This paper is published in the open archive of Mid Sweden University

DIVA http://miun.diva-portal.org
with permission of the publisher

Citation for the peer-reviewed published paper:

Karlsson A, Agnemo R. High consistency hydrogen peroxide bleaching of mechanical pulps

with varying amounts of fines. Nordic Pulp & Paper Research Journal. 2010;25(3):256-268.

URL to article at publishers site:

http://dx.doi.org/10.3183/NPPRJ-2010-25-03-p256-268
 Mechanical Pulping

High consistency hydrogen peroxide bleaching of

mechanical pulps with varying amounts of fines
Anette Karlsson and Roland Agnemo
widely used bleaching chemical that is particularly
KEYWORDS: Mechanical pulps, Brightness, Light useful when high brightness levels are required
scattering, Light absorption, Fines, Bleaching, Hydrogen however there is a limit to the brightness gain that
peroxide, Temperature can be achieved by increasing the amount of
hydrogen peroxide applied to a pulp (Presley, Hill
SUMMARY: Hydrogen peroxide is a widely used
bleaching chemical for mechanical pulps and is 1996). One limiting factor for the bleaching
particularly useful when high brightness levels are efficiency could be the mechanical pulp fines.
required. The objective of this work was to study fines as a Fines have traditionally been defined as the
limiting factor for reaching higher brightness levels in fraction of a pulp that passes through a 200-mesh
high consistency hydrogen peroxide bleaching of Norway wire (see e.g. Luukko 1998). All types of fines
spruce (Picea abies) thermomechanical (TMP) and stone contain more lignin, hemicelluloses and pectins, and
groundwood (SGW) pulps. The hydrogen peroxide less cellulose than long fibres (Kleen et al. 2003;
bleaching efficiency (i.e. light absorption coefficient Sundberg et al. 2003; Kangas, Kleen 2004, Haugan,
reduction) was studied as a function of temperature, alkali Gregersen 2006). Furthermore, the chemical
charge and fines content using an experimental design
structure of lignin has been shown to vary
based on MODDE software (Umetrics AB).
It is known that all types of fines contain more lignin, depending on its morphological origin in wood
hemicelluloses, pectins, metals and less cellulose than long (Boutelje, Eriksson 1984; Sorvari et al. 1986),
fibres. The light absorption coefficient was greater for which might imply that fibre and fines could behave
unbleached TMP than for unbleached SGW pulp and an differently when subjected to bleaching. Mechanical
increased quantity of fines increased the light absorption pulp fines have been found to lower the hydrogen
coefficient for both pulp types. The increase was, peroxide bleaching efficiency and metals such as
however, most pronounced for the TMP. Furthermore, the iron, manganese and copper were accumulated in
data suggests that both the fibre fraction and the fines the fines fraction (Roick et al. 1991; Leduc,
fraction are more coloured in the thermomechanical pulp. Daneault 2007). Metal ions, especially transitions
In most cases, increased amounts of fines in the pulp had a
metal ions such as iron can form strongly coloured
negative impact on the bleached pulp brightness in spite of
the improved light scattering ability accompanying the
complexes with lignin and extractives that
addition of fines. A level of fines exceeding 50% was contribute to the darkening of mechanical and
especially detrimental resulting in lower pulp brightness chemimechanical pulps (Gupta 1970; Polcin,
due to a higher light absorption coefficient. Rapson 1972; Janson, Forsskåhl 1989; Ni et al.
Increasing the bleaching temperature did not improve the 1997; Ni et al. 1999; Yoon et al. 1999).
bleaching efficiency. The optimal bleaching temperature at Petit-Conil and Laurent (2003) found that fibre
a hydrogen peroxide charge of 4.5% was found to be 70°C could be bleached to a higher brightness level when
within the tested interval of 70-110°C. For a fines content bleached separately, whereas fines had better
of 25%, the stone groundwood pulp displayed a brightness bleachability when bleached in the entire pulp. The
of 83% whereas the thermomechanical pulp achieved a
fines fraction consumed almost all the hydrogen
brightness of 79%.
peroxide and the consumption by the fines was
ADDRESSES OF THE AUTHORS: Anette Karlsson, explained by higher lignin and wood extractive
([email protected]) SCA R&D Centre AB, SE-851 contents than those of the corresponding fibres.
21 Sundsvall, Sweden, and Mid Sweden University, However, Allison and Graham (1989) studied
FSCN, SE-851 70 Sundsvall, Sweden, Roland Agnemo, hydrogen peroxide bleaching of mechanical pulp
([email protected]) Domsjö fabriker, SE-891 fractions from Radiata pine. Kinetic rate laws for
80 Örnsköldsvik, Sweden. peroxide bleaching were found to be similar for
Corresponding author: Anette Karlsson fibres and fines, indicating similar chromophore
reactivity and thus structure.
The demand for higher brightness in newsprint and Fines are not homogeneous and are generally
high quality printing papers, such as LWC divided into lignin-rich flake-like particles
(lightweight coated) paper and SC (super- originating from the middle lamella and primary cell
calendered) paper, has increased over a number of wall, and fibrillar-like particles originating from the
years. For these applications, the brightness of cellulose-rich secondary wall of the fibres
mechanical pulps, the main components of such (Honkasalo et al. 1983; Luukko 1998). The flake-
papers, must be improved. Hydrogen peroxide is a like fines have been found to have a higher

256 Nordic Pulp and Paper Research Journal Vol 25 no. 3/2010
bleachability compared to the fibrillar-like fines reach the maximum brightness level (Liebergott et
(Haugan, Gregersen 2006). al. 1987; Kappel, Sbaschnigg 1991). The alkali
The brightness response in the bleaching stage is charge could be reduced when the temperature was
dependent on the initial pulp brightness. Generally, increased (Kappel, Sbaschnigg 1991), but at higher
a higher initial unbleached brightness results in hydrogen peroxide charges the brightness became
higher bleached brightness under similar bleaching more sensitive to the caustic charge (Liebergott et
conditions (Kouk et al. 1989). Several factors al. 1987). Contradictory results have been reported
influence the hydrogen peroxide bleaching regarding the brightness obtained at high
performance such as alkalinity, temperature, temperature bleaching compared to conventional
retention time, consistency, and hydrogen peroxide bleaching with both lower (Kappel, Sbaschnigg
stability. Hydrogen peroxide bleaching is performed 1991; Hill et al. 1996; Logenius 2008) and higher
under alkaline conditions in order to produce the (Liebergott et al. 1987) final brightness values
perhydroxyl anion that is considered to be the active having been reported.
bleaching species. It is important to optimize the The present work aims to develop a better
alkali charge as an alkali charge that is too high may understanding of the factors limiting the
lead to pulp darkening, whereas an alkali charge that bleachability of fibre and fines to high brightness
is too low may lead to inefficient bleaching. pH levels. The effect of temperature and alkali charge
levels that are too high may lead to hydrogen during high consistency hydrogen peroxide
peroxide decomposition and exceed the rate of the bleaching of spruce thermomechanical and stone
bleaching reaction resulting in a reduced brightness groundwood pulps containing different amounts of
response. The formation of the perhydroxyl anion is fines will be presented and thoroughly discussed.
also enhanced by increasing the temperature
(Presley, Hill 1996) due to the decrease of the pKa Materials and Methods
value of the hydrogen peroxide anion with increased
Pulps
temperature (Andersson, Amini 1996). Temperature
and retention time are closely related, an increase in Commercially produced pulps, an unbleached
temperature accelerates the hydrogen peroxide thermomechanical pulp (TMP) and a stone
bleaching reactions but also hydrogen peroxide groundwood (SGW) pulp produced from Norway
decomposition. The combination of temperature and spruce (Picea abies) were used in the experiments
retention time have to be chosen so that a reasonable described in this paper. The TMP was metal
portion of the charged hydrogen peroxide is chelated in the mill. Pentanatriumdietylen-
consumed (Lindholm 1999), however a retention triaminpentaacetat (Na5DTPA) was added after the
time that is too long can consume all the hydrogen disc filter and before the first dewatering press. The
peroxide and induce yellowing due to alkaline and pulp was collected after the dewatering press at the
thermal darkening reactions (Leary, Giampaolo blowline from the secondary stage refiner. The
1999). Furthermore, the bleaching consistency SGW pulp was collected from a double wire press
influences the bleaching response that increases before bleaching and was not treated with a
continuously as pulp consistency is increased up to a chelating agent in the mill.
level of about 40% (Kappel, Sbaschnigg 1989).
Hydrogen peroxide is decomposed by transition Metal management
metals such as iron, manganese and copper A chelating stage was carried out in order to reduce
(Colodette et al. 1988), which is undesirable. To the metal content of the pulps prior to the oxidative
reach higher brightness levels, a conventional bleaching with hydrogen peroxide. The TMP was
hydrogen peroxide bleaching process therefore often chelated in the mill with a commercial Na5DTPA
includes a chelating stage where detrimental metal solution (Versenex 80 from Dow, Sweden) at a
ions are removed before bleaching. Hydrogen charge of 0.40%. The pulp was Na5DTPA treated at
peroxide bleaching of mechanical pulps is often ~60°C, pH~5.5 at 8% pulp consistency. After a
performed at 30% pulp consistency at temperatures retention time of 15-30 minutes the pulp was
of between 60 and 75°C, and the pulp is retained in dewatered to about 30% consistency. The
a tower for 1-3 hours (Meyer et al. 1990; fractionated TMP with a fibre/fines ratio of 25/75
Dessureault et al. 1994; Presley, Hill 1996; Dietz et and the SGW pulp was treated in the laboratory with
al. 2009). 0.40% Na5DTPA at room temperature and at 5%
High temperature hydrogen peroxide bleaching pulp consistency. After a retention time of 30
could offer some advantages over conventional minutes, the pulp was dewatered on a polyamide
hydrogen peroxide bleaching. With increasing wire (125 µm) in a Büchner funnel to about 25%
bleaching temperatures, less time was required to consistency, the filtrates being recirculated twice

Nordic Pulp and Paper Research Journal Vol 25 no. 3/2010 257
through the filter cake to avoid the loss of fibres and Hydrogen peroxide bleaching
fines material. To wash out the water-soluble metal- High consistency hydrogen peroxide bleaching was
DTPA complexes, 1 litre deionised water was added performed at a pulp consistency of 30%. Five g
to 200 g pulp in the funnel, repeated twice (a total of (oven-dry) pulp was mixed with the bleaching
3 litres), and pressed to a consistency of chemicals at room temperature for one minute using
approximately 40%. The iron and manganese a stainless steel coffee grinder. The pulp was
contents of the pulps are stated in Table 1. The thereafter immediately placed into a polyethylene
addition level for Na5DTPA is given as commercial plastic bag (thickness 90 µm from Genpack A/S,
product. The chelated pulps were stored in a freezer Denmark), which was sealed and immersed into
at -24°C until use. thermostatic water or glycol bath. The hydrogen
Preparation of pulps with different amounts of fines peroxide charge was 4.5% and the sodium hydroxide
charge was optimised and varied between 0.9 and
Unbleached TMP and SGW pulps were disintegrated
3.6% with the purpose of reaching the highest
in deionised water, diluted to 10 g/l, and fractionated
possible brightness. The sodium silicate (S38 with a
into a fibre and a fines fraction using a Britt dynamic
weight ratio between SiO2 and Na2O of 3.25,
drainage jar with a metal plate wire with 76 µm hole
supplied by PQ Europe, Sweden) charge was kept
diameter (ca. 200 mesh, from PRM Inc., USA). One
constant at 2.9%. The bleaching trials were
litre of the pulp suspension was poured into the
performed at 70°C and 90°C at a retention time of
drainage jar with the agitator at low speed. The
120 minutes and at 110°C at a retention time of 5, 10,
agitation was increased to 900 rpm and the outlet was
15 and 120 minutes. After the bleaching stage, the
opened. The pulp was drained until the level in the
pulp was cooled to room temperature and further
jar sank below the stirrer after which 1 litre deionised
used for determination of optical properties. One
water was added, this was repeated 3 times (a total of
gram of pulp (oven-dry) was diluted with 50 gram
4 litres). The fibre and fines fractions were mixed in
deionised water. After five minutes, the pulp
controlled portions to obtain pulps with different
suspension was dewatered in a Büchner funnel on a
fibre/fines ratios. The pulps were dewatered on a
Munktell 3 filter paper. The filtrate was collected and
polyamide wire of 125 µm in a Büchner funnel, the
used to determine residual hydrogen peroxide. The
filtrates being recirculated twice through the filter
addition level for sodium silicate is given as
cake and pressed to a consistency of approximately
commercial product. Hydrogen peroxide and sodium
40%. The target fines contents of the fractionated
hydroxide are given as 100% active components.
pulps were 0, 25, 50 and 75 weight%. Characteristic
data for unfractionated and fractionated pulps is
stated in Table 1.
Table 1. Freeness, fines content and optical properties together with metal, extractive and lignin contents for unbleached TMP and
SGW pulp at different fibre/fines ratios.
Fines Extractive Lignin
Freeness Brightness1 s457nm k457nm Fe Mn
Pulp content content content
ml CSF % ISO m2/kg m2/kg mg/kg mg/kg
wt% wt% wt%
TMP
Reference 55 31 62.1 64 7.4 1 42 0.9 28.8
Fractionated 5 59.2 43 6.1 1 9 0.2 26.0
Fractionated 28 60.2 63 8.3 2 11 0.8 28.5
Fractionated 49 60.2 80 10.5 3 11 1.2 31.3
Fractionated 75 59.9 90 12.1 6 20 1.8 34.2
Fractionated and
75 60.5 86 11.1 4 <0.05
DTPA washed
SGW pulp
Reference 50 32 67.6 69 5.4 7 2 1.1 29.8
Fractionated 5 66.1 52 4.5 3 1 0.3 27.5
Fractionated 27 67.1 68 5.5 7 1 0.8 29.0
Fractionated 55 67.6 83 6.5 10 2 1.2 30.7
Fractionated 77 67.5 92 7.3 16 3 1.8 33.0
1
Brightness determined on laboratory sheets (65 g/m2) according to ISO standard method 9416.

258 Nordic Pulp and Paper Research Journal Vol 25 no. 3/2010
Preparation of laboratory sheets for evaluation of lignin) and acid-soluble lignin. The acid-insoluble
optical properties lignin was determined gravimetrically according to
Laboratory sheets (60 g/m2, oven-dry) were formed TAPPI method T 222. The acid-soluble lignin was
in a laboratory sheet former using a 76-µm (ca. 200 determined using UV-spectroscopy at 205 nm
mesh) polyamide wire (from Derma, Sweden). The (ε = 128 dm3g-1cm-1) according to the procedure
optical properties (brightness, light scattering (s) and described by Dence (1992). Residual hydrogen
light absorption (k) coefficients) were determined peroxide was determined by iodometric titration
according to ISO standard method 9416. In this paper where the liberated iodine was titrated with 0.1 M
the s- and k-values were determined with the sodium thiosulphate.
brightness function, R457.
Brightness pads (approximately 200 g/m2, oven- Results and discussion
dry) for ISO brightness determination were prepared
Light absorption coefficient (k457nm) for unbleached
according to ISO standard method 3688 (using
pulps with various fines content
funnel and filter paper) and the ISO brightness was
determined according to ISO standard method 2470. In the pulping process, elevated temperatures in the
defibering stage can cause discoloration by reactions
Experimental design of chromophores and leucochromophores located in
One part of the experimental work presented in this the lignin (Gellerstedt, Pettersson 1980; Gellerstedt
paper was designed using the MODDE 5.0 software et al. 1983; Gratzl 1985; Chong et al. 1991). Colour
(Umetrics AB, Umeå, Sweden). The factors reversion induced by heat could also involve the pulp
examined (temperature, sodium hydroxide charge carbohydrates (Polcin, Rapson 1971; Luo et al. 1988;
and fines content) were varied simultaneously over a Holmbom et al. 1992; Grossmann, Ott 1994; Beyer et
set of planned experiments; the results were then al. 1995; Fischer et al. 1995; Tylli et al. 1997). Metal
connected by means of a mathematical model. The ions, especially transition metal ions such as iron can
model is used for interpretation, predictions and form strongly coloured complexes with lignin and
optimization of the process. The data collected by the extractives which contribute to the darkening of
experimental design is used to estimate the mechanical and chemimechanical pulps (Gupta 1970;
coefficients of the model. The model represents the Polcin, Rapson 1972; Janson, Forsskåhl 1989; Ni et
relationship between the response [brightness, light al. 1997; Ni et al. 1999; Yoon et al. 1999).
scattering coefficient (s457nm), light absorption In an unpressurized groundwood process, the
coefficient (k457nm) and hydrogen peroxide grinding temperature will not exceed 100°C
consumption] and the factors (temperature, sodium (Liimatainen et al. 1999), whereas in a TMP process
hydroxide charge and fines content). Data gathered the temperature in the first refiner can be 143°C–
using a central composite and face-centred (CCF) 158°C (Tienvieri et al. 1999), although the
design was used to generate prediction models. All temperature can exceed 170°C in a temperature peak
responses were modeled together using multiple (Härkönen et al. 2003). The typical brightness of
linear regressions (MLR). The experiments were unbleached Norway spruce pulps produced in the
conducted using factors at three levels, i.e. low, TMP process is 57-60% ISO and in the stone ground
medium and high. The following factors were varied: wood process 60-65% ISO (Lindholm 1999).
temperature (70°C, 90°C and 110°C), sodium Previous results indicate that steaming native spruce
hydroxide charge (0.9%, 1.8% and 2.7%) and wood at temperatures below 120°C causes no
fibre/fines ratio (TMP: 72/28, 51/49, 25/75 and SGW brightness losses (Logenius et al. 2005). However,
pulp: 73/27, 45/55, 23/77). The experimental design when steaming TMP from spruce at higher
for hydrogen peroxide bleaching of TMP and SGW temperatures, above 165°C, the brightness declined
pulp with different fibre/fines ratios is shown in by several units (Koskinen et al. 1997).
Table 4. The light absorption coefficient (k457nm) for
unbleached TMP and SGW pulp at various fines
Other analyses
content is shown in Fig 1 and listed in Table 1. The
Analyses and methods not mentioned elsewhere in light absorption coefficient was higher for
the experimental section are as follows: freeness unbleached TMP than for unbleached SGW pulp
(ISO standard method 5267-2), fines content (SCAN and an increased amount of fines increased the light
standard method CM 66:05) and metal ion content absorption coefficient for both pulp types. The
(SCAN standard method CM:38:96). The pulps were increase was, however, most pronounced for the
extracted with dichloromethane to determine the TMP. The data also suggests that both the fibre
extractive content according to SCAN standard fraction and the fines fraction are more coloured in
method C7:62. The lignin content of the pulps is the thermomechanical pulp. One reason for the
given as the sum of acid-insoluble lignin (Klason

Nordic Pulp and Paper Research Journal Vol 25 no. 3/2010 259
higher light absorption coefficient for TMP could be that for the TMP fibres (5% fines content), see
the higher temperatures during processing. The light Table 1. However, at high fines concentrations above
absorption coefficient is known to increase during 50%, the initial differences in light scattering
refining and coloured metal ion complexes are between the two pulps were reduced, and the pulps
suggested to be formed (Johansson, Gellerstedt exhibited a similar light scattering coefficient. This
2000; Johansson et al. 2002). The brightness of the was also shown by Lindholm (1980).
wood raw material may also differ depending on
Hydrogen peroxide bleaching at 70°C
how fresh it is, the ratio of sapwood (brighter) and
heartwood (darker), the growth conditions of the In hydrogen peroxide bleaching, it is important to
wood (fast growth gives brighter wood than slow optimize the alkali charge in order to reach the
growth) and the bark content that is detrimental for highest possible brightness level and to avoid too
the brightness (Brill 1985; Hartler 1986; Höglund, high alkali charges which will result in yellowing
Wilhelmson 1993). due to the prevailing alkaline conditions (Leary,
Giampaolo 1999). Unbleached TMP with different
k 457 nm, m2/kg fibre/fines ratios were bleached using 4.5% hydrogen
14 peroxide at a pulp consistency of 30% and at a
12 temperature of 70°C for 120 minutes. The sodium
10 hydroxide charge was varied between 0.9 and 3.6%.
TMP
8 In most cases, the highest brightness was about 79%
SGW and was achieved at an alkali charge of 2.7%.
6
4 Increasing the sodium hydroxide charge to 3.6% did
2 not increase the brightness any further for most of the
0
pulps (Fig 2). The brightness was somewhat
increased however for the pulp with the lowest
0 25 50 75 100
amount of fines (5% fines). An alkali interval
Fines content, % between 0.9 and 2.7% was therefore suitable for
further bleaching trials since an alkali charge of 3.6%
Fig 1. The light absorption coefficient (k457nm) as a function of
was too high and some alkali darkening reactions
fines content for unbleached TMP and SGW pulp.
occurred for the pulp with the highest amount of
The metal (iron, manganese), extractive and lignin fines (75% fines). The fibre/fines ratio of 51/49
contents increased with increased amount of fines reached the highest brightness of about 80%.
(see Table 1) as expected, despite the fact that the
fractionation procedure and re-addition of fines to the Brightness ISO%
pulp includes extensive washing of the pulp (see 81
51/49
experimental section and discussion below). The 80
manganese content of SGW pulp was lower probably 79 95/5
due to better chelating conditions in the laboratory 78
compared to the conditions prevailing in the mill. 77 72/28
The main source of metals in mechanical pulp is the 76
Ref
wood raw material and the mill water (Read et al. 75
1968; Christiansen, Michalowski 1989). The SGW 0
74 25/75
pulp contained 3-4 times more iron than the TMP, 0 1 2 3 4
even though it is known that the level of iron may NaOH, %
increase during the refining process (Colodette,
Dence 1989). The iron was accumulated in the fines Fig 2. The brightness (ISO 2470) of TMP with different
fraction (Table 1). fibre/fines ratios as a function of sodium hydroxide charge. The
charge of hydrogen peroxide was 4.5%.
Light-scattering coefficient (s457nm) for unbleached It is known that heat and light induces yellowing of
pulps with various fines content lignin-rich mechanical pulps (Gratzl 1985; Schmidt,
The properties of the fractions larger than the fines Heitner 1993; Leary 1994; Paulsson, Ragauskas
set the initial level of light scattering. The addition of 1998; Forsskåhl 2000). Furthermore, transition metal
fines will increase the light scattering coefficient and ions such as iron and copper can cause discoloration,
fines from different types of pulps all improve the either directly (upon ion exchange) or after
light scattering coefficient in a similar way and are subsequent ageing (Gupta 1970; Janson, Forsskåhl
directly additive up to a certain fines level (Lindholm 1989). High iron content has been reported in fibre
1980). The light scattering coefficient of unbleached and fines fractions (Allison, Graham 1989; Haugan,
SGW fibres was 52 m2/kg and 9 units higher than Gregersen 2006). Allison and Graham (1989)

260 Nordic Pulp and Paper Research Journal Vol 25 no. 3/2010
ascribed the high content to the fractionation process. hydrogen peroxide consumption between the two
The light absorption coefficient of the unbleached pulps was seen (Table 1).
fractionated pulp (28% fines) was higher than the The TMP fractionated and remixed pulps contained
reference pulp (31% fines), Table 1. Great care was more manganese, as is shown in Table 1, than the
taken to reduce possible sources of contaminants of SGW pulps.
the samples in the fractionation procedure by using To see how this difference influenced the
deionised water (low metal content), and by storing bleaching response for further trials, a DTPA wash
the samples in cold and dark conditions, but the was performed on the highest manganese containing
cause of the differences cannot be explained. pulp (TMP, fibre/fines ratio 25/75) and the
Nevertheless, the fractionated pulp (28% fines) manganese content was reduced from 20 mg/kg to
showed a similar bleaching response as the reference <0.05 mg/kg. The brightness of the pulps after
pulp (31% fines) (Fig 2). hydrogen peroxide bleaching for the DTPA-treated
High levels of transition metals, particularly and the untreated pulp was essentially the same,
manganese, iron and copper, will catalyse the 78.6% vs. 79.0%, although the DTPA-treated pulp
decomposition of hydrogen peroxide and reduce the consumed less peroxide, 57% compared to 74%,
possibility for reaching high brightness levels probably due to the decreased manganese-induced
(Andersson, Amini 1996). Manganese is loosely decomposition of hydrogen peroxide (cf. Presley,
bound to the wood matrix and can be removed by a Hill 1996). Increased amount of residual hydrogen
chelating agent (e.g. DTPA). Manganese can also be peroxide did not contribute to higher brightness.
removed by cation exchange with calcium or
Hydrogen peroxide bleaching at 110°C
magnesium. Furthermore acidic conditions will
liberate the manganese ions. Iron and copper are The perhydroxyl anion is generally accepted as the
strongly attached to the wood matrix and are active bleaching species in alkaline hydrogen
considerably more difficult to remove (Sundén et al. peroxide systems. The formation of the anion can be
2000). regulated by increasing or decreasing the pH and
The TMP reference (31% fines), which was temperature at a constant hydrogen peroxide
chelated and washed in the mill, had a manganese concentration. Hydrogen peroxide bleaching at
content of 42 mg/kg. The fractionated TMP and higher temperatures of mechanical pulps will
remixed pulp (28% fines) were exposed to a increase the reaction rates of both the bleaching and
considerable amount of deionised water during hydrogen peroxide decomposition reactions.
fractionation and had a manganese content of 11 Because of the increased reaction rates, it is
mg/kg (see Table 1). This difference in manganese necessary to make the appropriate adjustments in
content indicated that a part of the manganese in the retention time (Presley, Hill 1996). Logenius (2008)
reference pulp was to a large extent water-soluble studied high temperature hydrogen peroxide
and complexed with DTPA. As can be seen in Table bleaching of TMP and revealed that a retention time
2, the water-soluble complexed manganese did not of 5-10 minutes was optimal at 105°C. Too long a
interfere in the bleaching under the conditions and time can consume all the hydrogen peroxide and
metal levels tested; no significant difference in induce yellowing due to alkaline and thermal
darkening reactions.
Table 2. The brightness and hydrogen peroxide consumption for TMP at different fibre/fines ratios. Bleaching conditions: 70°C, 120
minutes, 30% pulp consistency. Chemical charges: 4.5% H2O2, 0.9-3.6% NaOH, 2.9% sodium silicate.
NaOH Fractionated TMP
charge Fractionated TMP Reference TMP Fractionated TMP Fibre/fines 25/75
% Fibre/fines 72/28 Fibre/fines 69/31 Fibre/fines 25/75 DTPA washed
Peroxide Peroxide Peroxide Peroxide
Brightness1 Brightness1 Brightness1 Brightness1
consumption consumption consumption consumption
% % % %
% % % %
0.9 76.3 39 76.3 38
1.8 78.3 53 78.5 54
2.7 79.0 66 79.1 69 79.0 74 78.6 57
3.6 79.3 79 79.0 85
1 2
Brightness determined on brightness pads (200 g/m ) according to ISO standard method 2470.

Nordic Pulp and Paper Research Journal Vol 25 no. 3/2010 261
 The hydrogen peroxide bleaching efficiency (i.e. Table 3. Optical properties and hydrogen peroxide consumption
light-absorption coefficient reduction) is probably for TMP at 25/75 fibre/fines ratio. Bleaching conditions: 110°C,
due to the balance between the chromophore- 5-15 minutes, 30% pulp consistency. Chemical charges: 4.5%
reducing reactions of hydrogen peroxide and the H2O2, 2.7% NaOH, 2.9% sodium silicate.
chromophore-creating reactions initiated by alkali. Peroxide
Time Brightness1 s457nm k457nm
When the hydroxide ion concentration increases consumption
min % m2/kg m2/kg
more than the perhydroxyl anion concentration, the %
alkaline chromophore-creating reactions start to 5 76.0 87 3.2 67
prevail over the bleaching reactions. A temperature 10 76.4 82 3.0 76
increase will favour hydrogen peroxide 15 76.4 82 3.0 77
decomposition rather than bleaching reactions due 1
Brightness determined on laboratory sheets (65 g/m2) according
to the higher activation energy (Moldenius, Sjögren to ISO standard method 9416.
1982).
Hydrogen peroxide bleaching at various temperatures
Hydrogen peroxide bleaching of TMP (fibre/fines evaluated with experimental design
ratio 25/75) and SGW pulp (fibre/fines ratio 23/77)
The TMP and SGW pulp bleaching trials were
at 110°C for 120 minutes with an alkali charge of
evaluated using the MODDE 5.0 software. Table 4
2.7% was too long, as expected, since all the
shows the experimental design and basic data
hydrogen peroxide was consumed resulting in low
analysed.
brightness of 70.3% and 67.2% respectively.
When fitting a model the most important diagnostic
Hydrogen peroxide bleaching trials using different
tool consists of the two companion parameters R2 and
retention times at 110°C revealed that 10-15 minutes
Q2. R2 is a measure of fit, i.e. how well the model fits
was optimal (Table 3). Increased retention time was
the data whereas Q2 indicates how well the model
not expected to give any higher brightness since 3/4
predicts new data. R2 varies between 0 and 1, where
of the hydrogen peroxide was consumed. A retention
1 indicates a perfect model. Similarly to R2, Q2 has
time of 10 minutes was therefore chosen for further
the upper limit of 1. A Q2>0.5 should be regarded as
bleaching trials.
good and Q2>0.9 as excellent.

Table 4. The MODDE experimental design and responses (optical properties and hydrogen peroxide consumption) for hydrogen
peroxide bleaching of TMP and SGW pulp with different fibre/fines ratios. Chemical charges: 4.5% H2O2, 2.9% sodium silicate.
k457nm Peroxide
Temperature NaOH Fines content Brightness1 s457nm
m2/kg consumption
°C % % % m2/kg
%
TMP SGW TMP SGW TMP SGW TMP SGW TMP SGW
70 0.9 28 27 76.5 79.4 63 69 2.3 1.8 39.4 44.4
70 0.9 75 77 76.9 78.1 98 93 3.4 2.9 44.6 81.1
70 1.8 49 55 79.0 81.4 77 80 2.2 1.7 50.0 69.1
70 2.7 28 27 79.2 82.8 57 65 1.5 1.2 65.9 61.1
70 2.7 75 77 78.7 80.9 85 86 2.5 2.0 73.8 85.1
90 0.9 49 55 77.1 76.5 80 85 2.7 3.1 35.6 76.0
90 1.8 28 27 77.4 79.2 60 67 2.0 1.8 60.6 68.9
90 1.8 49 55 77.9 78.1 75 81 2.3 2.5 57.7 81.8
90 1.8 49 55 78.4 78.2 75 82 2.2 2.5 53.9 80.9
90 1.8 49 55 78.7 78.1 76 81 2.2 2.5 54.3 81.1
90 1.8 75 77 77.0 76.6 86 94 2.9 3.4 65.5 90.9
90 2.7 49 55 79.3 80.0 67 73 1.8 1.8 71.5 89.3
110 0.9 28 27 75.1 78.8 63 71 2.6 2.0 33.1 42.4
110 0.9 75 77 75.8 77.7 94 98 3.6 3.1 36.4 74.2
110 1.8 49 55 77.4 79.9 79 79 2.6 2.0 46.0 64.7
110 2.7 28 27 77.3 81.2 55 63 1.9 1.4 59.4 68.0
110 2.7 75 77 76.4 78.7 82 85 3.0 2.4 75.6 85.6
1
Brightness determined on laboratory sheets (65g/m 2) according to ISO standard method 9416.

262 Nordic Pulp and Paper Research Journal Vol 25 no. 3/2010
From the MODDE model used in this work, the R2 work it was shown that an approximate 30%
values for the responses were >0.90 and the Q2 decrease in NaOH charge resulted in unchanged
values >0.84 which is considered to be very good. brightness but still lower than that obtained at 70°C.
Fig 3 shows the regression coefficients of the Kappel and Sbaschnigg (1991) reported a 30%
interaction model and their 95% confidence intervals. reduction in optimum NaOH charge (i.e. at optimal
The first three coefficients, also called linear terms, brightness) when the temperature was increased from
reveal the real effects of the three factors 60 to 95°C but the higher temperature resulted in
(temperature, NaOH charge, fines content). The last lower brightness than that obtained at 60°C.
coefficient shows whether there is any interaction The MODDE model coefficient plot indicates that
among the factors. The uncertainty of the coefficients alkali charge and fines content were important for the
is given by the confidence intervals. light scattering coefficient (Fig 4). It can be seen that
k 457 nm, m 2/kg fines had the greatest impact on the light scattering;
1.0 increased fines content increased the light scattering
coefficient. The second most influential factor was
0.5
the NaOH charge. Increased sodium hydroxide
charge decreased the light scattering coefficient,
SGW probably due to the improved bonding resulting in
0.0 TMP increased sheet density (Engstrand, Sjögren 1991).
The temperature and the interaction term
-0.5
“Temperature x NaOH” did not affect the light
scattering coefficient as the confidence interval
Temp NaOH Fines Temp x NaOH
included zero and was statistically insignificant.
-1.0
s 457 nm, m2/kg
20
Fig 3. Hydrogen peroxide bleaching of TMP and SGW pulp with
15
different fibre/fines ratios. Scaled and centred regression
coefficients of the interaction model for the light absorption 10
coefficient with 95% confidence interval as error bars. SGW
The coefficient plot displays the impact for each 5 TMP
factor for a response. The MODDE model coefficient
0
plot indicates that the main effects, i.e. temperature,
alkali (NaOH) charge, and fines content, were -5
important for the light absorption coefficient (Fig 3). Temp NaOH Fines Temp x NaOH
It can be seen that fines content had the strongest -10
impact on the light absorption coefficient. When the
fines content was increased from the medium level, Fig 4. Hydrogen peroxide bleaching of TMP and SGW pulp with
50%, to its high level, 75%, and the other factors different fibre/fines ratios. Scaled and centred regression
remained fixed at their medium levels, the light coefficients of the interaction model for the light scattering
absorption coefficient increased by 0.5 units. The coefficient with 95% confidence interval as error bars.
second most influential factor was the sodium The following charts in this paper are based on the
hydroxide charge, where an increased caustic charge predicted values from the MODDE 5.0 model. The
decreased the light absorption coefficient. The third predicted values of optical properties and hydrogen
factor, temperature, had the lowest impact on the peroxide consumption are also given in Table 5,
light absorption coefficient with increased Appendix.
temperature increasing the light absorption Hydrogen peroxide bleaching of TMP and SGW
coefficient to some extent. The interaction term pulp at 70, 90 and 110°C revealed that the highest
“Temperature x NaOH” was statistically insignificant brightness and lowest light absorption coefficient
as the confidence interval included zero and no was obtained at an alkali charge of 2.7%
interaction between sodium hydroxide and irrespective of the fines content of the pulp or the
temperature was seen, meaning that it was not bleaching temperature (see Table 5, Appendix).
beneficial to lower the sodium hydroxide charge Haugan and Gregersen (2006) reported that
when the temperature was increased in the chosen hydrogen peroxide bleaching of fines needed quite
intervals. The chosen interval for the NaOH charge high sodium hydroxide charges to reach the lowest
was 0.9, 1.8 and 2.7% and the difference in level was light absorption coefficients. The bleaching was
probably too large to see any NaOH charge reduction performed at 3% pulp consistency which is different
when the temperature increased. In Logenius (2008) from the high consistency conditions (30%) used in

Nordic Pulp and Paper Research Journal Vol 25 no. 3/2010 263
this investigation and those prevalent in industrial units for both pulps (Table 5, Appendix). The
processes. The lowest light absorption coefficients divergent difference in the light absorption
were obtained at a bleaching temperature of 70°C coefficient between unbleached TMP and SGW
(NaOH charge 2.7%), see Figs 5 and 6. pulp towards higher fines content (Fig 1) was, after
k 457 nm, m2/kg hydrogen peroxide bleaching (at optimal
4 conditions), rather constant and the difference after
bleaching was about 0.5 units (Fig 7).
3 TMP 25/75
k 457 nm, m2/kg
TMP 50/50 3
2
TMP 75/25
1 2 TMP
SGW
0
60 70 80 90 100 110 120 1
Temperature, °C
0
Fig 5. The light absorption coefficient (k457nm) as a function of
temperature for hydrogen peroxide bleached TMP with 0 25 50 75 100
fibre/fines ratio of 25/75, 50/50 and 75/25%. Bleaching Fines content, %
conditions: 4.5% H2O2, 2.7% NaOH, 2.9% Silicate, 70°C, 120
minutes, 30% consistency. Predicted values. Fig 7. The light absorption coefficient (k457nm) as a function of
fines content for hydrogen peroxide bleached TMP and SGW
pulp. Bleaching conditions: 4.5% H2O2, 2.7% NaOH, 2.9%
k 457 nm, m2/kg
Silicate, 70°C, 120 minutes, 30% consistency. Predicted
4 values, 95% confidence interval as error bars.
3 The SGW pulp reached a higher final brightness
SGW 25/75
than the TMP (Fig 8) due to a lower light absorption
SGW 50/50
2 coefficient (Fig 7). Although increased fines content
SGW 75/25 increased the light absorption coefficient, the TMP
1 brightness was increased when the fines content
increased from 25% to 50% (Fig 8). This was due to
0 the increase in the light scattering coefficient (Fig 9).
60 70 80 90 100 110 120 Increasing the fines content further from 50% to 75%
Temperature, °C decreased the brightness for both pulps since the
increased light scattering coefficient (Fig 9) could
Fig 6. The light absorption coefficient (k457nm) as a function of not compensate for the increased light absorption
temperature for hydrogen peroxide bleached SGW pulp with coefficient (Fig 7).
fibre/fines ratio of 25/75, 50/50 and 75/25%. Bleaching
conditions: 4.5% H2O2, 2.7% NaOH, 2.9% Silicate, 70°C, 120 Brightness, %
minutes, 30% consistency. Predicted values. 85
Increased temperature did not reduce the k-value to 84
83
the same extent. In the case of SGW pulp with the 82 SGW
highest fines content (75%), hydrogen peroxide 81
bleaching at 90°C for 120 minutes seemed to be too TMP
80
long as almost all peroxide (97%) was consumed 79
resulting in a light absorption coefficient that was 78
77
probably too high (Fig 6). The SGW pulp seemed to 0
76
be more sensitive for the 90°C temperature and 120
0 25 50 75 100
minutes retention time compared to TMP (cf. Fig 5
and Fig 6) where the light absorption coefficient Fines content, %
and the hydrogen peroxide consumption were higher Fig 8. The brightness as a function of fines content for hydrogen
(Table 5, Appendix), although the SGW pulp peroxide bleached TMP and SGW pulp. Bleaching conditions:
contained a small amount of manganese. 4.5% H2O2, 2.7% NaOH, 2.9% Silicate, 70°C, 120 minutes,
The iron content was, however, somewhat higher as 30% consistency. Predicted values, 95% confidence interval as
shown in Table 1. Increasing the temperature from error bars.
70°C up to 110°C decreased the brightness by 2

264 Nordic Pulp and Paper Research Journal Vol 25 no. 3/2010
 s 457 nm, m2/kg
100 Acknowledgements
80
SGW Financial support from the Knowledge Foundation is gratefully
60 acknowledged. The authors thank Professor Per Engstrand,
TMP Mid Sweden University, Sundsvall Sweden for valuable
40 discussions and Dr Magnus Paulsson, Eka Chemicals AB,
Bohus, Sweden and Mid Sweden University, Sundsvall,
20
Sweden for commitment and valuable comments..
0
Literature
25 50 75
Fines content, %
Allison, R.W. and Graham, K.L. (1989): Peroxide bleaching of
Fig 9. The light scattering coefficient (s457nm) as a function of mechanical pulp fractions from Radiata pine. J. Pulp Paper Sci.
fines content for hydrogen peroxide bleached TMP and SGW 15(4), 145-150.
pulp. Bleaching conditions: 4.5% H2O2, 2.7% NaOH, 2.9% Andersson, J.R, and Amini, B. (1996): Hydrogen peroxide
Silicate, 70°C, 120 minutes, 30% consistency. Predicted bleaching, In: Dence, C.W. and Reeve, D.W. (eds.), Pulp
values, 95% confidence interval as error bars. Bleaching: Principles and Practice, TAPPI Press, Atlanta, GA,
USA, pp. 411–442.
Conclusions Beyer, M., Baurich, Ch. and Fischer, K. (1995): Mechanismen
der licht- und wärmeinduzierten vergilbung von faserstoffen, Das
The objective of this work was to study fines as a Papier 49(10A), V8-V14.
limiting factor for reaching higher brightness levels Boutelje, J. and Eriksson, I. (1984): Analysis of lignin in
in hydrogen peroxide bleaching of mechanical pulps. fragments from thermomechanical spruce pulp by ultraviolet
For unbleached TMP and SGW pulps, an increased microscopy, Holzforschung 38(5), 249-252.
amount of fines in the fractions increased, as Brill, J.W. (1985): Effects of wood and chip quality on TMP
expected, the light absorption coefficients and the properties, Int. Mech. Pulp Conf, Stockholm, Sweden, May 6-10,
amounts of extractives, lignin and transition metals SPCI, Stockholm, Sweden, pp.153-161.
(Fe, Mn). Unbleached TMP had a higher light Chong, J.M., Nanayakkara, N.P.D. and Whiting, P. (1991):
absorption coefficient than unbleached SGW pulp, Model compound studies of thermal reversion chemistry, J. Pulp
and both the fibres and fines were more coloured. Paper Sci. 17(1), J18-J21.
The bleaching temperature in the interval 70-110°C Christiansen, S.H. and Michalowski, R.J. (1989): Putting
did not affect the light scattering coefficient of the metals in their place: A new solution, PIMA 71(12), 21-25.
pulps. The light scattering coefficients decreased, as Colodette, J.L., Rothenberg, S. and Dence, C.W. (1988):
expected, for all TMP and SGW pulp fibre/fines Factors affecting hydrogen peroxide stability in the brightening of
fractions with an increased charge of sodium mechanical and chemimechanical pulps. Part 1: Hydrogen
hydroxide. The hydrogen peroxide bleaching peroxide stability in the absence of stabilizing systems, J. Pulp
efficiency (i.e. light absorption coefficient reduction) Paper Sci. 14(6), 126-132.
in the 70-110°C temperature interval was found to be Colodette, J.L. and Dence, C.W. (1989): Factors affecting
best at the lowest bleaching temperature, when hydrogen peroxide stability in the brightening of mechanical and
residence time and alkali charge were also taken into chemimechanical pulps. Part IV: The effect of transition metals in
consideration. In most cases increased amounts of Norway spruce TMP on hydrogen peroxide stability, J. Pulp
fines in the pulp had a negative effect on the Paper Sci. 15(3), 79-83.
bleached pulp brightness. This was to some extent Dence, C.W. (1992): The determination of lignin, In: Lin, S.Y and
compensated with an increased light scattering Dence, C.W (eds.), Methods in Lignin Chemistry, Springer-
Verlag Berlin Heidelberg, Germany, pp. 34-61.
coefficient. The SGW pulp displayed a better final
brightness level, 2-4 brightness units higher than the Dessureault, S., Lafrenière, S., Barbe, M.C., Leduc, C. and
brightness of the TMP under similar bleaching Daneault, C. (1994): Bleaching process for the production of
mechanical and chemi-mechanical pulps of high brightness, Pulp
conditions. However, the SGW pulp seemed to be
Paper Can. 95(7), 264-272.
more sensitive to the 90°C temperature and longest
(120 minutes) retention time compared to the TMP. Dietz, T., Hopf, B., Schmidt, K. and Süss, H.U. (2009):
Aspects of optimisation of mechanical pulp bleaching with
At a hydrogen peroxide charge of 4.5% and at a fines
hydrogen peroxide, Appita J. 62(5), 335-338.
content of 25%, the stone groundwood pulp
Engstrand, P., Sjögren, B. (1991): The significance of
displayed a brightness of 83% whereas the thermo-
carboxylic groups for the physical properties of mechanical pulp
mechanical pulp reached a brightness of 79%. fibers, 6th Int. Symp. Wood Pulp. Chem., Melbourne, Australia,
April 28- May 3, APPITA, Parkville, Australia, vol. 1, pp. 75-79.

Nordic Pulp and Paper Research Journal Vol 25 no. 3/2010 265
Fischer, K., Koch, H. and Beyer M. (1995): Mechanisms of light Kappel, J. and Sbaschnigg, J. (1989): Peroxide bleaching of
and heat induced yellowing – A comparison, 8th Int. Symp. groundwood pulp at consistencies up to 40% and temperatures
Wood Pulp. Chem., Helsinki, Finland, June 6-9, KCL Finnish up to 95°C, 75th Annual Meeting of the Technical Section of the
Pulp and Paper Research Institute, Jyväskylä, Finland, vol. 1, Canadian Pulp and Paper Association, Montréal, Canada,
pp. 429-434. January 31-February 1, PAPTAC, Montréal, Canada, pp. B35-
Gellerstedt, G. and Pettersson, B. (1980): Autoxidation of B39.
lignin, Svensk Papperstidning 83(11), 314-318. Kappel, J. and Sbaschnigg, J. (1991): Bleaching of
Gellerstedt, G., Pettersson, I. and Sundin, S. (1983): Light- groundwood pulp at temperatures up to 95°C, Pulp Paper Can.
induced and heat-induced yellowing of mechanical pulps, 92(9), 229-234.
Svensk Papperstidning 86(15), R157-R163 Kleen, M., Kangas, H. and Laine, C. (2003): Chemical
Gratzl, J.S. (1985): Lichtinduzierte vergilbung von zellstoffen – characterization of mechanical pulp fines and surface layers,
ursache und verhütung, Das Papier 39(10A), V14-V23. Nord. Pulp Paper Res. J. 18(4), 361-368.
Grossmann, K. and Ott, G. (1994): Licht- und hitzevergilbung Koskinen, J., Salerma, M. and Forsskåhl, I. (1997): Hot-shot
unterschiedlicher papierfaserstoffe – ein vergleich, Das Papier rapid heat aging of mechanical pulps. Int. Mech. Pulp. Conf.,
48(10), 623-627. Stockholm, Sweden, June 9-13, SPCI, Stockholm, Sweden, pp.
Gupta, V.N. (1970): Effect of metal ions on brightness, 377-379.
bleachability and colour reversion of groundwood, Pulp Paper Kouk, R.S., Meyrant, P. and Dodson, M.G. (1989): Hydrogen
Mag. Can. 71(18), 69-77. peroxide bleaching of spruce high-yield pulp at varying degrees
Forsskåhl. I. (2000): Brightness reversion. In: Stenius P. (ed.). of sulfonation, J. Pulp Paper Sci. 15(4), 151-157.
Forest Products Chemistry. Fapet Oy, Helsinki, Finland, pp. 278- Leary, G.J. (1994): Recent progress in understanding and
332. inhibiting the light-induced yellowing of mechanical pulps, J. Pulp
Hartler, N. (1986): Wood quality requirements in mechanical Paper Sci. 20(6): J154-J160.
pulping, Nord. Pulp Paper Res. J. 1(1), 4-10. Leary, G. and Giampaolo, D. (1999): The darkening reactions
Haugan, M. and Gregersen, Ø (2006): Hydrogen peroxide of TMP and BTMP during alkaline peroxide bleaching, J. Pulp
bleaching of mechanical pulp fines, Nord. Pulp Paper Res. J. Paper Sci. 25(4), 141-146.
21(1), 105-110. Leduc, C. and Daneault, C. (2007): Impact of mechanical pulp
Hill, R., Hill, P., Huey, J. and Plumet, L. (1996): High fines on the efficiency of peroxide bleaching of TMP pulp, Cellul.
temperature peroxide bleaching of mechanical pulps, Pulping Chem. Technol. 41(7-8), 399-404.
Conference, Nashville, TN., USA, October 27-31, TAPPI, Liebergott, N., Van Lierop, B. and Heitner, C. (1987): Rapid
Atlanta, GA, USA, pp.407-409. high temperature peroxide brightening of high yield pulps –
Honkasalo, J., Knuts, K. Laine, J.E. and Nygren-Konttinen, silicate free hydrogen peroxide treatment, Int. Mech. Pulp Conf.,
A. (1983): The fines fraction of mechanical pulp – A litterature Vancover, British Columbia, Canada, June 2-5, CPPA,
review, Paperi Puu 65(11), 693-699. Vancouver, Canada, pp.75-80.
Holmbom, B., Ekman, R. and Eckerman, C. (1992): Liimatainen, H., Haikkala, P., Lucander, M., Karojärvi, R. and
Degradation products formed during light and heat treatment of Tuovinen, O. (1999): Grinding and pressure grinding, In:
spruce groundwood, J. Pulp Paper Sci. 18(4), J146-J150. Sundholm, J. (ed.), Mechanical Pulping, Fapet Oy, Helsinki,
Finland, pp. 105-158.
Härkönen, E., Kortelainen, J., Virtanen, J. and Vuorio, P.
(2003): Fiber development in TMP main line, Int. Mech. Pulp Lindholm, C.A. (1980): Comparison of some papermaking
Conf, Québec, Canada, June 2-5, PAPTAC, Montréal, Canada, properties of groundwood, pressure groundwood and
pp. 171-178. thermomechanical pulp by means of artificial blends of pulp
fractions, Paperi Puu 62(10), 593-606.
Höglund, H. and Wilhelmson, K. (1993): The product must
determine the choice of wood type in mechanical pulping, Int. Lindholm, C.-A. (1999): Bleaching, In: Sundholm, J. (ed.),
Mech. Pulp Conf., Oslo, Norway, June 15-17, PTF, Oslo, Mechanical Pulping, Fapet Oy, Helsinki, Finland, pp. 311-343.
Norway, pp.1-22. Logenius, L., Agnemo, R. and Höglund, H. (2005): High
Janson, J. and Forsskåhl, I. (1989): Color changes in lignin- temperature bleaching of mechanical pulps from spruce, Int.
rich pulps on irradiation by light, Nord. Pulp Paper Res. J. 4(3), Mech. Pulp. Conf., Oslo, Norway, June 7-9, PTF, Oslo, Norway,
197-205. pp. 80-84.
Johansson, M., Zhang, L. and Gellerstedt, G. (2002): On Logenius, L. (2008): The influence of high temperature on the
chromophores and leucochromophores formed during the discoloration and degradation of carbohydrates and on hydrogen
refining of wood, Nord. Pulp Paper Res. J., 17(1), 5-8. peroxide bleaching of spruce TMP, PhD thesis, Mid Sweden
University, Sundsvall, Sweden.
Johansson, M. and Gellerstedt, G. (2000): Chromophoric
content in wood and mechanical pulps, Nord. Pulp Paper Res. J. Luo, Ch., Putz, L. and Göttsching, L. (1988): Untersuchungen
15(4), 282-286. zur licht- und hitzeinduzierten Vergilbung von gebleichten
ligninhaltigen Primär- und Sekundärfaser-stoffen, Das Papier
Kangas, H. and Kleen, M. (2004): Surface chemical and
42(10A), V55-V67.
morphological properties of mechanical pulp fines, Nord. Pulp
Paper Res. J. 19(2), 191-199. Luukko, K. (1998): On the characterization of mechanical pulp
fines, Paperi Puu 80(6), 441-448.

266 Nordic Pulp and Paper Research Journal Vol 25 no. 3/2010
Meyer, K.A., Kappel, J. and Petschauer, F. (1990): Criteria for Read, D.W., Eade, B.D. and Slingsby, N.R. (1968): The origin
the selection of optimum bleaching systems for mechanical pulp, and some effects of contaminating metal ions in the groundwood
Pulping Conference, Toronto, Canada, October 14-17, TAPPI bleaching environment, Pulp Paper Mag. Can. 69(13), 51-58.
Press, Atlanta, GA, USA, pp. 291-297. Roick, T., Schempp, W. and Krause, T. (1991): Mechanical
Moldenius, S. and Sjögren, B. (1982): Kinetic models for pulp fines: some causes for their poor bleachability, Papier
hydrogen peroxide bleaching of mechanical pulps. J. Wood 45(10A), 23-26.
Chem. Technol. 2(4), 447-471. Schmidt, J.A. and Heitner, C. (1993): Use of UV-visible diffuse
Ni, Y., Li, Z. and van Heiningen, A.R.P. (1997): Minimization of reflectance spectroscopy for chromophore research on wood
the brightness loss due to metal ions in process water for fibers: a review, Tappi J. 76(2), pp. 117-123.
bleached mechanical pulps, Pulp Paper Can. 98(10), 72-75. Sorvari, J., Sjöström, E., Klemola, A. and Laine, J.E. (1986):
Ni, Y., Ng, A. and Mosher, M. (1999): A model compound study Chemical characterization of wood constituents, especially lignin,
–The formation of colored metallic extractive complexes and in fractions separated from middle lamella and secondary wall of
their effect on the brightness of TMP pulp, J. Wood Chem. Norway spruce (Picea abies), Wood Sci. Technol. 20(1), 35-51.
Technol. 19(3), 213-223. Sundberg, A., Pranovich, A.V. and Holmbom, B. (2003):
Paulsson, M. and Ragauskas, A.J. (1998): Chemical Chemical characterization of various types of mechanical pulp
modification of lignin-rich paper. Part 9. Effect of dry heat and fines, J. Pulp Paper Sci. 29(5), 173-178.
moist heat on the accelerated yellowing of untreated and Sundén, A., Brelid, H., Rindby, A. and Engström, P. (2000):
acetylated high-yield pulps, Nord. Pulp Paper Res. J. 13(3), 191- Spatial distribution and modes of chemical attachment of metal
197. ions in spruce wood, J. Pulp Paper Sci. 26(10), 352-357.
Petit-Conil, M. and Laurent, A. (2003): Bleaching behaviour of Tienvieri, T., Huusari, E., Sundholm, J., Vuorio, P.,
softwood TMP, PGW or CTMP fibre fractions, Int. Mech. Pulp. Kortelainen, J., Nystedt, H. and Artamo, A. (1999):
Conf., Quebec, Canada, June 2-5, PAPTAC, Montréal, Canada, Thermomechanical pulping, In: Sundholm, J. (ed.), Mechanical
pp. 105-114. Pulping, Fapet Oy, Helsinki, Finland, pp. 157-221.
Polcin, J. and Rapson, W.H. (1971): Sapwood and heartwood Tylli, H., Forsskåhl, I. and Olkkonen, C. (1997): Effect of heat,
groundwood of Western hemlock and Jack pine Part II. Heat light and infrared radiation on chemimechanical pulp studied by
stability of extractives, Pulp Paper Mag. Can. 72(10), T324-T330 fluorescence spectroscopy, 9th Int. Symp. Wood Pulp. Chem.,
Polcin, J. and Rapson, W.H. (1972): Sapwood and heartwood Montréal, Canada, June 9-12, CPPA, Montréal, Canada, pp.
groundwood of western hemlock and jack pine. Part III. Influence 116-1-116-4.
of solvent extraction on the bleaching of pulps, Pulp Paper Mag. Yoon, B.-H., Wang, L.-J. and Kim, G.-S. (1999): Formation of
Can. 73(1), 86-92. lignin-metal complexes by photo-irradiation and their effect on
Presley, J.R. and Hill, R.T. (1996): Peroxide bleaching of colour reversion of TMP, J. Pulp Paper Res. J. 25(8), 289-293.
(chemi) mechanical pulps, In: Dence, C.W. and Reeve, D.W.
(eds.), Pulp Bleaching: Principles and Practice, TAPPI Press,
Atlanta, GA, USA, pp. 459-489. Manuscript received October 22, 2009
Accepted June 4, 2010

Nordic Pulp and Paper Research Journal Vol 25 no. 3/2010 267
Appendix
Table 5. The predicted (MODDE 5.0) optical properties and consumed hydrogen peroxide for hydrogen peroxide bleached TMP and
SGW pulp with different fibre/fines ratios.
Peroxide
Temperature NaOH Fines Brightness1 s457nm k457nm
consumption
°C % % % m2/kg m2/kg
%
TMP SGW TMP SGW TMP SGW TMP SGW
70 0.9 25 76.1 79.4 62 68 2.3 1.8 41.2 45.2
70 1.8 25 77.6 81.3 58 66 1.9 1.4 53.6 54.0
70 2.7 25 79.1 83.2 55 64 1.5 1.0 65.9 62.8
70 0.9 50 77.8 79.5 83 79 2.6 2.1 34.9 60.5
70 1.8 50 79.0 81.2 78 76 2.2 1.6 49.3 65.9
70 2.7 50 80.2 82.9 73 74 1.8 1.2 63.7 71.3
70 0.9 75 76.9 78.2 97 94 3.4 2.9 43.7 78.0
70 1.8 75 77.7 79.7 90 90 2.9 2.4 60.1 80.1
70 2.7 75 78.6 81.1 84 86 2.5 1.9 76.6 82.1
90 0.9 25 75.6 77.0 59 71 2.3 2.5 45.8 56.1
90 1.8 25 76.9 78.5 55 68 1.9 2.1 59.4 66.9
90 2.7 25 78.2 80.1 51 65 1.5 1.7 73.0 77.7
90 0.9 50 77.3 77.1 81 83 2.7 2.8 39.5 71.3
90 1.8 50 78.3 78.5 75 78 2.3 2.4 55.1 78.8
90 2.7 50 79.2 79.8 70 74 1.9 1.9 70.8 86.2
90 0.9 75 76.4 75.8 94 97 3.4 3.6 48.3 88.8
90 1.8 75 77.1 76.9 87 92 3.0 3.1 65.9 92.9
90 2.7 75 77.7 78.0 80 86 2.6 2.7 83.6 97.0
110 0.9 25 74.9 78.8 61 70 2.6 2.0 34.1 40.0
110 1.8 25 75.9 80.0 57 66 2.3 1.7 48.9 52.8
110 2.7 25 77.0 81.2 53 62 1.9 1.3 63.8 65.7
110 0.9 50 76.6 79.0 83 82 3.0 2.3 27.8 55.2
110 1.8 50 77.3 80.0 77 77 2.6 1.9 44.7 64.7
110 2.7 50 78.0 80.9 72 71 2.2 1.5 61.5 74.2
110 0.9 75 75.7 77.6 96 97 3.7 3.1 36.6 72.8
110 1.8 75 76.1 78.4 89 90 3.3 2.7 55.5 78.9
110 2.7 75 76.4 79.1 82 83 2.9 2.3 74.4 85.0
1
Brightness determined according to ISO standard method 9416.

268 Nordic Pulp and Paper Research Journal Vol 25 no. 3/2010