Utilisation of tube flow fractionation in fibre and ...

$: Utilisation of tube flow fractionation in fibre and ...$
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


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


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


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 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


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


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

]


Attached ink

Detached ink

a)73

58

50

59

7674

5553

58

72

30

35

40

45

50

55

60

65

70

75

80


Bri

ghtn

ess

[ISO

-%]


b)

43


hyperwashed pulp pads in the case of SC after 15 minutes pressure screening


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.


hyperwashed pulp pads in the case of ONP after 15 minutes pressure screening


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


RI [

ppm

]


Attached ink

Detached ink

a)

61

5047

49

59

49

39

4947 47

30

35

40

45

50

55

60

65

70

75

80


Bri

ghtn

ess

[ISO

-%]


b)

81 123

822

615

106228

451

1955

1441

276

0

200

400

600

800

1000

1200

1400

1600

1800

2000


RI [

ppm

]


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


Bri

ghtn

ess

[ISO

-%]


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)


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


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).


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


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


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


ABCDEFG

UNIVERS ITY OF OULU P.O.B . 7500 F I -90014 UNIVERS ITY OF OULU F INLAND


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


Senior Researcher Eila Estola

Director Sinikka Eskelinen

Professor Jari Juga


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)


TECHNICA


TECHNICA

OULU 2011

C 382

Ossi Laitinen


UNIVERSITY OF OULU,FACULTY OF TECHNOLOGY,DEPARTMENT OF PROCESS AND ENVIRONMENTAL ENGINEERING

C 382

ACTA

Ossi Laitinen


Utilisation of tube flow fractionation in fibre and ...

Documents

Transcript of Utilisation of tube flow fractionation in fibre and ...