1982) On the suitability of the McNett classifier for the fibre-length classification of pulps. Paperi ja Puu Erikoisnumero 4 (1982): 213–236.
Lin Y-C and Jen C-P (2002) Mechanism of hydrodynamic separation of biological objects in microchannel devices. Lab Chip. 2(2002): 164–169.
Liukkonen J (2006) Determination of paper furnish flocculation with a fractionator. MSc thesis, Chemical Technology, Lappeenranta University of Technology (Finnish).
Mason SG (1950) The motion of fibres in flowing liquids. Pulp and paper Magazines of Canada, April 1950, 93–100.
Merza J & Haynes RD (2001) Batch pulping versus drum pulping: The impact on deinking performance. Proc. 6th Research Forum on Recycling, Tappi 2001, Magog, Canada, 1 – 4 October 2001: 45–51.
Niskanen T (2010) On automatic ink measurement in deinking plant. MSc thesis, Department of Process and Environmental Engineering, University of Oulu.
Olgård G (1970) Fractionation of fiber suspensions by liquid column flow. Tappi J 53(7): 1240–1246.
Pascal R and Silvy J (1991) La Caractérisation en Continu des Pates à Papier. Revue A.T.I.P. 45(7), Décembre 1991, 253–261 (French).
Pascal R and Silvy J (1993) Hydrodynamic fractionation applied to characterization of paper pulp. Récent progrès en génie des procédés 7, no 29, 233–238 (French).
Provder T (1997) Challenges in particle size distribution measurement past, present and for the 21st century. Prog. Organic Coatings 32(1997): 143–153.
Rao R & Kuys K (1995) Deinkability of Aged Paper. Proc. 49th Appita Annual Conference, Hobart, Australia, p. 601–608.
Rao R & Stenius P (1998) Mechanisms of ink release from model surfaces and fiber, J Pulp Pap Sci 24(6): 183–187.
Rosinski J (1995) Aging and Deinking of Soy Printed Newsprint. Prog Pap Recycling 4(3): 55–62.
Silvy J and Pascal R (1968) L’observation en continu des suspensions fibreuses: Methode pratique de controlle de fabrication. Revue A.T.I.P. 23(1968), 205–214 (French).
Silvy J and Pascal R (1992) Device for determining the characteristics of particles in suspensions in a liquid. US patent 5.087.823.
Schimpf M, Caldwell K and Giddings J C (2000) Field flow fractionation handbook. A Wiley-Interscience publication, USA 2000.
60
Tiikkaja E (2007) Application of an optical fibre analyser and a tube flow fractionator to the estimation of quality potential of TMP - Experimental study. Acta Univ Oul C 285 (Finnish).
Tijssen R, Bos J and Kreveld M. (1986) Hydrodynamic chromatography of macromolecules in open microcapillary tubes. Anal. Chem. 58(14): 3036–3044.
Walmsley M and Silveri L (1999) Relationship among image analysis measurements, dirt count, brightness and ERIC, In: Doshi MR and Dyer JM (eds.), Paper recycling challenge. Volume IV. Process control and mensuration. Appleton, WI, USA: Doshi and Associates: 45–56.
61
Original publications
I Laitinen O, Kemppainen K, Stoor T & Niinimäki J (2010) Fractionation of pulp and paper particles selectively by size. BioResources 6(1): 672–685.
II Laitinen O, Löytynoja L & Niinimäki J (2006) Tube flow fractionator − A simple method for laboratory fractionation. Paperi ja Puu - Paper and Timber. 88(6): 351–355.
III Laitinen O, Körkkö M & Niinimäki J (2008) The effects of aging in different raw material furnishes. Progress in Paper Recycling 18(1): 9–15.
IV Laitinen O, Kemppainen K, Körkkö M, Stoor T & Niinimäki J (2009) The effect of pre-wetting of recycled raw material before pulping in a deinking process. Progress in Paper Recycling 18(4): 24–32.
Reprinted with permission from BioResources (I), Paper and Timber (II) and
Technical Association of the Pulp and Paper Industry (III-IV).
Original publications are not included in the electronic version of the dissertation.
62
A C T A U N I V E R S I T A T I S O U L U E N S I S
Book orders:Granum: Virtual book storehttp://granum.uta.fi/granum/
S E R I E S C T E C H N I C A
366. Palukuru, Vamsi Krishna (2010) Electrically tunable microwave devices using BST-LTCC thick films
367. Saarenpää, Ensio (2010) Rakentamisen hyvä laatu : rakentamisen hyvän laaduntoteutuminen Suomen rakentamismääräyksissä
368. Vartiainen, Johanna (2010) Concentrated signal extraction using consecutivemean excision algorithms
369. Nousiainen, Olli (2010) Characterization of second-level lead-free BGAinterconnections in thermomechanically loaded LTCC/PWB assemblies
370. Taskila, Sanna (2010) Improved enrichment cultivation of selected food-contaminating bacteria
371. Haapala, Antti (2010) Paper machine white water treatment in channel flow :integration of passive deaeration and selective flotation
372. Plekh, Maxim (2010) Ferroelectric performance for nanometer scaled devices
373. Lee, Young-Dong (2010) Wireless vital signs monitoring system for ubiquitoushealthcare with practical tests and reliability analysis
374. Sillanpää, Ilkka (2010) Supply chain performance measurement in themanufacturing industry : a single case study research to develop a supply chainperformance measurement framework
375. Marttila, Hannu (2010) Managing erosion, sediment transport and water quality indrained peatland catchments
376. Honkanen, Seppo (2011) Tekniikan ylioppilaiden valmistumiseen johtavienopintopolkujen mallintaminen — perusteena lukiossa ja opiskelun alkuvaiheessasaavutettu opintomenestys
377. Malinen, Ilkka (2010) Improving the robustness with modified boundedhomotopies and problem-tailored solving procedures
378. Yang, Dayou (2011) Optimisation of product change process and demand-supplychain in high tech environment
379. Kalliokoski, Juha (2011) Models of filtration curve as a part of pulp drainageanalyzers
380. Myllylä, Markus (2011) Detection algorithms and architectures for wireless spatialmultiplexing in MIMO-OFDM systems
381. Muhos, Matti (2011) Early stages of technology intensive companies
C382etukansi.kesken.fm Page 2 Thursday, May 19, 2011 1:40 PM
ABCDEFG
UNIVERS ITY OF OULU P.O.B . 7500 F I -90014 UNIVERS ITY OF OULU F INLAND
A C T A U N I V E R S I T A T I S O U L U E N S I S
S E R I E S E D I T O R S
SCIENTIAE RERUM NATURALIUM
HUMANIORA
TECHNICA
MEDICA
SCIENTIAE RERUM SOCIALIUM
SCRIPTA ACADEMICA
OECONOMICA
EDITOR IN CHIEF
PUBLICATIONS EDITOR
Senior Assistant Jorma Arhippainen
Lecturer Santeri Palviainen
Professor Hannu Heusala
Professor Olli Vuolteenaho
Senior Researcher Eila Estola
Director Sinikka Eskelinen
Professor Jari Juga
Professor Olli Vuolteenaho
Publications Editor Kirsti Nurkkala
ISBN 978-951-42-9448-8 (Paperback)ISBN 978-951-42-9449-5 (PDF)ISSN 0355-3213 (Print)ISSN 1796-2226 (Online)
U N I V E R S I TAT I S O U L U E N S I SACTAC
TECHNICA
U N I V E R S I TAT I S O U L U E N S I SACTAC
TECHNICA
OULU 2011
C 382
Ossi Laitinen
UTILISATION OF TUBE FLOW FRACTIONATION IN FIBRE AND PARTICLE ANALYSIS
UNIVERSITY OF OULU,FACULTY OF TECHNOLOGY,DEPARTMENT OF PROCESS AND ENVIRONMENTAL ENGINEERING
C 382
ACTA
Ossi Laitinen
C382etukansi.kesken.fm Page 1 Thursday, May 19, 2011 1:40 PM