Partial exchange of ozone by electron beam irradiation shows better viscosity control and less oxidation in cellulose upgrade scenarios

Carbohydrate Polymers 265 (2021) 118037

Contents lists available at ScienceDirect

Carbohydrate Polymers
journal homepage: www.elsevier.com/locate/carbpol

Partial exchange of ozone by electron beam irradiation shows better
viscosity control and less oxidation in cellulose upgrade scenarios
Oliver P. Sarosi a, 1, Daniela Bammer a, 1, Elisabeth Fitz a, 1, Antje Potthast b, *
a

Kompetenzzentrum Holz GmbH, Altenbergerstraße 69, A-4040, Linz, Austria
Institute of Chemistry of Renewable Resources, Department of Chemistry, University of Natural Resources and Life Sciences, Konrad-Lorenz-Straße 24, A-3430, Tulln,
Austria

b

A R T I C L E I N F O

A B S T R A C T

Keywords:
Kraft pulp
Xylan
Hexenuronic acid
Chemical-free
Bleaching

Electron beam irradiation (EBI) is an alternative treatment for intrinsic viscosity (IV) control in cellulose pulps,
but has never been integrated in full bleaching sequences for comparison to conventional methods. Both euca­

lyptus kraft (EK) paper pulp and beech sulfite (BS) dissolving pulp were subjected to totally chlorine free (TCF)
bleaching sequences comprising either EBI, ozone, or both for IV control. Additionally, effects of EBI on hex­
enuronic acid (HexA) and xylan were investigated. IV was adjusted to 450–500 mL g− 1 and properties including
carbonyl content, kappa, brightness, alkali-resistance and sugar composition were compared. Pulps produced
with EBI had a higher alkali-resistance, uniformity and less cellulose oxidation. However, the degree of bleaching
(DoB) was low without the use of ozone. HexA content in a birch pulp was halved by EBI. Isolated xylans were
more resistant to irradiation than cellulose with little decrease of molar masses and moderate oxidation.

1. Introduction
With an increasing demand for environmentally friendly textiles, the
market for regenerated cellulose fibers (RCF) is projected to show a 4.2
% compounded annual growth rate, despite the recent events sur­
rounding COVID-19 (GlobeNewswire, 2020). To enter the RCF market,
the upgrade of existing low-value paper pulp mills to high-value dis­
solving pulp mills or swing mill operation has been considered in the
past (Lundberg, Axelsson, Mahmoudkhani, & Berntsson, 2012; Sappi
Limited, 2013).
Converting paper pulp to dissolving pulp requires the removal of
hemicellulose through either acid prehydrolysis before pulping or
alkaline extraction or enzymatic degradation during bleaching (Hut­
terer, Schild, & Potthast, 2016; Hutterer, Kliba, Punz, Fackler, & Pot­
thast, 2017; Sixta, 2006). Dissolving pulp generally requires a high
content of α-cellulose, a brightness above 90%ISO, a narrow molecular
weight distribution (MWD), and low IV. Thus, an upgraded paper mill or

swing mill needs a method for IV reduction, which can be achieved by
adjusting the process parameters, such as cooking intensity, ozone
charge during TCF bleaching or dwell times during steeping and accel­
erated aging. However, intensified cooking conditions lead to a sub­
stantial cellulose yield loss and a conversion of α-cellulose to undesirable

alkali-soluble fragments (Agarwal & Gustafson, 1997; Kubes, Fleming,
Macleod, Bolker, & Werthemann, 1983). Furthermore, alkaline steeping
and pulp pre-aging oxidize the pulp, introducing both carbonyl and
carboxyl groups, which are responsible for pulp brightness reversion
ăf, So
ăderlund, & Germgård, 2006; Mozdynie­
(Ahn et al., 2019; Kvarnlo
wicz, Nieminen, & Sixta, 2013).
Ozone rapidly decomposes under aqueous conditions during
bleaching, yielding various radical and peroxo-species (Staehelin,
Buehler, & Hoigne, 1984). Ozone itself and all subsequent radical spe­
cies can cleave the glycosidic bond of cellulose, especially if the tem­
perature is raised and transition metals such as iron and cobalt are
present, favoring the Fenton-type decomposition of ozone (Kishimoto &

Abbreviations: BS, beech sulfite pulp; CCOA, [2-(2-aminooxyethoxy)-ethoxy]-amide; DoB, degree of bleaching; DP, degree of polymerization; DS, degree of
substitution; EBI, electron beam irradiation; ECF, elemental chlorine free; EK, eucalyptus kraft pulp; FDAM, 9H-fluoren-2-yl-diazomethane; IV, intrinsic viscosity;
HexA, hexenuronic acid; KX, kraft xylan; Mn, number average molecular weight; Mw, weight average molecular weight; MWD, molecular weight distribution; odtp,
oven-dried ton of pulp; SX, sulfite xylan; REG, reducing end group; RCF, regenerated cellulose fiber; TCF, totally chlorine-free.
* Corresponding author at: Muthgasse 18, Department Chemie, A-1190, Vienna, Austria.
E-mail addresses: (O.P. Sarosi), (D. Bammer), (E. Fitz),
(A. Potthast).
1
Werkstraße 2, A-4860, Lenzing.
/>Received 8 February 2021; Received in revised form 15 March 2021; Accepted 1 April 2021
Available online 7 April 2021
0144-8617/© 2021 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license ( />

O.P. Sarosi et al.

Carbohydrate Polymers 265 (2021) 118037

Nakatsubo, 1998; Ni, Kang, & van Heiningen, 1996; Tripathi, Bhardwaj,
& Ghatak, 2018). Initially, the ozone treatment shows a higher reaction
rate towards residual lignin and after its depletion, cellulose degradation
is promoted, offering dose-dependent IV control (Kang, Zhang, Ni, & van
Heiningen, 1995; Lachenal, Mishra, & Chirat, 2013). However, an
increased rate of cellulose degradation is accompanied by a similar rate
of undesirable carbonyl group formation due to the low selectivity of
ozone and its decomposition products (Pouyet, Chirat, Potthast, &
Lachenal, 2014; Zhang, Ni, & van Heiningen, 2000). Since ozone has
such a rapid reaction rate, its diffusion into the pulp and thus cellulose
degradation is limited to amorphous and crystalline surface regions,
leading to broadening of the MWD and, in extreme cases, to a bimodal
distribution of cellulose due to the unequal treatment of different celư
ăholm, & Lindstro
ăm, 2001).
lulose fractions (Berggren, Berthold, Sjo
In contrast, γ-irradiation and EBI homogeneously penetrate the pulp
on a macroscopic and microscopic level, leading to ionization in both
amorphous and crystalline regions (Hammer, Christensen, Conroy, King,
& Pogue, 2011; Yang, Zhang, Wei, Shao, & Hu, 2010). EBI causes radical
formation throughout the cellulose molecule, which leads to cleavage of
the glycosidic bond at statistically distributed positions along the cel­
lulose chain (Burkart, 1999). Not only can the IV be adjusted to a
desirable level through a dose-effect relationship, but also the dispersity
(Ð) is reduced (Sarosi, Bischof, & Potthast, 2020; Sixta, Andrea, & Kraft,
2007). Additionally, EBI treatments of cellulosic pulps have a high yield
while avoiding degradation to the monomer level and a statistical
preference for longer cellulose chains, retaining high contents of α-cel­

lulose at irradiation intensities of 20 kGy or below (Sarosi et al., 2020;
Yang et al., 2010). EBI has been reported to have oxidative effects,
leading to the formation of carbonyl groups on the cellulose backbone at
elevated irradiation levels of 50 kGy and above (Henniges, Hasani,
Potthast, Westman, & Rosenau, 2013). However, by differentiation be­
tween newly formed reducing end groups (REGs) and keto groups from
ăhrling, Potthast, Rosenau, Lange,
oxidation by the CCOA method (Ro
Ebner et al., 2002), it has been shown that at low irradiation levels of 20
kGy or below, the oxidation by EBI is lower than the detection limit,
refuting its long believed highly-oxidative nature (Henniges et al.,
2013). In a previous study, the protective effect of residual pulp com­
ponents, such as hemicellulose and lignin, towards chain scission and
oxidation were investigated (Sarosi et al., 2020). Hemicellulose was
found to protect cellulose from chain scission and backbone oxidation,
which was attributed to its sacrificial function as a protective barrier,
shielding cellulose from radicals from the surrounding water. While for
lignosulfonates the antioxidant properties of residual lignin had a
beneficial effect, lignin in kraft pulps caused backbone oxidation.
Another important species to observe is HexA. It is formed from
4-O-methylglucuronic acid during alkaline kraft pulping and causes
additional consumption of bleaching chemicals and brightness insta­
bility when not removed (Jiang, Lierop, & Robson, 2000; Rosenau et al.,
2017). HexAs are hydrolyzed during hot acid and ozone stages, and
while their role in pulping and bleaching has been sufficiently studied
(Antes & Joutsimo, 2015; Brogdon, 2009; Gomes, Longue, Colodette, &
Ribeiro, 2014), the impact of ionizing irradiation on HexAs has only
rarely been investigated (Tsuji-Katsukawa, Miyawaki, & Koyanagi,
2012).
The goal of this study was to highlight benefits and drawbacks of IV

control by EBI when using it as an integral part of TCF bleaching se­
quences as a full or partial replacement of ozone. Experimental focus
was put on the controlled reduction of IV to provide a chemical-free and
more homogeneous and more flexible alternative to ozone for cellulose
depolymerization in paper pulp upgrade applications. The final pulp
properties of a eucalyptus (Eucalyptus globulus) kraft paper and a beech
(Fagus sylvatica) sulfite dissolving pulp were compared after using EBI
and/or ozone for IV adjustment. Further experiments were performed to
elucidate the impact of EBI on isolated xylan and HexAs. Using EBI as a
partial or full replacement of ozone may facilitate IV control especially
in high-Mw kraft pulps, which allows for process debottlenecking and

chemical savings during bleaching.
2. Experimental
2.1. Pulps
Two different pulps were used for this study. Both oxygen-bleached
eucalyptus (Eucalyptus globulus) kraft paper pulp (EK) and oxygenbleached beech (Fagus sylvatica) sulfite dissolving pulp (BS) were
generously contributed from an industry partner as research samples.
Both pulps were never-dried and lab-washed. In between use, all pulps
were stored at − 20 ◦ C. After each bleaching stage, the IV was measured
to assure correct progression. HexA-rich birch kraft pulp was generously
contributed by the University of Natural Resources and Life Sciences
(Vienna, Austria).
2.2. Xylan
Beech sulfite xylan was isolated from steeping lye by acidic precip­
itation. Eucalyptus kraft xylan was prepared by cold caustic extraction
and acidic precipitation of a fully bleached paper pulp. 300 g of air-dried
pulp were subjected to CCE by swirling at 2.5 % consistency for 30 min
at 25 ◦ C, using 10 % (w/v) NaOH lye. The dissolved xylan was separated
by filtration and regenerated by acidification to pH 2.5 using 20 %

H2SO4. The sediment xylan was purified by 4 repeated cycles of
centrifugation and stirring in fresh softened water. Finally, the xylan was
dried at 60 ◦ C over-night, which resulted in a yield of 22 g xylan.
2.3. Bleaching stages
2.3.1. Irradiation
EBI was conducted at room temperature as described in a previous
study (Sarosi et al., 2020). Irradiation was performed at Mediscan GmbH
(Kremsmünster, Austria) using a Rhodotron TT100-IBA-X electron
accelerator according to EN ISO 13485 and ISO 11137. Pulps were
prepared for irradiation by forming pulp sheets with a thickness of 2− 3
cm, a diameter of 20 cm and a moisture content of 72–75 %. The pulp
sheets were attached to cardboard panels and positioned in vertical
irradiation trays. To guarantee homogeneous irradiation, total pulp
thickness was kept below 4 cm. The irradiation dose was set to (and
verified by a dosimeter as) either 1.25 kGy (1.30 kGy) or 2.5 kGy (2.52
kGy) for the beech sulfite pulp and 5.0 kGy (5.2 kGy) or 10.0 kGy (10.3
kGy) for the EK pulp. Irradiated pulps were swirled in hot water at 5 %
consistency for 5 min and then filtered. For irradiation of HexA-rich
birch kraft pulp and xylan, ~1.5 g of dry sample was filled into micro­
reaction tubes (PP) and attached to cardboard sheets before passing
through the electron beam. Additional irradiation levels were set to (and
verified as) 50.0 kGy (50.2 kGy), 100 kGy (101.3 kGy) or 200 kGy
(202.0 kGy). Doses above 50 kGy were applied by multiple passes at
reduced doses on alternating sides.
2.3.2. A-stage
EK pulps received a sulfuric acid stage for metal removal before
ozone or peroxide treatment. For hot acid treatments, preheated pulp
and softened water were mixed in plastic bottles (PP) at 3.0 % consis­
tency and the pH was slowly adjusted to 2.5 using sulfuric acid (100 g
L− 1). The bottles were sealed, mixed thoroughly by vigorous shaking

and incubated for 30 min at 60 ◦ C (A) or 90 ◦ C (A-hot), depending on
whether or not an IV reduction, respectively, was desired, with repeated
shaking after 15 min. The reaction was terminated by vacuum filtration
of the bottle contents over a quartz frit and washing of the pulp 4–6
times with double the pulp volume of hot, softened water and vacuum
suction between washes.
2.3.3. Z-Stage
Ozone bleaching was conducted in a medium-consistency (MC)
2


O.P. Sarosi et al.

Carbohydrate Polymers 265 (2021) 118037

mixer with closed lid and thorough pulp fluidization. The treatment and
subsequent pulp workup were performed by trained lab workers. The
treatment conditions were the following: 4.0–16.9 kg odtp− 1 (kilogram
per oven-dried ton of pulp) of ozone charge, depending on the type of
pulp and treatment route, pH 2.5, 45 ◦ C, 10 % consistency. The total
ozone charge for each pulp was split into smaller partial charges and IV
was measured in-between charges while the pulp remained in the
reactor. Once the IV was adjusted to a satisfactory level, the pulp was
filtered and washed 4–6 times with double the pulp volume of hot,
softened water. Actual ozone consumption was determined by streaming
the gaseous phase before and after bleaching through a potassium iodide
solution and titrating against sodium thiosulfate using starch as
indicator.

GPC data according to:

REG =

1
× 106
Mn

(1)

where Mn is the number average molecular weight.
HexA analysis relied on the available TAPPI standard using spec­
troscopic quantification after hydrolysis (Chai, Zhu, & Li, 2001; TAPPI,
2007).
3. Results & discussion
3.1. Bleaching sequence conduct and observations
The goal of the adjustment was to achieve IV between 450–500 mL
g− 1 (degree of polymerization (DP) 1045–1205), which is suitable for
RCF processes. To investigate the differences between EBI and ozone
during a full bleaching sequence, three approaches were chosen for each
pulp. Table 2 summarizes the different sequences for each pulp. Oxygendelignified pulps were used as the starting point for all experiments. The
first variant used EBI as a full replacement of ozone. Electron doses were
calculated based on dose-effect fit functions from previous experiments
with similar pulps (Sarosi et al., 2020). The second approach used both
EBI and ozone in lower doses to represent the integration of EBI in an
existing bleaching sequence and to highlight potential chemical savings
and quality improvements. Finally, the third sequence applied a con­
ventional sequence based on ozone as reference. Xylan removal from the
paper pulps can be performed by cold caustic extraction and enzymatic
treatments after TCF bleaching, but was omitted in this study to avoid
the interference of effects such as additional IV reduction, changes in Ð,
or alteration of the chemical composition (Duan, Verma, Li, Ma, & Ni,

2016; Hutterer et al., 2016).
Fig. 1 shows the IV decrease caused by each stage. EBI of the kraft
pulp delivered values within the calculated target range. A dose of 10
kGy halved the IV, while the 5 kGy treatment achieved 87 % of that IV
reduction. This is in accordance with previous studies, since EBI is
known to randomly cleave the cellulose chain, which is expressed by a
linear increase of the number of chain scissions and an exponential
decrease of IV and weight average molecular weight (Mw) (Chen, Ma, Li,
Miao, & Huang, 2017; Henniges, Okubayashi, Rosenau, & Potthast,
2012). Since BS had a lower initial IV, a smaller irradiation dose was
applied. However, the sulfite pulp showed a similar IV decrease by both
1.25 and 2.5 kGy, despite precise dose control. This may be ascribed to
the high sensitivity of Mw at the initial, low dose range.
In some cases, the IV was slightly increased after hot acid (A) or the
first ozone (Z) stage. One explanation may lie in the hydrolysis and
removal of short chains, thus increasing the median chain length.
However, the sugar outflow (Fig. 2, Table 3) did not indicate significant
carbohydrate release in those stages and it remains unclear where this IV
increase originates from. The ozone stages were conducted by applying
stepwise charges with IV measurements in between, which allowed for
precise control. When comparing the IV reduction capacity of ozone
treatments expressed as the quotient of %IV reduction and ozone dose
(data not shown), the presented results fall in line with numbers that
other researchers have found during the ozone treatment of comparable
pulps (He, Liua, & Tian, 2018; Pouyet et al., 2014; Tripathi et al., 2018).
By incremental application of ozone charges it is shown that
dose-normalized %IV reduction of BS pulps is larger after the first
charge, while EK pulps receive an equal reduction intensity with each
charge. This may indicate that BS pulp has approximated a kinetic
bottleneck after the first ozone stage. In contrast, the high residual lignin

and xylan content in EK partially protect cellulose from degradation,
which inhibits %IV decrease, resulting in a lower rate across all ozone
stages. Additionally, data from the literature suggests that fiber acces­
sibility and pulp reactivity of a paper pulp is typically much lower than
that of a dissolving pulp, limiting ozone diffusion in the former (He

2.3.4. P-stage
Every sequence was terminated by a strong hydrogen peroxide stage
with constant charges to both increase the DoB and stabilize the IV by
β-elimination of carbonyl “hot-spots” on the cellulose backbone (Yang,
2016). Hydrogen peroxide bleaching was conducted in plastic bottles
(PP). The treatment conditions were the following: 4 kg odtp− 1
hydrogen peroxide, 6 kg odtp− 1 sodium hydroxide, 10 % consistency, 90
◦
C, 180 min with thorough shaking every 30 min. Both pulp and soft­
ened water were pre-heated and the calibrated amounts of hydrogen
peroxide (~35 %, exact content was measured before use) and sodium
hydroxide stock solution (50 g L− 1) were added to the water just before
mixing. The bottle was sealed, shaken vigorously and incubated at 90 ◦ C
for 180 min with additionally shaking every 30 min. The reaction was
terminated by vacuum filtration of the bottle contents over a quartz frit
and washing of the pulp between 4–6 times with double the pulp volume
of hot, softened water. Contact with atmospheric oxygen at high alka­
linity was minimized by working quickly and keeping the pulp covered
during workup. Residual hydrogen peroxide and alkalinity were deter­
mined by titration as described above or against 0.1 M HCl, respectively.
In cases where the peroxide was not completely consumed due to low
pH, the bleaching stage was repeated, employing the calculated residual
hydrogen peroxide charge. After hydrogen peroxide treatment, the
bleaching sequence was terminated by swirling the pulp at 5 % consis­

tency for 10 min in hot water and adjusting the pH to below 3.0 using
SO2-infused water with subsequent washing.
2.4. Pulp analysis
Kappa number is measured according to TAPPI T236cm-85 where
pulp chromophores are reacted with a fixed amount of KMnO4 solution
and its consumption is evaluated by neutralization with KI followed by
titration against Na2S2O3. Pulp, brightness, is analyzed according to ISO
2470-1:2009 where the reflectiveness of uniform pulp sheets is
measured optically and IV was were measured according to TAPPI
T236cm-85, ISO 2470-1:2009 and ISO 5351 5351, respectively, which
uses rheological data from pulp solutions in cupri-ethylenediamine and
a flow-through viscometer. Alkali-resistant fractions R10 and R18 were
determined according to DIN 54355, which analyzes the pulp solubility
in 10 % or 18 % NaOH solution, respectively. Cellulose crystallinity was
measured by FT-Raman spectroscopy with reference data from X-ray
ăder et al., 2006).
wide angle scattering (Ro
The sugar composition of pulps and aqueous samples was analyzed
by total hydrolysis and HPLC as previously described (Sarosi et al.,
2020).
Measuring the MWD was coupled to fluorescence labeling of
carbonyl or carboxyl groups, as described in the CCOA or FDAM method,
ăhrling, Potthast, Rosenau, Lange,
respectively (Bohrn et al., 2006; Ro
Borgards et al., 2002). Instrumentation, settings and data evaluation was
performed as previously described (Sarosi et al., 2020). For fluorescence
detection of FDAM-labelled samples, a different fluorescence detector,
RF 535 (Shimadzu, Japan), was used at λex 280 nm and λem 312 nm.
As a rough estimation, the number of REGs can be calculated from
3

O.P. Sarosi et al.

Carbohydrate Polymers 265 (2021) 118037

Table 1
General properties of the investigated hardwood pulps before treatment.
Pulp

Mw (GPC-MALS)
(kg mol− 1)

IV (mL g− 1)

Kappa number

Brightness (%)

R10; R18 (%)

Hemicellulose content (%)

Oxygen-bleached eucalyptus kraft pulp (EK)
Oxygen-bleached beech sulfite pulp (BS)

466
336

943

643

8.6
1.9

63.5
77.5

89.3; 91.6
89.1; 93.8

19.0
3.4

and highest in EK pulps, with values of 0.8 % for EK-EBI and EKI-mixed
and 1.8 % for EK-ozone.

Table 2
Pulp treatment sequences and electron and ozone dose. *The first number is the
measured actual dose that was required and the second number in parentheses is
the initially calculated dose requirement.
Pulp

Bleaching
sequence

Electron dose
(kGy)*

Ozone charge (kg

odtp− 1)

Name

EK
EK
EK
BS
BS
BS

OO-EBI-A-P
OO-EBI-A-Z-P
OO-A(hot)-Z-P
EO-EBI-P
EO-EBI-Z-P
EO-Z-P

10.3 (10.0)
5.2 (5.0)
–
2.52 (2.5)
1.3 (1.25)
–

–
6.4
16.9
–
4.4

4.0

EK-EBI
EK-mixed
EK-ozone
BS-EBI
BS-mixed
BS-ozone

3.2. Comparison of final pulp properties
3.2.1. IV, MWD and carbonyl group distribution
After completing each bleaching sequence, the pulps were analyzed
by the CCOA method (Table 4, Fig. 3). The underlying GPC data shows,
that all pulps have an Mw of 227–295 kg mol− 1, which translates to
calculated IVs of 561–685 mL g− 1. This deviates significantly from the
measurements by the cupri-ethylenediamine (CUEN) method, which
showed similar ratios between the pulps, albeit 150–200 mL g− 1 lower
numbers due to β-elimination in the CUEN solution. In both EK and BS
pulps, dispersity is significantly higher when ozone was used for IV
control. This effect is aggravated in the EK pulps, where a stronger IV
reduction was required. This is a result of the diffusion behavior and
high reactivity of ozone, which limits polysaccharide degradation to
short-chain amorphous and outer crystalline regions, leaving long-chain
core regions mostly intact. Thus, EK-ozone shows the strongest peak
broadening as it required the highest dose of ozone. Additionally, it is
the only pulp that showed slight degradation to the monomer level
during the Z-stage, caused by the prolonged dwell times in the acidic
reactor as a consequence of the step-wise process conduct (Fig. 2).
EBI treatments led to a statistically distributed increase of carbonyl
groups, raising the carbonyl profiles by both backbone oxidation and

new REG. This is most visible in, but not limited to, the low-Mw region.
While carboxy group formation by EBI has been reported for amorphous
cotton materials, lignocellulosic pulps barely undergo carboxyl group
formation at low irradiation doses, especially in hardwood pulps in
which a high hemicellulose content is present (Bouchard, M´
ethot, &
Jordan, 2006). This is caused by the limited diffusion of molecular ox­
ygen inside the material. Hence, EBI in the present study delivers
carbonyl groups as the major oxidation product at the given irradiation
levels.
In analogy to the IV reduction, oxidation is heterogeneous for
chemical treatments (Potthast, Rosenau, & Kosma, 2006). The carbonyl
profiles (Fig. 3, A, B) of both EK and BS reference ‘ozone’ pulps display
the highest carbonyl group content in the low-Mw region. This is due to

O = oxygen delignification; EBI = electron beam irradiation stage; A = hot
sulfuric acid stage; A(hot) = A-stage with increased temperature; E = alkaline
extraction stage; Z = ozone stage; P = hydrogen peroxide stage.

ăpcke, Ibarra, & Ek, 2008; Miao et al., 2015). Finally, the
et al., 2018; Ko
peroxide stage (P) required an increased proportion of alkali to
compensate for β-elimination reactions due to the higher number of keto
groups from EBI and ozone stages. The final IVs of EK-mixed, EK-ozone
and BS-EBI were within the 450–500 mL g− 1 target, while EK-EBI,
BS-mixed, and BS-ozone delivered lower values, which allows the EBI
or ozone doses to be lowered in those cases (Fig. 1).
Carbohydrate release was monitored by measuring the sugar
composition of crude bleaching filtrates, both before and after total
hydrolysis, to differentiate monomeric and polymeric fragments,

respectively (Fig. 2). EBI is known to release predominately oligomeric
and polymeric hemicelluloses from the pulp (Sarosi et al., 2020). The
extent of hemicellulose release is roughly correlated to the pulps’
hemicellulose content and to the irradiation dose. Acid and ozone stages
released minimal amounts of polymeric cellulose and hemicellulose
with the exception of EK-ozone (Fig. 2, E), which shows cellulose
degradation to the monomer level, likely due to its highest ozone dose
and longest dwell time in the reactor at acidic conditions. During
hydrogen peroxide bleaching, the alkaline conditions facilitate removal
of low-molecular fragments, which mostly consist of hemicellulose.
Overall yield loss during bleaching, calculated as total sugar outflow
from the pulp, was lowest in BS pulps, with values slightly above 0.1 %,

Fig. 1. IV of eucalyptus kraft (A) and beech sulfite (B) pulps after EBI, hot acid, incremental ozone, or hydrogen peroxide stages, respectively. Error bars correspond
to 1.0 % inherent standard error of the method.
4


O.P. Sarosi et al.

Carbohydrate Polymers 265 (2021) 118037

Fig. 2. Progressive sugar release from all bleaching stages of eucalyptus kraft and beech sulfite pulps. Bleaching filtrates were analyzed before and after total
hydrolysis, to differentiate between monomeric and polymeric carbohydrates, respectively. A = EK-EBI; B = BS-EBI; C = EK-mixed; D = BS-mixed; E = EK-ozone; F =
BS-ozone.

the overwhelming generation of REG in low-Mw fragments, which is also
reflected in a larger fraction of DP < 200 species. Oxidation of REG in
the low-Mw region to carboxylic acid and lactone species, which are not
covered by CCOA labelling, was secondary, as the carbonyl profiles of

“ozone” variants indicated more carbonyl groups than “EBI” and
“mixed” in that area.
The IV reduction caused by the P-stage is greater in EBI variants due

to the facilitated backbone carbonyl formation in the high-Mw region.
Chain cleavage through β-elimination was shown to have a greater effect
on IV if the keto group is located on a long chain than on a short chain.
In ΔDSCO (degree of carbonyl substitution including REG) plots
(Fig. 3, C, D) the respective CCOA signals of pulps from ‘EBI’ and ‘mixed’
treatment routes were subtracted from the signal of the conventional
“ozone” variants to compare the oxidation profiles caused by each
5


O.P. Sarosi et al.

Carbohydrate Polymers 265 (2021) 118037

Table 3
Sugar release data of each individual bleaching stage of eucalyptus kraft and beech sulfite pulps. *Measured values were lower due to formation of hydrox­
ymethylfurfural during total hydrolysis.
Pulps
EK-EBI

EK-mixed

EK-ozone
BS-EBI
BS-mixed
BS-ozone

Stage

Cellulose monomeric (mg odgp− 1)

Cellulose polymeric (mg odgp− 1)

Hemicellulose monomeric (mg odgp− 1)

Hemicellulose polymeric (mg odgp− 1)

EBI
A
P
EBI
A
Z
P
Ahot
Z
P
EBI
P
EBI
Z
P
Z
P

0

0
0
0
0
0
59

64
0
0
126
0
0
116

35
0
8
0
0
58
34

5606
467
1987
3151
522
498
3838

0

0

0

326

7879
163
0
10
0
0
122
5
109

7879*
298
21
98
15
24
220
26
221

207

100
0
94
0
6
15
6
21

855
8535
519
519
544
77
690
114
800

Table 4
CCOA data of all final pulps.
Pulps

Mn
(kg mol− 1)

Mw (kg mol
− 1
)

Mz (kg
mol− 1)

DP<
100
(%)

DP<
200
(%)

DP<
2000
(%)

DP>
2000
(%)

Ð

IV (mL g− 1)

REG (μmol g− 1)

C = O (μmol g− 1)

EK-EBI
EK-mixed
EK-ozone

BS-EBI
BS-mixed
BS-ozone

58
59
45
50
37
36

227
246
254
295
259
245

534
639
850
753
838
738

6
6
10
8
10

11

11
10
11
6
7
7

61
61
57
56
59
59

22
24
22
30
24
23

3.9
4.2
5.6
5.9
7.1
6.7

561
595
611
685
620
595

17.2
17.0
22.2
19.9
27.3
27.5

11.7
10.9
17.5
20.1
21.1
28.4

REG = reducing end groups; IV = intrinsic viscosity; DP = degree of polymerization; Mn, Mw or Mz = number, weight or z-average molecular mass.

approach. As indicated by the absolute carbonyl group contents
(Table 4), progressive replacement of ozone by EBI for IV control lead to
a lower carbonyl group profile throughout the MWD. The low-Mw region
showed the strongest differences with ozone-treated pulps featuring
more REG due to aforementioned diffusion limitations. Only BS-EBI
shows a moderately higher carbonyl group content than BS-ozone
around a log(Mw) of ~3.8. In bleaching sequences with combined

treatments of ozone and hydrogen peroxide, a greater number of
carbonyl groups disappears through oxidation of REGs than in sequences
with lower or without ozone stages. Yet, the higher number of total
carbonyl groups of reference “ozone” pulps permits the conclusion that
“mixed” and “EBI” variants exhibit a lower amount of true cellulose
backbone oxidation and thus better brightness stability (Ahn et al.,
2019).

resistant with R10 and R18 fractions on average 1.7 % and 0.4 % higher,
respectively, than the other two BS variants. Due to the sparse avail­
ability of data in the literature on alkali-resistance of pulps after EBI, one
must consider other means of determining alkali-resistance, such as
measuring the α-cellulose content (Ritter, 1929). Other researchers have
found an exponential decrease in the α-cellulose content of sugarcane
bagasse of up to 80 % by EBI with a dose of up to 40 kGy (Ribeiro,
Oikawa, Mori, Napolitano, & Duarte, 2013). Such behavior was not
indicated in the present results. Another study suggested that γ-irradi­
ation in doses of up to 10 kGy had no effect on the α-cellulose content of
bamboo paper kraft pulp, with minor reductions at 30 kGy and strong
reductions beyond that irradiation level (Yang et al., 2010). Generally,
an increasing level of irradiation causes polymeric chain length to fall
below the dissolution limit in 10 % or 18 % NaOH, respectively. How­
ever, at the given irradiation levels in both EK and BS pulps, the bulk of
holocellulosic chain length remained above the dissolution limit in both
EK and BS pulps at the given irradiation levels.

3.2.2. Alkali-resistance R10 and R18
Pulp alkali-resistance in 10 % (R10) or 18 % (R18) NaOH showed
distinct differences between each treatment variant (see supplementary
information). Cellulose chain cleavage inevitably leads to a decrease in

the R10 and R18 fractions. However, since EBI statistically favors longer
chains and ozone is diffusion-limited to outer areas for the fiber, the
alkali-resistance of the “EBI” variant is less compromised than in the
“ozone” pulp. This is indicated by the MWD (Fig. 3, A, B), where pulps
with EBI displayed lower Ð and less shifts towards dissolution limits in
10 % and 18 % NaOH, respectively. However, EK-EBI and EK-ozone
display equal R10 and R18 fractions with EK-mixed having high R10
and R18 fractions with on average 2.2 % and 1.3 % higher values,
respectively, than the other two. BS-mixed had similar R10 and R18
values to BS-ozone due to their equal ozone stages and the low irradi­
ation dose of the former. Hence, the behavior of those pulps during
alkalization and viscose making may be the same. BS-EBI is more alkali-

3.2.3. Pulp sugar composition
The sugar composition of the final pulps (Fig. 4, Table 5) was very
similar within each group of EK or BS variants. This is reasonable, since
the yield loss of all treatment sequences was below 1.8 %. Marginal
differences were observed in the EK series, where xylan content
decreased from 17.1 % in EK-ozone over 17.0 % in EK-mixed to 16.7 %
in EK-EBI, but glucose recovery decreased in the same order from 80.3 %
over 79.9 % to 78.9 %, respectively. These small differences can be
ascribed to side reactions such as formation of furfural and HMF during
sample preparation. Compared to the respective starting material, EK
pulps lost between 1.8–2.2 % of hemicellulose content, while BS did not
a show significant reduction (Table 1). The pulp crystallinity index of all
pulps was 54.7–55.9 % and 54.0–54.5 % for EK and BS pulps,
6


O.P. Sarosi et al.

Carbohydrate Polymers 265 (2021) 118037

Fig. 3. Molecular weight distribution and carbonyl group profiles of eucalyptus kraft (A) and beech sulfite (B) pulps from variable treatments and their respective
differential carbonyl group profiles. Vertical lines R10 and R18 represent the lower limit of alkali resistance in 10 % or 18 % NaOH solution, respectively. ΔDS plots
are generated by subtracting the CCOA signal of ‘EBI’ or ‘mixed’ from the reference ‘ozone’ pulp signal, which shows the divergence of carbonyl group profiles within
EK (C) and BS (D) pulps, respectively.

respectively, with differences between the treatment-variants well
below the relative standard deviation. If pulp crystallinity can be
assumed as one indicator for pulp reactivity, differences of the latter
between the pulps may be minimal (Ferreira, Evtuguin, & Prates, 2020).
While reactivity measurements according to Fock or Treiber were not
possible in the current study, other researchers have provided detailed
insight. A study by Gondhalekar et al. found that crystallinity of a
hardwood dissolving pulp was decreased by up to 9% and reactivity was
increased by 11 % at a dose of 5 kGy (Gondhalekar, Pawar, & Dhumal,
2019). Based on this data, reactivity changes by EBI can be assumed to
be equal to those by ozone treatments.
3.2.4. Degree of bleaching (DoB)
The pulps’ DoB was quantified by kappa number and ISO brightness
measurements (Fig. 5). Out of the EK pulps, only EK-ozone with the
conventional sequence reached brightness values that meet dissolving
pulp specifications. BS-mixed and BS-ozone showed similar and high
DoB due to their equal Z-stage. EK-EBI, EK-mixed and BS-EBI showed
DoB below dissolving pulp specifications due to a lack of sufficient
chromophore removal or destruction. EBI at the employed doses does
not introduce enough disruptions in the extended π-electron system of
lignin and other chromophores to have an effect on pulp brightness
(Sarosi et al., 2020). In contrast with cellulose, lignin shows good radical

stabilization and is resistant against lower irradiation doses and thus
remains mostly intact in the pulp after irradiation (Dizhbite, Telysheva,
Jurkjane, & Viesturs, 2004; Faustino, Gil, Cecília, & Duarte, 2010).
While EBI is known to form reactive oxygen species in atmospheric
oxygen and water, such as ozone, peroxyl and hydroxyl radicals, the
amount of generated bleaching agents in moist pulps is too low to have a

Fig. 4. Total hydrolysis sugar composition of final eucalyptus kraft and beech
sulfite pulps with variable bleaching sequences.
Table 5
Data of total hydrolysis sugar composition of final eucalyptus kraft and beech
sulfite pulps.
Pulps

Glucose (%)

Xylose (%)

Mannose (%)

Galactose (%)

EK-EBI
EK-mixed
EK-ozone
BS-EBI
BS-mixed
BS-ozone

78.9

79.9
80.3
94.8
94.7
94.3

16.7
17.0
17.1
2.7
2.7
2.7

0.1
0.1
0.1
0.8
0.7
0.7

0.2
0.2
0.2
0.0
0.0
0.0

7

O.P. Sarosi et al.

Carbohydrate Polymers 265 (2021) 118037

REG in SX are not labeled by CCOA, and the calculated number of REG is
more than double the total number of measured carbonyl groups,
regardless if Mn or Mw is used for calculation. Overall, isolated xylan is
more resistant to degradation by EBI than cellulose pulp. However, as
shown by the sugar analysis of wash filtrates after irradiation (Fig. 2),
xylan is primarily leached from the pulps. This may indicate that xylan
Mw decreases barely below the dissolution limit after EBI, or that the
accessibility of xylan fibers otherwise recalcitrant to dissolution is
improved by EBI.
3.4. The influence of EBI on HexA
The kappa number of EK pulps may be partially impacted by HexA,
which is attached to the xylan side chains in kraft pulps (Jiang et al.,
2000; Vuorinen, Fagerstră
om, Buchert, Tenkanen, & Teleman, 1999). As
shown before, EK contains considerable amounts of xylan. This gave rise
to a small series of experiments in which a HexA-rich birch kraft pulp
was irradiated. The birch pulp was fully bleached, yet had a distinct
yellow hue caused by the high HexA content. The HexA content was
traced using three methods: regular kappa measurements, uronic acid
groups by the FDAM method, and HexA quantification based on the
available TAPPI standard (Bohrn et al., 2006; TAPPI, 2007).
Fig. 7 shows that EBI causes a linear increase of uronic acid groups at
elevated irradiation doses, which is analogous to the carbonyl group
formation found in other studies (Bouchard et al., 2006; Henniges et al.,
2012). As has been shown, kappa numbers decrease with increasing
irradiation doses (Sarosi et al., 2020). While the previous data showed a

marginal kappa increase at elevated irradiation doses due to carbonyl
group formation, it was not observed here.
Similar to other parameters like cellulose Mw or kappa number, the
HexA content was reduced rapidly at low irradiation doses, with a leveloff effect at elevated doses. In the dose range of 5–10 kGy, relevant for
paper pulp upgrade, the HexA content was reduced by 10–26 %, while
the highest reduction of 49 % was observed at a dose of 50 kGy. HexA
removal by EBI gives rise to improved pulp brightness stability, since
HexA is known as a precursor for multiple chromophores (Rosenau
et al., 2017). HexA content reduction is caused either by chemical
modification or by removal of hemicelluloses. Hemicellulose degrada­
tion by EBI to soluble fragments is observed, although the degree of
hemicellulose removal can only account for a small part of the decrease
in HexA content (Chen et al., 2016; Ribeiro et al., 2013). Interestingly,
HexA content saw a slight increase at 200 kGy indicating the formation
of double bonds, which are sensitive for analysis, at elevated irradiation
doses.
Since HexAs react with potassium permanganate during kappa
determination, they are considered “false lignin,” the extent of which
can be calculated (Vuorinen et al., 1999). To determine that fraction,
both variables, HexA content and Kappa number, were transformed into
permanganate molar equivalents. The KMnO4 consumption was directly
taken from kappa measurements and a 1:1 ratio was used for calculating
KMnO4 equivalents from the HexA concentration (Table 7) (Chai et al.,
2001). Results indicate that up to one third of the kappa number is
caused by HexAs. The “false lignin” fraction decreases with irradiation,
which indicates that HexAs are more susceptible to degradation by EBI
than lignin, which is reasonable considering the structure and radical
scavenging activity of the latter (Dizhbite et al., 2004).

Fig. 5. Kappa number and brightness of eucalyptus kraft and beech sulfite

pulps after their respective treatments. Vertical lines represent the upper kappa
limit of 1.0 and lower brightness limit of 90 % that are recognized as typical for
dissolving pulps. Error bars indicate the inherent standard error of the kappa
(5%) and brightness (0.2 %) measurement method, respectively.

great effect on kappa number or brightness at the given dose (Cleland &
Galloway, 2015; Gehringer, 1997; Sarosi et al., 2020). Since the “mixed”
variant of each pulp gave a significantly higher DoB compared to the
“EBI” route, especially in the case of EK pulp, a hybrid use of EBI and
ozone in paper pulp upgrade is at least plausible, if the hemicellulose
fraction is removed and the bleaching intensity is adjusted accordingly.
3.3. The influence of EBI on isolated xylan
EBI of isolated hemicellulose has rarely been investigated in previous
studies (Chen et al., 2016; Ma et al., 2014). Isolating xylan before irra­
diation may reveal effects that are otherwise overshadowed by the
presence of cellulose. Hence, both isolated eucalyptus kraft xylan (KX)
and beech sulfite xylan (SX) were irradiated and analyzed by the CCOA
method. The Mw reduction of xylan was less pronounced than for
cellulosic components since the fragment size was already low (Table 6).
Non-irradiated KX had a relatively high Mw, which was reduced to
almost level-off Mw after a dose of just 10 kGy, while SX was barely
affected by EBI (Table 6). Other researchers have observed a similar
resistance of the Mw of hemicellulose towards irradiation (Ma et al.,
2014). The MWD of KX showed peak broadening at high irradiation
levels, which may indicate the simultaneous occurrence of chain
cleavage and cross-linking, with the former being more pronounced
(Fig. 6). Carbonyl group content increased moderately for SX and
strongly for KX at high irradiation doses. In the latter case, a significant
part of the carbonyl groups originate from newly formed REG. The
majority of REG groups of SX are oxidized since it originates from an

acidic magnesium bisulfite process. Therefore, a significant portion of
Table 6
Weight average molecular mass, carbonyl group content and calculated REG of
eucalyptus kraft and beech sulfite xylan after irradiation at varying doses.

Mw (kg mol− 1)
Carbonyl group content
(μmol g− 1)
REG calculated from Mn
(μmol g− 1)
REG calculated from Mw
(μmol g− 1)

Sample

0 kGy

10
kGy

100
kGy

200
kGy

KX
SX
KX
SX

KX
SX
KX
SX

47.6
5.5
71.6
67.8
25.8
183.2
21.0
156.0

14.4
4.9
66.1
69.5
87.2
202.4
69.3
154.6

11.9
4.4
89.3
82.7
83.8
228.8
83.8

186.9

13.0
4.6
216.1
84.7
160.5
218.8
77.2
180.5

4. Conclusion
In the present study, a eucalyptus kraft paper pulp and a beech sulfite
dissolving pulp were subjected to different bleaching sequences, con­
sisting of either an EBI, an ozone stage, or a combination of both, with
the aim of lowering the IV to levels applicable for RCF processes. Overall
yield loss of the sequences measured by carbohydrate outflow was
around 0.1 % for all beech dissolving pulps, 0.8 % for EK-EBI and EKmixed pulps and 1.8 % for EK-ozone pulp. EBI posed a tool for

REG = reducing end groups.
8


O.P. Sarosi et al.

Carbohydrate Polymers 265 (2021) 118037

Fig. 6. Molecular weight distribution and carbonyl group profiles of eucalyptus kraft (A) and beech sulfite (B) xylan after irradiation at varying dose.

displayed the highest alkali-resistance out of all tested pulps, high uni­

formity, the lowest carbonyl group content, a suitable IV, and a
bleaching degree slightly below dissolving pulp specifications. On the
other hand, acid sulfite pulping generates a pulp that already has a low
IV, making the use of EBI for IV control less important, despite similar
dispersity and oxidation advantages. EBI is a suitable treatment method
for IV reduction of high-Mw pulps, which can be used to substitute
otherwise intensive chemical treatments and release potential process
bottlenecks. However, since EBI had no significant effect on lignin at the
employed irradiation levels, pulps with EBI-only sequences suffered
from low bleaching degrees. Herein, the combined use of EBI and ozone
posed a good compromise, unifying advantages of both treatments.
Additional irradiation experiments on HexA-rich birch pulp and isolated
xylan samples revealed a reduction of the former of up to 49 % EBI by a
dose of 50 kGy. While xylan sample’s Mw were barely above or within
lower level-off regions, the degree of oxidation increased moderately for
beech sulfite xylan and strongly for eucalyptus kraft xylan. Overall,
changes imparted in xylan by EBI are weaker than in cellulose pulps at
equal irradiation doses, especially if the xylan Mw is already low.

Fig. 7. Birch pulp HexA content, uronic acid content, and kappa values after
EBI with varying dose.

CRediT authorship contribution statement
Oliver P. Sarosi: Conceptualization, Investigation, Data analysis,
Writing - original draft. Daniela Bammer: Investigation. Elisabeth
Fitz: Writing - review & editing, Supervision. Antje Potthast: Concep­
tualization, Data analysis, Manuscript review & editing, Supervision.

Table 7
Permanganate equivalent concentrations of irradiated birch pulp calculated

from Kappa number or HexA measurements and the fraction of “false lignin”
caused by HexAs. *The 1.25 kGy sample was determined as an outlier in both
kappa and HexA measurements due to irregularities during measurements.
Irradiated birch
pulp sample

KMnO4 equivalent
“Kappa” (μmol
g− 1)

KMnO4 equivalent
“HexA” (μmol g− 1)

Fraction of “false
lignin” caused by
HexA (%)

Reference
1.25 kGy*
2.5 kGy
5.0 kGy
10.0 kGy
50.0 kGy
100 kGy
200 kGy

60.9
42.1*
65.0
64.1

57.7
53.2
51.2
50.3

21.9
18.5*
21.4
19.7
16.2
11.5
11.1
15.1

36.0
44.0*
32.8
30.8
28.0
21.5
21.7
30.0

Funding
This research was funded by the Austrian Research Promotion
Agency (FFG), Grant Number 844608. Open access was supported by
BOKU Vienna Open Access Publishing Fund.
Declaration of Competing Interest
The authors report no conflict of interests.
Acknowledgements

efficient IV reduction, giving rise to a potential use in upgrading high-IV
paper pulps to dissolving pulps. The CCOA measurements supported this
hypothesis, revealing a lower carbonyl group content and improved
carbonyl profiles when progressively replacing the ozone stage with EBI.
Additionally, EBI-treated variants had a lower dispersity of molar
masses and a smaller shift towards the low-Mw region due to the sta­
tistically high chance of cleaving long chain cellulose, which is prefer­
able for dissolving pulp. This was also expressed by higher alkaliresistant fractions R10 and R18 in EBI-treated pulps. “EK-mixed” pulp

Financial support was provided by the Austrian government and by
the provinces of Lower Austria, Upper Austria and Carinthia, as well as
by Lenzing AG. We also express our gratitude to the University of Nat­
ural Resources and Life Sciences (BOKU), Vienna and Lenzing AG for
their in-kind contributions. Furthermore, we thank the responsible
project manager at Lenzing AG Robert Bischof. Last but not least, special
thanks to all lab technicians for their patience and laboratory support,
especially Sonja Schiehser and Markus Huemer.
9


O.P. Sarosi et al.

Carbohydrate Polymers 265 (2021) 118037

Appendix A. Supplementary data

Henniges, U., Okubayashi, S., Rosenau, T., & Potthast, A. (2012). Irradiation of cellulosic
pulps: Understanding its impact on cellulose oxidation. Biomacromolecules, 13(12),
4171–4178. />Hutterer, C., Kliba, G., Punz, M., Fackler, K., & Potthast, A. (2017). Enzymatic pulp

upgrade for producing high-value cellulose out of a Kraft paper pulp. Enzyme and
Microbial Technology, 102(7), 67–73. />enzmictec.2017.03.014
Hutterer, C., Schild, G., & Potthast, A. (2016). A precise study on effects that trigger
alkaline hemicellulose extraction efficiency. Bioresource Technology, 214, 460–467.
/>Jiang, Z. H., Lierop, B., & Robson, B. (2000). Hexenuronic acid groups in pulping and
bleaching chemistry. Tappi Journal, 83(1), 167–175. />biortech.2013.01.049
Kang, G., Zhang, Y., Ni, Y., & van Heiningen, A. R. P. (1995). Influence of lignins on the
degradation of cellulose during ozone treatment. Journal of Wood Chemistry and
Technology, 15(4), 413–430. />Kishimoto, T., & Nakatsubo, F. (1998). Non-chlorine bleaching of kraft pulp. V.
Participation of radical species in ozonation of methyl 4-O-ethyl--Dglucopyranoside. Holzforschung, 52(2), 185. />hfsg.1998.52.2.185
Kă
opcke, V., Ibarra, D., & Ek, M. (2008). Increasing accessibility and reactivity of paper
grade pulp by enzymatic treatment for use as dissolving pulp. Nordic Pulp and Paper
Research Journal, 23(4), 363–368. />Kubes, G. J., Fleming, B. I., Macleod, J. M., Bolker, H. I., & Werthemann, D. P. (1983).
Viscosities of unbleached alkaline pulps. II. The G-factor. Journal of Wood Chemistry
and Technology, 3(3), 313333. />Kvarnlă
of, N., Să
oderlund, C. A., & Germgård, U. (2006). The effect of modifying the
oxidative pre-aging conditions in the manufacture of viscose from wood pulp. Paperi
Ja Puu, 88(3), 175–180.
Lachenal, D., Mishra, S., & Chirat, C. (2013). The effect of introducing ozone in elemental
chlorine free bleaching of eucalyptus kraft pulp Pulp. Tappi Journal, 12(11), 39–45.
/>Lundberg, V., Axelsson, E., Mahmoudkhani, M., & Berntsson, T. (2012). Energy analysis
for conversion of a kraft pulp mill into a dissolving pulp mill. Chemical Engineering
Transactions, 29, 13–18.
Ma, X., Zheng, X., Zhang, M., Yang, X., Chen, L., Huang, L., et al. (2014). Electron beam
irradiation of bamboo chips: Degradation of cellulose and hemicelluloses. Cellulose,
21(6), 3865–3870. />Miao, Q., Tian, C., Chen, L., Huang, L., Zheng, L., & Ni, Y. (2015). Combined mechanical
and enzymatic treatments for improving the Fock reactivity of hardwood kraft-based
dissolving pulp. Cellulose, 22(1), 803–809. />Mozdyniewicz, D., Nieminen, K., & Sixta, H. (2013). Alkaline steeping of dissolving pulp.

Part I: Cellulose degradation kinetics. Cellulose, 20, 1437–1451. />10.1007/s10570-013-9926-2
Ni, Y., Kang, G. J., & van Heiningen, A. R. P. (1996). Are hydroxyl radicals responsible
for degradation of carbohydrates during ozone bleaching of chemical pulp? Journal
of Pulp and Paper Science, 22(2), 53–57.
Potthast, A., Rosenau, T., & Kosma, P. (2006). Analysis of oxidized functionalities in
cellulose. In D. Klemm (Ed.), Polysaccharides II (pp. 1–48). Berlin, Heidelberg:
Springer Berlin Heidelberg.
Pouyet, F., Chirat, C., Potthast, A., & Lachenal, D. (2014). Formation of carbonyl groups
on cellulose during ozone treatment of pulp: Consequences for pulp bleaching.
Carbohydrate Polymers, 109, 85–91. />Ribeiro, M. A., Oikawa, H., Mori, M. N., Napolitano, C. M., & Duarte, C. L. (2013).
Degradation mechanism of polysaccharides on irradiated sugarcane bagasse.
Radiation Physics and Chemistry, 84, 115–118. />radphyschem.2012.06.034
Ritter, G. J. (1929). Determination of alpha-cellulose. Industrial & Engineering Chemistry
Analytical Edition, 1(1), 5254. />Ră
oder, T., Moosbauer, J., Fasching, M., Bohn, A., Fink, H.-P., Baldinger, T., et al. (2006).
Crystallinity determination of native cellulose - comparison of analytical methods.
Lenzinger Berichte, 86, 132136.
Ră
ohrling, J., Potthast, A., Rosenau, T., Lange, T., Borgards, A., Sixta, H., et al. (2002).
A novel method for the determination of carbonyl groups in cellulosics by
fluorescence labeling. 2. Validation and applications. Biomacromolecules, 3(7),
969975. />Ră
ohrling, J., Potthast, A., Rosenau, T., Lange, T., Ebner, G., Sixta, H., et al. (2002).
A novel method for the determination of carbonyl groups in cellulosics by
fluorescence labeling. 1. Method development. Biomacromolecules, 3(5), 959–968.
/>Rosenau, T., Potthast, A., Zwirchmayr, N. S., Hettegger, H., Plasser, F., Hosoya, T., et al.
(2017). Chromophores from hexeneuronic acids: Identification of HexA-derived
chromophores. Cellulose, 24(9), 3671–3687. />Sappi Limited. (2013). Sappi fine paper North America announces successful completion of
$170M major capital investment in North American operations Accessed Date: 19 May
2020, Retrieved from />Sarosi, O. P., Bischof, R. H., & Potthast, A. (2020). Tailoring pulp cellulose with Electron

beam irradiation: Effects of lignin and hemicellulose. ACS Sustainable Chemistry &
Engineering, 8(18), 7235–7243. />Sixta, H. (2006). Prehydrolysis. Handbook of pulp (pp. 325–345). Weinheim: Wiley-VCH.
Sixta, H., Andrea, P., & Kraft, G. (2007). Method for producing cellulose. Patent No.:
WO2007128027A1.

Supplementary material related to this article can be found, in the
online version, at doi: />References
Agarwal, N., & Gustafson, R. (1997). A contribution to the modeling of kraft pulping. The
Canadian Journal of Chemical Engineering, 75, 8–15. />cjce.5450750104
Ahn, K., Zaccaron, S., Zwirchmayr, N. S., Hettegger, H., Hofinger, A., Bacher, M., et al.
(2019). Yellowing and brightness reversion of celluloses: CO or COOH, who is the
culprit? Cellulose, 26(1), 429–1444. />Antes, R., & Joutsimo, O. P. (2015). Effect of modified cooking on bleachability of
Eucalyptus globulus and Eucalyptus nitens. Bioresources, 10(1), 597–612. https://
doi.org/10.15376/biores.10.1.597-612
Berggren, R., Berthold, F., Sjă
oholm, E., & Lindstră
om, M. (2001). Fiber strength in relation
to molecular mass distribution of hardwood kraft pulp. Nordic Pulp and Paper
Research Journal, 16(4), 333–338. />Bohrn, R., Potthast, A., Schiehser, S., Rosenau, T., Sixta, H., & Kosma, P. (2006). The
FDAM method: Determination of carboxyl profiles in cellulosic materials by
combining group-selective fluorescence labeling with GPC. Biomacromolecules, 7,
1743–1750. />Bouchard, J., M´ethot, M., & Jordan, B. (2006). The effects of ionizing radiation on the
cellulose of woodfree paper. Cellulose, 13(5), 601–610. />s10570-005-9033-0
Brogdon, B. (2009). A fundamental review and critical analysis of hexenuronic acids and
their impact in elemental chlorine-free bleaching. 2009 TAPPI engineering, pulping an
environmental conference.
Burkart, P. (1999). Cellulose. Activation of dissolving pulp by electron beam irradiation.
A new route to a safer and more profitable viscose process. Polymer News, 24(6),
194–197.
Chai, X.-S., Zhu, J., & Li, J. (2001). A simple and rapid method to determine

hexeneuronic acid groups in chemical pulps. Journal of Pulp and Paper Science, 27,
165–170.
Chen, J., Wang, L., Su, X., Wang, K., Wu, X., Chen, L., et al. (2016). Structure,
morphology, thermostability and irradiation-mediated degradation fractions of
hemicellulose treated with γ-irradiation. Waste and Biomass Valorization, 7(6),
1415–1425. />Chen, Q.-Y., Ma, X.-J., Li, J., Miao, Q., & Huang, L. (2017). Effect of the utilization of
electron beam irradiation on the reactivity of bamboo dissolving pulp. Bioresources,
12(3), 6251–6261. />Cleland, M. R., & Galloway, R. A. (2015). Ozone generation in air during electron beam
processing. Physics Procedia, 66, 586–594. />phpro.2015.05.078
Dizhbite, T., Telysheva, G., Jurkjane, V., & Viesturs, U. (2004). Characterization of the
radical scavenging activity of lignins––natural antioxidants. Bioresource Technology,
95(3), 309–317. />Duan, C., Verma, S. K., Li, J., Ma, X., & Ni, Y. (2016). Combination of mechanical,
alkaline and enzymatic treatments to upgrade paper-grade pulp to dissolving pulp
with high reactivity. Bioresource Technology, 200, 458–463. />10.1016/j.biortech.2015.10.067
Faustino, H., Gil, N., Cecília, B., & Duarte, A. (2010). Antioxidant activity of lignin
phenolic compounds extracted from kraft and sulphite black liquors. Molecules, 15,
9308–9322. />Ferreira, J. C., Evtuguin, D. V., & Prates, A. (2020). Effect of cellulose structure on
reactivity of eucalyptus acid sulphite dissolving pulp. Cellulose, 27(8), 47634772.
/>ă
Gehringer, P. (1997). Radiation processing of aqueous systems (pp. 15). Seibersdorf: OFZS,
ă
Osterreichisches
Forschungszentrum Seibersdorf.
GlobeNewswire. (2020). Amid the COVID-19 crisis and the looming economic recession, the
Cellulosic Man-Made Fibers market worldwide will grow by a projected 2.2 Million Tons,
during the analysis period Accessed Date: 27 January 2021, Retrieved from https
://www.globenewswire.com/news-release/2020/06/03/2043336/0/en/Amid-the
-COVID-19-crisis-and-the-looming-economic-recession-the-Cellulosic-Man-Made
-Fibers-market-worldwide-will-grow-by-a-projected-2-2-Million-Tons-during-the
-analysis-period.html.

Gomes, V. J., Longue, D., Colodette, J. L., & Ribeiro, R. A. (2014). The effect of eucalypt
pulp xylan content on its bleachability, refinability and drainability. Cellulose, 21(1),
607–614. />Gondhalekar, S. C., Pawar, P. J., & Dhumal, S. S. (2019). Use of electron beam irradiation
for improving reactivity of dissolving pulp in viscose process. Journal of
Radioanalytical and Nuclear Chemistry, 322, 67–72. />Hammer, B. E., Christensen, N. L., Conroy, M. J., King, W. J., & Pogue, N. (2011).
Penetration depth measurement of a 6 MeV electron beam in water by magnetic
resonance imaging. Physical Review Special Topics Accelerators and Beams, 14, 11.
He, T., Liua, M., & Tian, X. (2018). Kinetics of ozone bleaching of eucalyptus kraft pulp
and factors affecting the properties of the bleached pulp. Bioresource Technology, 13
(1), 425–436. />Henniges, U., Hasani, M., Potthast, A., Westman, G., & Rosenau, T. (2013). Electron
beam irradiation of cellulosic materials—Opportunities and limitations. Materials, 6
(5), 1584–1598. />
10


O.P. Sarosi et al.

Carbohydrate Polymers 265 (2021) 118037

Staehelin, J., Buehler, R. E., & Hoigne, J. (1984). Ozone decomposition in water studied
by pulse radiolysis. 2. Hydroxyl and hydrogen tetroxide (HO4) as chain
intermediates. The Journal of Physical Chemistry, 88(24), 5999–6004. https://doi.
org/10.1021/j150668a051
TAPPI. (2007). Hexenuronic acid content of chemical pulp (T 282 pm-07).
Tripathi, S., Bhardwaj, N., & Ghatak, H. (2018). Additives to decrease cellulose chain
scission during ozone bleaching of wheat straw pulp. Nordic Pulp and Paper Research
Journal, 33(3), 430–438. />Tsuji-Katsukawa, S., Miyawaki, S., & Koyanagi, T. (2012). Development of a technique
for removal of hexenuronic acid in ECF-bleached pulp using UV irradiation. Kami Pa
Gikyoshi, 66(3), 287292. />
Vuorinen, T., Fagerstră

om, P., Buchert, J., Tenkanen, M., & Teleman, A. (1999). Selective
hydrolysis of hexenuronic acid groups and its application in ECF and TCF bleaching
of kraft pulps. Journal of Pulp and Paper Science, 25(5), 155–162.
Yang, Y. (2016). β-Elimination side reactions. Side reactions in peptide synthesis (pp.
33–42). Oxford: Academic Press. />Yang, G., Zhang, Y., Wei, M., Shao, H., & Hu, X. (2010). Influence of γ-ray radiation on
the structure and properties of paper grade bamboo pulp. Carbohydrate Polymers, 81
(1), 114–119. />Zhang, X. Z., Ni, Y., & van Heiningen, A. (2000). Kinetics of cellulose degradation during
ozone bleaching. Journal of Pulp and Paper Science, 26(9), 335–340.

11