FIBIC FuBio Cellulose programme report

Ohjelmatunnukset
Future Biorefineries
Products from Dissolved Cellulose
Programme Report 2011-2014

13
Ohjelmatunnukset
Future Biorefinery Cellulose
Programme Report 2011-2014

Copyright Finnish Bioeconomy Cluster FIBIC 2013. All rights reserved.
This publication includes materials protected under copyright law, the copyright for which is held by
FIBIC or a third party. The materials appearing in publications may not be used for commercial pur-
poses. The contents of publications are the opinion of the writers and do not represent the official
position of FIBIC. FIBIC bears no responsibility for any possible damages arising from their use.
The original source must be mentioned when quoting from the materials.
ISBN 978-952-67969-4-9 (paperback)
ISBN 978-952-67969-5-6 (PDF)
Layout: Brand United Ltd
Printing: Kirjapaino Lönnberg
Scarf photo (cover) by: Mikko Raskinen
CONTENT
Foreword...........................................................................................................................................................5
Promising results in creating new cellulose-based products in novel value chains............6
Introduction.................................................................................................................................................... 8
Ionic liquid based dissolution and regeneration processes........................................................12
Water based dissolution and regeneration processes.................................................................40
Textile value chain and recommendations for future research priorities
related to FuBio Cellulose textile fibres.............................................................................................60
New cellulose products............................................................................................................................ 72
Cationic cellulose based chemicals................................................................................................... 102

FOREWORD
The Finnish forest industry is undergoing active renewal. This is being pursued partly due to
changes in the traditional business environment, but also as a response to the opportunities
presented by the emerging global bioeconomy. New wood-based products and related new and
previously less known value chains are under scrutiny. Alongside paper, board and tissue, which
remain the backbone of the pulp industry, the value and versatility of wood as a raw material is
being intensively explored to its fullest potential, with a key focus on wood as a viable renewable
alternative to petroleum-based resources.
The five-year research programme Future Biorefinery (FuBio) was launched in 2009 by the Finnish
Bioeconomy Cluster FIBIC (formerly Forestcluster Ltd.). During the first phase of FuBio, several
new pathways for wood-based bio-products were studied and pre-evaluated. This laid the ground
for the next phase of the programme, which was dedicated to exploiting the most promising
results of phase one in order to create new value chains and future business opportunities for the
participating companies.
In this second phase, launched 2011, FuBio was split into two separate programmes, FuBio Joint
Research 2, focusing on bioeconomy research, and the present programme FuBio Products from
Dissolved Cellulose. During the programme planning stage, commercial interest towards man-
made cellulosic fibres grew to new heights with the price of dissolving pulp peaking in early 2011.
This market pull had a clear impact on the planning process. In addition, the first phase of FuBio
had introduced some interesting processing alternatives for dissolving and regenerating cellulosic
fibres and other shaped particles. These new technologies formed the basis of the new programme.
FuBio Products from Dissolved Cellulose focused on developing new processes for dissolving
and regenerating cellulose. The aim was twofold: to produce new fibres for use in textiles and
nonwoven products, and to produce new cellulose-based materials – such as thermoformable
cellulose and cationic cellulose derivatives – for use in water treatment. Understanding the
business environment and cost structures as well as value chain formation and value generation
were identified as critical, and these aspects were thus built in as a separate, generic work package.
Considerable effort was focused on techno-economical modelling and business area and value
chain analysis. It was of primary importance for the programme to include as programme partners
companies active downstream in the studied value chains – i.e. potential new customers for pulp
manufacturing companies. Another objective of the programme was to introduce demonstration
products or materials in order to generate interest among potential future customers. This was
largely successful, and also public interest was raised through the demo products.
FuBio Products from Dissolved Cellulose achieved the majority of its set targets and succeeded
in bringing Finland’s forest industry an important step closer to its ultimate goal of renewal. The
success stories highlighted in this report are a true embodiment of this achievement.
Kari Kovasin
Metsä Fibre, Industrial Coordinator of Programme

FUBIO CELLULOSE PROGRAMME REPORT6
New value chains will have a major impact
on the renewal of the Finnish forest industry.
Research performed in FIBIC programmes has
already shown the power of novel value chains
in creating new processes and products based
on new raw materials.
The FuBio Cellulose programme (FuBio
Products from Dissolved Cellulose) has been
the first programme in the FIBIC framework to
be focused on a value chain. This approach to
a platform for promoting value creation in the
Finnish forest industry of the future has proved
to be a good choice. At the same time the
programme has been able to improve general
awareness of new wood-based biorefinery
opportunities among the industrial companies
of the sector.
The specific target of the programme was
to develop novel sustainable processes for
production of staple fibres, new cellulose-based
materials and water treatment chemicals.
The industrial partners of the programme
believe that the targets have been met very
well, although the original ambition level was
high, as it has to be in this kind of research.
The partners represented a broad selection of
forestclustercompanies.Inthefollowingchapters
the companies highlight the business relevance
of the results achieved in the programme.
A breakthrough in ionic liquids
A clear breakthrough in the FuBio Cellulose
programme was made in research into ionic
liquids used in the production of textile
fibres. This work has achieved its targets and
generated very valuable results.
Ionic liquids have been a key area in the
programme. A team of researchers from the
University of Helsinki and Aalto University has
been developing new ionic liquids systems that
have high potential in industrial applications.
They have even been able to spin regenerated
fibres produced from ionic liquids. This work
was highlighted when a dress knitted from birch
cellulose fibre was displayed at a Marimekko
fashion show in the spring of 2014.
The combined efforts of the University of
Helsinki and Aalto University have contributed
to the formation of ionic liquids competence on
which the future application-oriented research
can be built.
PROMISING RESULTS
IN CREATING NEW CELLULOSE-BASED PRODUCTS
IN NOVEL VALUE CHAINS
Kari Kovasin, Metsä Fibre:
“The industrial concepts of ionic liquids will
require further development, but the roadmap
has been clearly defined by the results achieved
in the FuBio Cellulose programme.”
Jari Räsänen, Stora Enso:
“Fibre processing by using ionic liquids is very
high on our research agenda for the future.”
Margareta Hulden, Suominen Nonwovens:
“The results in the ionic liquids research indicate
many new product opportunities for wood-based
pulp as a feedstock. As our company makes
nonwovens, we would be a downstream user of the
potential new products, but we feel that it is always
beneficial to be involved at an early stage in the
development of raw materials that we can utilize.”

FUBIO CELLULOSE PROGRAMME REPORT 7
Feasibility studies proved to be of great
importance
Several industrial participants have expressed
appreciation for the techno-economic feasibility
analyses of selected value chains and novel
cellulose-based processes performed in the
FuBio Cellulose programme.
Without understanding the characteristics of
new value chains that may provide companies
in the pulp and paper industry with new
businesses, it is extremely difficult to develop
the concepts required and to steer the research
in the correct direction.
Moreover, by considering the techno-
economic aspect at an early stage of the
project, the management group and project
teams were aided in the prioritisation and
selection of development paths.
A good basis for further development
The programme and its results have been very
innovative and future-oriented. And it has given a
proofofconceptforthehigh-valueprecommercial
research in forest-based sector in Finland.
The joint research teams and knowhow
platform created by the programme will make
it easy to continue the work. Many companies
already have plans for how they will use some
of the results in the company-specified future
development projects.
The participants agree that the results point
in the right direction, but more work is needed
before business potential can be determined.
A good example of common interest for the
combined research efforts and cooperation is
the new Advanced Cellulose for Novel Products
(ACel) programme that FIBIC launched in the
summer of 2014.
Esa Hassinen, UPM-Kymmene:
“Upgrading wood-based cellulose to higher-value
products fits well with our strategic targets.
Broadening the use of wood and fibres, and using
biomaterials for new and existing applications,
are good candidates for future businesses.”
Kari Saari, Kemira:
“The knowledge created in activation and
modification of cellulose can be utilized broadly
in future programmes. Kemira will continue to
study production and utilization of cellulose-
based products in different applications. We
expect good ongoing collaboration with the
competent partners we have had in the FuBio
Cellulose programme.”
Advances in several areas
Great advances were made in several research
areas of the FuBio Cellulose programme. For
example, it developed and demonstrated a new
cellulose dissolving-regeneration process with
high sustainability and quality features, which
is seen as a promising new alternative.
Non-fibre products, such as absorbent
materials, have also been regenerated. They will
be suitable for a great variety of end-products
in the future.
Another major focus was to develop water-
soluble cellulosic products to be used in various
water treatment or paper manufacturing
processes. They are good bio-based candidates
for the future processes.

INTRODUCTION
Background
Future Biorefinery (FuBio) has been a strategic
focus area of the Finnish Bioeconomy Cluster
(FIBIC) during 2009-2014. The overall objective
of FuBio research and development was to
establish in Finland globally-competitive
knowledge platforms for the renewal of
existing forest industry and the creation of new
business. The focus has been on creating new
value chains in which biomass-based materials
and chemicals are applied in substantial global
markets. The potential markets of focus are
both well-known to the forest industry (e.g.
fibre-based packaging) as well as essentially
new (e.g. textiles, nonwovens, polymers,
resins and thermoformable composites).
The creation of new biorefinery value chains
requires deep understanding of the biomass
structure. In addition, new processing
technologies must be developed hand in
hand with new biorefinery concepts and their
related value chains. Understanding of the
markets and freedom-to-operate are also
needed, and the first steps towards future
industrial partnerships must be taken.
The first, two-year Future Biorefinery
programme (FuBio Joint Research 1) was
completed in May 2011. This research was
thereafter continued through two separate
programmes, FuBio Joint Research 2 (FuBio
JR2) and FuBio Products from Dissolved
Cellulose (FuBio Cellulose). FuBio Cellulose,
the focus of this report, was a value chain
oriented programme building on the knowledge
generated in FuBio Joint Research 1 on novel
cellulose solvents and the modification of
dissolved cellulose to produce bio-based
materials and chemicals.
The motivation for the FuBio Cellulose value
chain programme stemmed from the needs of
the radically evolving forest industry. Upgrading
of wood cellulose to higher value products fits the
strategic targets of generating new end uses for
wood and fibre and using biomaterials in new and
existing applications. In Finland, industrial interest
and activity towards novel, wood cellulose-based
regenerated fibre products are on the rise. This is
due mainly to promising market trends, especially
in textile fibres, combined with environmental
considerations related to the current dominant
raw material, cotton. In addition to fibre, other
products such as films, beads and other cellulosic
particles can also be regenerated, opening
opportunities for a wide range of end products.
Furthermore, targeted functionalization or
manipulation of the cellulose chain paves the way
towards generation of water soluble cellulosic
products for use in water treatment and paper
manufacture, as well as an exciting new target
area – thermoplastic cellulose.
FuBio Cellulose focused on the development
and evaluation of novel sustainable processing
concepts for selected cellulose products. The
programme aimed to promote the development
of the basic knowledge and techniques
of sustainable wood cellulose dissolution,
regeneration and functionalisation developed
in FuBio 1 towards process concepts suitable
for industrial feasibility evaluation through
cellulose-focused and process- and product-
oriented high-quality research.
The three-year FuBio Cellulose research
programmehadtotalbudgetof11.6millioneuros.
The Finnish Funding Agency for Innovation
(Tekes) provided 60% of the financing, with
the remainder sourced from the participating
companies and research institutes.

Programme goals and structure
The main goal of FuBio Cellulose was to develop
novel sustainable processes for the production
of i) regenerated cellulose staple fibres, ii) novel
functional materials based on cellulose beads,
nonwovens or thermoformable structural
materials, and iii) cationic cellulose chemicals for
water treatment. The programme was composed
of five interlinked work packages contributing
to these selected focus areas (see Figure 1). The
programme’s value-chain approach aimed at
buildingtechnologiesandexpertiseindissolution,
regeneration and product development within
the selected areas, thus providing a platform
for future value creation for the Finnish forest
industry and cellulose converting industry value
chains. Concrete process concepts were built
based on the selected research paths. Techno-
economic evaluations were carried out for
selected concepts and these guided the technical
process development work throughout the
programme. Market analyses provided valuable
information on the value chains in general and on
the value generation mechanisms of the selected
value chains.
The first target of the programme was to
develop a new process for the production of
cellulosic staple fibres from dissolving grade
pulp. The main emphasis was on sustainable,
techno-economically feasible process concepts
that could replace the current industrial NMMO-
based lyocell process or viscose process. The
research focused on two approaches: ionic liquid
based dissolution and fibre regeneration, and
water-based dissolution and fibre regeneration.
The ionic liquid based process development drew
on the knowledge on cellulose-dissolving ionic
liquids developed (at University of Helsinki) and
the new dry-wet spinning equipment line built (at
Aalto University) by FIBIC during FuBio JR1. The
research on water-based process development
focused on generating basic understanding of the
factors affecting the dissolution and regeneration
of pulp cellulose to state-of-the-art water-based
Biocelsol system, which was used as a reference.
In both approaches, special emphasis was given
to demonstration of the properties of novel
regenerated fibres in clothing applications and
modelling the technical and economic feasibility
of the most promising novel processes.
The second target of the programme was
to develop processes for the production of
two new high-volume products or product
platforms based on cellulose beads, nonwovens
and thermoformable structures from dissolved
cellulose, without the use of spinning
regeneration. The end-product areas of focus
were hygienic products, packaging, and medical
componentcarriers,allselectedbasedonmarket
studies carried out in the programme. The
research focused on processes for producing
absorbing cellulose materials, thermoformable
cellulose derivatives and slow-release cellulose
beads. Special attention was paid to economic
factors and the properties of the cellulose
materials produced. Demonstrations of the
most promising materials were targeted in all of
the focus end-product areas.
The third target of the programme was to
develop a new process for producing a water-
soluble, cellulose-based polyelectrolyte
chemical product. The research focused on the
development of a techno-economically feasible
synthesis route for cationic water-soluble
cellulose derivatives from wood pulp. Two main
synthesis lines – water-based and organic solvent
based – were targeted after the initial screening
phase. The most promising synthesis products
produced at laboratory scale were tested as
paper and/or water processing chemicals and
benchmarked in selected applications against
commercial reference chemicals.

Management of the programme
TheFuBioCelluloseprogrammewasadministered
by a Management Group (MG) comprising
representatives from industry and academia.
Execution of the programme was headed by a
Programme Manager together with Industrial
and Scientific Coordinators. Daily management
tasks were performed in each Work Package (WP)
under the leadership of the WP manager.
The main tasks of the Management Group were
to supervise the progress of the programme with
respect to the objectives of the FuBio Cellulose
programme plan, and to assess the scientific
progress and techno-economic feasibility of the
results. The MG had the following members:
• Heikki Hassi, Carbatec, until March 2013
• Esa Hassinen, UPM-Kymmene (Eeva
Jernström until September 2012)
• Margareta Huldén, Suominen
• Ilkka Kilpeläinen, University of Helsinki,
Scientific Coordinator
• Kari Kovasin, Metsä Fibre, Chairman,
Industrial Coordinator
• Jukka Laakso, Tekes
• Markku Leskelä, FIBIC (Lars Gädda until April
2012)
• Jari Räsänen, Stora Enso
• Kari Saari, Kemira
• Anna Suurnäkki, VTT, Programme Manager
Dissemination of the FuBio Cellulose
programme results was achieved with a number
of different tools, the most important being
the FIBIC research portal, accessible to the
FuBio Cellulose programme participants, and
the FIBIC Ltd website open to the wider public
(http://fibic.fi/programmes/fubio-cellulose).
Both internal and public programme seminars
were held annually. The public seminars held
jointly with the FuBio JR2 programme brought
together experts from academic and industrial
fields and provided a comprehensive overview
of the research activities and results of both
the FuBio Cellulose programme and the whole
Future Biorefinery entity in Finland.
Participants and international
cooperation
The FuBio Cellulose programme brought
together the leading forest cluster companies,
selected value chain companies in nonwoven
and staple fibre areas, and public research
groups related to chemical pulping technology,
cellulose material science, modelling and
simulation and cellulose product applications
in Finland. Six companies (seven until 2013)
and six Finnish universities and research
institutes participated in the programme. In
addition, material demonstration work was also
subcontracted from external partners.
Industrial partners
• Carbatec, withdrawn 2013
• FIBIC
• Kemira
• Metsä Fibre
• Stora Enso
• Suominen
• UPM-Kymmene
Research organizations
• Lappeenranta University of Technology
• Tampere University of Technology
• University of Helsinki
• University of Oulu
• VTT Technical Research Centre of Finland
• Åbo Akademi
International collaboration was integral to
the FuBio Cellulose programme. The research
and company networks generated play an
important role in the further development of
the wood cellulose based value chains and
the Finnish knowledge base supporting this

development. The programme partners worked
collaboratively with several research groups
from five countries: Germany, Latvia, Poland,
Portugal, Spain, and Sweden. Close links with
the international scientific community will be
maintained and strengthened in the future,
particularly in the research areas of cellulose
dissolution and regeneration as fibres by novel
methods, chemical modification of cellulose
and cellulose structure characterization.
Programme participants have been active
in presenting the programme results at
international conferences and workshops.
Furthermore, programme results have been and
will continue to be published in scientific journals
as peer reviewed papers. The programme
results have also been communicated with the
value chain companies outside the programme
consortium. The novel wood cellulose based
textile fibres produced using the processes
developed in FuBio Cellulose drew national
attention in 2014 with the presentation of a
dress manufactured from these fibres by design
company Marimekko (see: http://fibic.fi/results).
Figure 1. FuBio Cellulose programme structure.
Dissolution of cellulose
• New ionic liquid based processes
• New water-based processes
• Modification
Cellulose-based chemicals
• Cellulose activation
• Synthesis routes for cationic polymers
• Synthesis of cationic particles
• Application & scale-up
New products
Absorbents for hygiene products
Thermoformable structures
Products based on cellulose beads
Material demonstration
Markets and economics
Textiles and nonwovens via
spinning regeneration
• Regeneration to fibres
• Modification
• Nonwovens
• Modelling cellulose in processing
The FuBio Cellulose programme has been
closely linked to the FuBio Joint Research 2
programme, especially in the development of
cellulose dissolving ionic liquids. It is also the
basis for the new Advance Cellulose to Novel
Products (ACel, 2014-2017) programme of FIBIC
Ltd. Many of the programme’s researchers
have also been involved in other on-going,
related national and international projects. This
has ensured active information exchange and
synergistic knowledge generation among the
Finnish and international research community.
FuBio Cellulose research groups participated,
for example, in the European Community’s 7th
Framework Programme projects and several
COST actions.
The FuBio Cellulose programme’s results
support industry-driven projects aimed at
developing novel business based on wood
cellulose. Active participation of industrial
partners within the programme has ensured
effective information flow from research to
innovation,thusspeedingbusinessdevelopment
among the participating companies.

CONTAC T PE RSON
Kristiina Poppius-Levlin, kristiina.poppius-levlin@vtt.fi
PAR TN E RS
Aalto University
Glocell
Lappeenranta University of Technology
Metsä Fibre
Pöyry Management Consulting
Stora Enso
University of Helsinki
University of Oulu
UPM-Kymmene
VTT Technical Research Centre of Finland
IONIC LIQUID-BASED
DISSOLUTION AND
REGENERATION
PROCESSES

FUBIO CELLULOSE PROGRAMME REPORT
ABSTRACT
13
The main objectives were to develop novel sustainable ionic liquid-based (IL) solvent sys-
tems with the capability to dissolve cellulose pulp of sufficiently high molecular weight to
achieve the targeted mechanical fibre properties upon regeneration and to develop com-
mercially viable cellulose staple fibre spinning processes for cellulose/IL solutions.
Detailed knowledge of the rheological behaviour of the IL-cellulose solutions, i.e. dopes,
is a prerequisite for determination of the viscoelastic properties and further processing
of the dopes. Various pulps of different grade and origin were analysed and dissolved in
different ILs. Obtained insights and knowledge of dope properties were crucial for the
development of the spinning window, i.e. for the prediction of optimal spinning conditions.
The hitherto unreported and distillable IL [DBNH][OAc] proved to be an excellent solvent
for the production of cellulosic fibres with strength properties significantly higher than
those of other man-made commercial fibres. Cellulosic textile fibres were produced with
tensile strength properties (>50 cN/tex) exceeding the initial target (≥35 cN/tex). Two de-
monstration products were manufactured: a scarf made of eucalyptus pulp and a dress (in
collaboration with Marimekko) made of Enocell birch dissolving pulp.
To assess sustainable chemical modifications of pulp cellulose prior to dissolution, a wide
range of chemical reactions were carried out in commercially available and novel, distillable
ILs. A sustainable acetylation process of pulp cellulose in distillable IL, i.e. [DBNH][OAc], was
of high potential as the mechanical properties of the chemically modified and spun fibres
were good. A new cellulose modification method – cellulose alkoxy carbonylation – was also
developed using ILs as a direct dissolution solvent.
In recovery and recycling studies of [emim] [OAc], polymeric ultrafiltration and nanofilt-
ration (NF) membranes as well as a TiO2
ceramic NF membrane gave good retention of
organics while not retaining the IL. Reverse osmosis was able to remove some water from
IL-water solution ([DBNH][OAc]). In addition, pervaporation showed potential as a method
for separating water from IL. Testing with ion-exchange resins showed their potential to
remove possible metals from spinning bath solution. The most promising concept for DIL
(distillable IL) recovery was based on evaporation and distillation technology.
Keywords:
carbonate cellulose, chemical modification, distillable ionic liquid, dissolving pulp, dry-jet
wet fibre spinning, fibre, ionic liquids, ion-exchange, membrane separation, purification
technology, lyocell process, nanofiltration, pervaporation, reverse osmosis, rheology,
solute exclusion, sustainable

1. Work background
Increasing global demand for consumer goods
is generating robust growth in the textile
fibre market. Total fibre consumption in 2030
is predicted to rise to more than 130 million
tonnes, with a predicted share of cellulosic
fibre of ca. 30%. Paired with the stagnation of
cotton production, this will create an annual
shortage of 15 million tonnes of cellulosic fibre.
This ‘cellulose gap’ opens up new opportunities
for man-made cellulosic fibres. For the Finnish
forest industry, wood cellulose upgrading to
higher value products fits the strategic target
of generating new uses for wood and wood-
based fibres. The promising market trends
in textile fibres have aroused interest and
activity towards novel, wood cellulose-based
regenerated fibre products. Accordingly, one
of the strong platforms identified as the main
outcome from the FuBio1 programme, which
ended in May 2011, was “New knowledge on
cellulose dissolution in novel, recyclable ionic
liquids”. A further goal was set to convert the
generated competences into market-driven
value chains.
Many attempts have been made to develop
alternative regenerated cellulosic fibre
processes that are competitive or even superior
to the well-established viscose process. So far,
only one technology fulfilling these criteria,
lyocell, is in industrial use. The process is based
on pulp dissolution in N-methylmorpholine-
N-oxide (NMMO) to form a spinning solution.
However, certain intrinsic properties of NMMO
render the solvent prone to thermal run-
away reaction and cellulose degradation, thus
necessitating an appropriate stabilizer. This
limits the versatility of the process.
Approximately a decade ago, ionic liquids were
identified as powerful direct cellulose solvents.
Their thermal and chemical stability can be
utilized to circumvent problems associated with
NMMO. Of the relatively few ionic liquids that
have been studied, the majority are imidazolium
based, thus having moderate thermal stability,
and some reactivity towards cellulose via
carbene formation. Moreover, only little progress
in the formation of cellulosic fibres from IL
solutions has been reported so far.
A big challenge for IL-based pulp dissolution
and regeneration systems is efficient and
economical IL recovery. ILs need to be circulated
and reused efficiently. For this purpose, new,
easily recyclable ILs also had to be developed
as, for example, distillable ILs were not available.
2. Objectives
The main objectives were to develop novel
sustainable ionic liquid-based (IL) systems
with the capability to dissolve cellulose pulp of
sufficiently high molecular weight necessary
to achieve the targeted mechanical fibre
properties upon regeneration. Chemical
pulp modifications during dissolution were
aimed at enhancing water uptake of the
regenerated fibres. The overall goal was
to develop commercially viable cellulose
staple fibre spinning processes for cellulose/
IL solutions and to improve the properties
(fibrillation and mechanical properties) of
the obtained regenerated fibres so that they
can be demonstrated in textile structures. In
order to develop an efficient, commercially
and environmentally viable IL-based process,
recycling of ILs is of crucial importance.

3. Research approach
The overall approach was to develop and
demonstrate a novel IL-based textile value
chain spanning from wood and chemical pulp
production to pulp dissolution in ILs, fibre
regeneration, yarn spinning and, finally, fabric
production (Figure 1).
In the commercial lyocell process, pulp is
dissolved in N-methylmorpholine N-oxide
(NMMO) monohydrate (non-derivatizing
solvent) to achieve cellulosic textile fibres.
To circumvent the problems associated with
NMMO, the potential of different ionic liquids
(IL) for the production of man-made cellulosic
fibres was studied.
AcommerciallyavailableIL,[emim][OAc](1-ethyl-
3-methylimiadzolium acetate), known to be an
excellent cellulose solvent, was used for initial
trials. Several dissolving pulps were dissolved
and the rheological properties of the resulting
solutions were assessed. An understanding of
the factors governing the rheological properties
of cellulose solution was of great importance
for solution processing. Thus, a classical shear
rheometer to assess the viscoelastic properties
and an extensional rheometer to determine
the elongational-rheological properties were
used. The goal was to establish a relationship
between pulp properties (molecular weight
distribution), the rheological properties of the
spin dope, and the spinnability. Spinnability
describes the extrusion behaviour of the dope
and the filaments’ stability in the air gap when
stretched, i.e. extensional stress exerted.
The chemical stability of cellulose in the IL is
important for the final fibre properties and for
the development of a recycling strategy. Thus,
respective pulp solutions in [emim][OAc] were
tested with time-temperature degradation
tests. All properties were compared to NMMO.
Once a basic understanding was established,
other ILs were tested. Besides known cellulose-
dissolving ILs, promising novel, distillable ILs
were developed and their suitability in fibre
spinning was tested. Hence, respective pulp
solutions were prepared and characterized in
detail.
The strategy for chemical modification of
pulp cellulose in ILs was to modify cellulose
to low DS (degree of substitution) in order to
Figure 1. Ioncell-F textile chain – from wood to garment.

allow disruption of the crystallinity of cellulose
and hence increase the water absorptivity
of the resulting regenerated fibres. Similar
incorporation of other alternative functionalities
besides low DS may also afford novel properties
and increased water retention. Therefore, the
research approach was to look for sustainable
chemical modification procedures for
modification of cellulose in the chosen ionic
liquid for fibre spinning. At the early stages the
fibre-spinning process was not established, so
a wide range of ionic liquids was tested. Many
types of chemical modification were also tested
to see which ones would be atom efficient, cause
minimal degradation and be sustainable. It was
initially intended that the regenerated cellulose
properties would guide the development, but
this approach proved impractical. Instead, it was
found to be more effective to examine a wide
range of chemistries to see which ones were
suitable and then transfer these to the resource-
intensive fibre spinning directly.
In the spinning trials, the commercially available
ionic liquid [emim][OAc] was chosen as the first
IL to be tested. Different dissolving pulps used
in the lyocell process were used to benchmark
the first results. Since the formation of a single
monofilament is more straightforward, this was
studied first. Subsequently, other novel volatile
ILs were implemented. One new IL, [DBNH][OAc]
(1,5-diazabicyclo(4.3.0)non-5-enium acetate)
showed excellent spin stability, thus enabling
the effects governing the multi-filament
spinning process to be studied. Different pulps
of lower quality (higher hemicellulose and
residual lignin content) were also spun, and the
effects of chemical cellulose modification on
spin stability and fibre properties were studied.
A comprehensive set of analytical tools was
employed to shed light on the mechanisms
of solution spinning and to characterize the
resulting fibres not only in terms of their
mechanical properties but also their (supra-)
molecular structure.
In the functionalization of regenerated fibres,
the goal was to improve fibre properties, such
as to reduce fibre fibrillation, through chemical
modification. In recent years, robust, quick, and
high fidelity chemical reactions tolerating both
water and oxygen have been developed under
the context of click chemistry. Alongside this,
irreversibleadsorptionofcertainpolysaccharides,
such as carboxymethyl cellulose (CMC), is a
well-established phenomenon. These two
concepts can be combined to provide a generic
modular platform. In the first step, modified
polysaccharide chains with clickable functional
groups were physically adsorbed on the cellulose
surface.Secondstepwastheactualclickreaction,
in which the desired molecule was covalently
attached to the modified polysaccharide in-situ
as already adsorbed on the surface.
To ensure the economic viability of IL-based pulp
dissolution and regeneration systems, ILs need
to be circulated and reused efficiently. Use of
pressure-driven membrane separation processes
for purification, recovery and concentration of
ILs is one potential approach. To examine this
approach, the filterability and tolerance of selected
membranes towards the ILs – [emim][OAc] and
distillable [DBNH][OAc] – were first studied. After
identifying suitable membranes, their usability
for removing impurities, such as carbohydrates
dissolved during the process, was proved.
The efficient removal of water from the ionic
liquids used in the cellulose dissolution process
is also a prerequisite for the feasibility of
the ionic liquid based fibre spinning process.
Pervaporation proved to be a potential energy-
efficient separation, purification and recovery
technology for this purpose.
Preliminary techno-economic screening of
DIL (distillable ILs) recovery concepts was
carried out, including identification of different
recycling concepts and estimation of the main
production costs of the concepts.

4. Results
4.1 Programme pulps and ILs
Pulps
Three different common pulps were selected for
the programme in order to be able to compare
results between different partners in the
programme. Additional pulps were also used in
different programme activities as necessary. The
programme pulps were acquired, characterized
with a number of methods and delivered to all
partners in the programme (FBC-pulp1: Domsjö
softwood sulfite pulp; FBC-pulp2: Eucalyptus
urograndis pre-hydrolysis kraft pulp, Bahia
Solucell; and FBC-pulp3: Borregaard spruce
sulfite pulp) (Table 1).
Ionic liquids (ILs)
The main ionic liquids used in the programme
are shown in Table 2 and the structures are
presented in Figure 2. In the case of the TMG
and DBN-based distillable ionic liquids, the
propionate versions were also briefly studied.
4.2 Dope properties
Detailed knowledge of the rheological behaviour
of the IL-cellulose solution, i.e. dopes, is a
prerequisite for determination of viscoelastic
properties and for further processing of the
dopes. Various cellulosic solutes (pulps with
different cellulose content and intrinsic viscosity
levels) were used for dope preparation. Figure
3 shows the complex viscosities and dynamic
moduli of three pulps in [emim][OAc]. Although
the softwood and beech sulfite pulps have
similar intrinsic viscosity values (540 and 520
ml/g, respectively) their respective IL-solutions
differ significantly. On the other hand, solutions
of the eucalyptus pre-hydrolysis kraft pulp (FBC-
pulp2) and beech sulfite pulp are similar in terms
of their viscoelastic properties although the pulp
viscosities were different.
FBC-pulp1, Domsjö
Sulfite, SW/Spruce-Pine
FBC-pulp2, Bahia, HW/
Euca, PHK
FBC-pulp3, Borregaard,
SW/Spruce Sulfite
Viscosity, ml/g 520 470 1520
Kappa no 0.48 0.3 4.1
Glucan, rel% 89.8 95.8 90.8
Xylan, rel% 1.4 2.6 3.6
Mannan, rel% - - 2.1
Mn, g/mol (SEC) 47 600 34 300 32 100
Mw, g/mol (SEC) 405 000 358 000 1 307 000
PD (Mw/Mn) (SEC) 8.5 10.4 40.7
Mn, g/mol (MALLS) 41 300 62 900 67 200
Mw, g/mol (MALLS) 530 500 196 400 792 000
DP<100, w% 4.1 3.7 7.3
DP>2000, w% 39 14.6 58.7
PD (Mw/Mn) (MALLS) 12.9 3.1 11.8
Crystallinity, % 54 59 -
Fibrils, nm (Lateral
dimens.)
4.2 4.8 -
Aggregates, nm 14.2 36 -
Table 1. Programme pulps and their properties.

Name of IL Abbreviation Comments, main uses in the
programme
1-ethyl-3-methylimidazolium acetate [emim][OAc]] For dope property and
rheological studies; benchmark
literature IL
1-ethyl-3-methylimidazolium
methylhydrogenphosphonate
[emim][MeHPO3] For IL phosphonate anionization
of cellulose
N,N,N,N-tetramethylguanidinium acetate [TMGH][OAc] 1st
generation distillable IL
1,5-diazabicyclo(4.3.0)non-5-enium acetate [DBNH][OAc] Dope property studies; current
fibre-spinning IL
1-methyl-1,5-diazabicyclo(4.3.0)non-5-enium
dimethylphosphate
[mDBN][Me2
PO4
] Alternative low-viscosity non-
distillable structure for rheology
testing
methyltrioctylphosphonium acetate [P8881
][OAc] Phase-separable ionic liquid
Table 2. Main ILs used in the programme.
Figure 2. Main ionic liquids studied during the programme.
Figure 3. Complex viscosity and dynamic moduli of FBC-pulp2 (Euca PHK pulp, 470 ml/g) (blue); FBC-pulp1
(Domsjö 540 ml/g) (black), beech sulfite pulp (Lenzing, 520 ml/g) (red) in [emim]OAc (all solutions 10 wt-
%, at 60°C).

It is known that cellulose undergoes degradation
in IL solutions. To study the stability in detail, a 10
wt-% solution of eucalyptus pre-hydrolysis kraft
pulp (FBC-pulp2) in [emim][OAc] was prepared
and then stored at different temperatures
for various periods. The solutions were then
characterized in terms of rotational shear and
extensional viscosity before the cellulose was
regenerated and its intrinsic viscosity measured.
Figure 4 (a) reveals substantial degradation of
cellulose at temperatures of 90°C or higher.
However, the cellulose is not affected when
stored at 60°C for 24 h. The cellulose degradation
is reflected in the viscoelastic properties of the
respective solutions (Figure 4 b) and can thus
Figure 4. DP of the regenerated cellulose of IL-cellulose solutions stored at different temperatures as a
function of storage time (a) and Zero shear viscosity from respective solutions (b); DP calculated from the
intrinsic Cuen viscosity.
a) b)
Figure 5. Molecular weight distribution (a) and zero-shear viscosity (full symbols) and crossover moduli
(open symbols) as a function of COP-angular frequency (b).
be assessed without laborious regeneration of
the cellulose. Extensional studies show a very
sensitive response in elongational relaxation
time to cellulose degradation. This is important
for predicting the spinnability of various dopes.
It should be noted that substantial degradation
already occurs during the dissolution process.
Thus, theeffectof propylgalate(PG)as a stabilizer
– as used in lyocell solutions – was studied. The
addition of PG reduced degradation substantially.
In order to study the influence of molecular
weight distribution on the spinnability of the
resulting dope, different (native and degraded)
pulps were mixed and dissolved in [emim][OAc]
(Figure 5.) Only Blend 2 showed good spinnability.
a) b)

Thezero-shearviscosityandtheangularfrequency
and dynamic modulus of the crossover point (COP)
of a cellulose-IL solution need to be within specific
ranges to obtain successful spinning. A zero-shear
viscosity between 27000 and 30000 Pa·s and a
crossover point between 0.8 and 1.2 s-1
and 3000
and 5500 Pa, respectively, seem to be required.
Furthermore, it seems that the spinnability of a
cellulose-IL solution is very sensitive to the high
molecular weight fraction of the cellulosic solute
andtothepolydispersityindex(PDI).Forsuccessful
spinning, a high molecular weight content greater
than 20% and a PDI higher than 3 appeared to be
favourable.
Main achievements
• Obtained insights and knowledge of dope
properties were crucial for the development
of the spinning window, i.e. prediction of
optimum spinning conditions.
• Various pulps of different grade and origin
were analysed and dissolved in ionic liquids.
Even low-grade pulps were successfully
spun in appropriate spinning conditions.
4.3 Chemical modification of pulp
cellulose in ILs
In examining the chemical modification of pulp
cellulose in ILs, several sustainable strategies
were developed with the aim of imparting
novel properties to the regenerated fibres,
such as increased water absorptivity, reduced
fibrillation or fire-resistance.
Cellulose etherification with epoxides
A typical reaction scheme for cellulose
etherification, i.e. preparation of hydroxypropyl
cellulose, is shown in Figure 6. Cellulose
(typically 5-10% w/w) was dissolved in different
ILs, typically [emim][OAc], [DBNH][OAc] and
compositions of DMSO and [P8881
][OAc] (0-
40% w/w DMSO). After dissolution, propylene
oxide was added (10 eq) and the mixtures were
heated for a set time period.
Using recyclable IL systems, [DBNH][OAc] and
[P8881
][OAc] gave better product quality and
cleanliness than [emim][OAc]. The reaction is,
however, still not very atom-efficient (10 eq of
epoxide used) and cellulose and ionic liquid were
found to degrade to some extent under the used
conditions, even in the presence of catalysts.
Cellulose alkoxycarbonylation using
dialkylcarbonates
New cellulose derivatives – cellulose alkyl
(methyl or ethyl) carbonates – were successfully
prepared (Figure 7). The optimum procedure for
their preparation is by dissolution of cellulose
in 10 wt% DMSO:[P8881
][OAc] (phase-separable
ionic liquid electrolyte) and using dimethyl or
diethylcarbonate. Reaction also succeeded
in [P8881
][OAc] and [emim][OAc]. Products with
a DS (degree of substitution) up to 1 were
obtained. However, [DBNH][OAc] did not give
the desired product.
See-through and flexible cellulose methyl-
carbonate films were successfully prepared by
solvent casting from pyridine.
Corey-Kim oxidation
Corey-Kim oxidation is a method of selectively
converting alcohols to aldehydes or ketones.
The reaction was confirmed to occur in LiCl/
DMA (lithium chloride / dimethyl acetamide), but
overall the amounts of cellulose soluble at the low
temperatures required by the method, combined
with the inability to recycle all reagents made it
unlikely that this procedure could be transferred
to the spinning dopes. Despite this being a novel
and unpublished reaction in the literature, the
decision was made not to continue this work.
Cellulose esterification using anhydrides or
esters
Esterification of cellulose with carboxylic
anhydrides and esters was highly successful. A
sustainable method of cellulose acetylation was
developed. Fibres have been spun from these
dopes and the initial results look promising.

Main achievements
• Several new, potential strategies for
chemical modification of pulp cellulose in
different ILs were developed.
• Esterification of cellulose in ILs with
carboxylic anhydrides and esters was very
successful. Fibres have been spun from the
dopes and the initial results look promising.
• Transesterification of a phosphonate ionic
liquid with cellulose produced water-soluble,
film-castable and fire-retardant cellulose.
Unfunctionalized cellulose was found to be
regenerated by dispersing in dilute acid,
resulting in a novel cellulose regeneration
process.
• Alkoxycarbonylation of cellulose using
the green reagents dimethyl and
diethylcarbonate succeeded in ionic liquids.
Figure 6. Typical cellulose etherification conditions and reaction in DMSO:[P8881
][OAc].
Figure 7. Scheme for preparation of cellulose alkyl carbonates using dimethyl carbonate (DMC) or
diethylcarbonate (DEC) in [P8881
][OAc]:DMSO solutions.

4.4 Cellulose textile fibres via spinning
regeneration
Production of staple fibres – Ioncell-F process
A novel process for producing staple fibres from
ILs was developed and named Ioncell-F(iber)
in analogy to the lyocell process. Staple fibres
from various pulps were produced successfully
using a distillable IL ([DBNH][OAc]) as solvent.
Optimized multi-hole spinnerets with a two-
stage conical diminution (first cone 60°, second
cone 10°) and a spin capillary aspect ratio of
L/D = 0.2 showed no melt fracture. Fibres spun
from eucalyptus pre-hydrolysis kraft pulp
(FBC-pulp2)-[DBNH][OAc] solutions and their
extraordinarily high draw ratio are illustrated
in Figure 8. Mechanical properties of Ioncell
fibres are significantly better than those of other
man-made cellulosic fibres, such as viscose,
modal and lyocell (Tencel) (Figure 9).
Figure 8. Fibres spun from FBC-pulp2 (eucalyptus PHK)-[DBNH][OAc] solutions (a). Linear density (titer) and
tensile strength (tenacity) of fibres spun from FBC-pulp2-[DBNH][OAc] solutions as a function of draw (b).
a) b)
Figure 9. Mechanical properties of Ioncell fibres in comparison to other man-made cellulosic fibres.

Figure 10. SEM (a) and GPC (b) analysis of fibres spun from FBC-pulp2 (eucalyptus PHK)-[DBNH][OAc].
a)
b)
Fibre analysis
Standard fibre properties such as linear
density (titer) and tensile strength (tenacity)
were measured on a routine basis. In addition,
orientation and crystallinity were measured
via optical birefringence and X-ray analyses,
respectively.Morphologywasassessedbymeans
of SEM. Similar to lyocell fibres, the Ioncell fibres
showed high tenacity values that were retained
under wet conditions. Crystalline and total
orientation that was high and increasing with the
draw ratio. The fibres showed a typical fibrillar
morphology (Figure 10, a). Also, determination of
the molecular weight of the cellulose before and
after spinning showed that there is no significant
degradation during the spinning process (Figure
10, b). This is important not only for the fibre
properties but also regarding purification and
recycling of the IL.

Demonstrations of textile production
A demonstration run was performed to
demonstrate the applicability of the fibres for
textile production. Ca. 20 litres of [DBNH][OAc]
were synthesized at Helsinki University and
approximately 300 g of Ioncell staple fibres
spun at Aalto University. Together with the
Department of Design (School of Arts, Design
and Architecture, Aalto University) the fibres
were ring spun to a yarn at the Swedish School
of Textiles (University of Borås, Sweden), dyed
and flat-bed knitted (Figure 11). Fibre and yarn
properties are summarized in Table 3. Both
the IL-fibres and the IL-yarns have significantly
higher tenacity than commercial viscose fibres.
The official presentation of the scarf (Figure
11g) at the FIBIC annual seminar (autumn 2013)
attracted the attention of a Finnish textile and
design company Marimekko, who expressed
an interest in jointly producing a full garment.
Fibres from birch dissolving pulp (Stora Enso)
were subsequently similarly processed to
produce a dress (Figure 12), which was exhibited
at Marimekko’s Autumn and Winter Fashion
show (Helsinki railway station, March 2014).
This spin-off project of the FuBio Cellulose
programme is summarized in a video available
online (http://youtu.be/AGFDPyzN1C8).
Figure 11. Process steps during yarn manufacture and the final knitted product: a) carding of [DBNH][OAc]-
spun staple fibres; b) sliver feeding to the drafting machine; c) preparing the roving; d) feeding the roving;
e) ring spinning; f) plying; g) flat-bed knitted scarf.

fibreyarn
[DBNH][OAc] viscose
linear density (dtex) 1.9 1.5
dry tenacity (cN/tex) 47 23
elongation (%) 9.4 22.5
fibre length (mm) 37 40
finish no yes
linear density (tex) 54.3 62.7
tenacity (cN/tex) 34.4 17.3
elongation (%) 7.4 18.2
CV (%) 13.6 9.1
Table 3. Properties of yarns spun from [DBNH][OAc] and commercial viscose fibres.
Figure 12. Marimekko’s multi-functional dress produced from 100% Ioncell fibre (birch).

Extensional rheology experiments have thus
been conducted by means of a Capillary
Break-up Extensional Rheometer (CaBER) to
characterize the air gap phenomena. Once the
filament enters the spin bath, a complex solvent
exchange leads to the coagulation of the
cellulose and formation of the solid fibre. The
solvent exchange in the spin bath is suspected
to proceed via spinodal decomposition, which
largely preserves the molecular orientation
created in the spin capillary and air gap.
Batch experiments were conducted to study
the diffusion kinetics of the solvent and anti-
solvent. Upon solvent exchange, the water
content in the filament increases gradually
from the surface while, concomitantly, the
solvent level decreases. This causes a radial
gradient where the transition from filament
to fibre passes through various gel-states
(Figure 13). These gel states were assessed in
Figure 13. Cut through a simulated filament in the coagulation bath. The graphs show the solvent and
water content, respectively, as the coagulation proceeds. The solvent content decreases from blue to red.
X-axis shows the radial distribution, y-axis along the fibre.
Structure formation process
A better understanding of the effects and
factors governing the structure formation of the
cellulosic fibre, i.e. the transition from solution
to solid state, is needed to tailor and improve
the solvent-based spinning process. Several
stages in the spinning process influence the
final (supra-)molecular structure of the cellulose
polymer chains in the fibre. Shear stress in the
spin capillary causes pre-orientation of the
polymers in solution. Thus, each prepared dope
was subjected to a routine shear rheological
characterization. When the liquid filament
enters the air gap, the shear stress is released
instantaneously and the polymer chains tend
to re-assume a random-coil formation which
leads to die-swell. This is counteracted by
the draw acting on the filament. The exerted
elongational stress causes a further orientation
of the cellulose chains.

Figure 14. KS15 (left) and KS42 (right) piston spinning unit.
KS42 KS15
piston diameter 42 mm 15 mm
cylinder volume 500 ml 17 ml
extrusion velocity range 0.4 – 5.0 ml/min 0.007 – 0.06 ml/min
Table 4. Specification of spinning units.
terms of their elastic strength and moduli in
order to determine the weakest point of the
filament in the spin bath. It was shown that the
structure formation can differ markedly due to
different diffusion constants and gel strengths
of different ILs.
New ILs were constantly tested for their suitability
as fibre spinning solvents. In order to process
also small (lab-scale) amounts of IL and thus gain
more flexibility, a small piston spinning unit (KS15)
was integrated into the existing spinning line,
thus enabling full use of all previously installed
equipment (Figure 14). The piston speed was
reduced accordingly to create the same shear
stress conditions generated in the bigger unit
(KS42). The characteristics are summarized in
Table 4.

Main achievements
• [DBNH][OAc] proved an excellent solvent for
cellulose fibre spinning.
• The mechanical properties of Ioncell fibres
are significantly higher than those of other
man-made cellulosic fibres.
• The chemical stability of the cellulose and
mechanical properties of the resulting
fibres clearly exceeded the goals set at the
beginning of the programme. Cellulosic
textile fibres were produced with tensile
strength properties (>50 cN/tex) exceeding
the initial target (≥35 cN/tex).
• Different dissolving pulps were spun
successfully with only minimal difference
in final mechanical properties. Paper grade
pulp was converted into textile fibres with
tenacity values of 48 cN/tex.
• Two demonstration products were
manufactured: a scarf made of eucalyptus
and a dress (in collaboration with
Marimekko) made of Enocell birch dissolving
pulp. The products were presented at the
FIBIC Annual Seminar November 2013 and at
Marimekko’s autumn and winter collection
fashion show (March 2014).
4.5 Functionalization of regenerated
cellulose fibres
The aim of the chemical functionalization
of regenerated and spun fibres task was to
give the fibres more added-value and better
properties, such as reduced fibre fibrillation.
Crosslinking of spun fibres
Due to the high orientation of Tencel (lyocell)
fibres, the fibres have a high tendency for
fibrillation. The degree of fibrillation can be
taken as a direct indication of the abrasion
resistance of the fibre. Results using click
chemistry for crosslinking Tencel reference
fibres to reduce the unwanted fibrillation
tendency of highly oriented fibres showed
promising results similar to those of
commercial triazine crosslinker (Figure 15).
The fibrillation index was reduced from 2.5
to 1.0. Similar positive effects of crosslinking
are also expected with other fibre types.
Fibre crosslinking chemistries based on
the adsorption of pre-modified CMC and
click chemistry and those with commercial
crosslinking agent 2-sodiumhydroxy-4,6-
dichloro-1,3,5-triazine are shown in Figure 16.
Main achievements
• Pre-modified CMCs were irreversibly adsorbed
onto regenerated cellulose fibres. Further
functionalization of cellulose fibres was
demonstrated using click chemistry reaction.
• Fibre fibrillation was reduced using a click-
chemistry based crosslinker.
4.6 Recovery of ionic liquids
Efficient recovery of ILs in IL-based solution
and spinning processes is a prerequisite for
an environmentally and economically feasible
process.
Pressure-driven membrane separations
Different UF (ultrafiltration) and NF
(nanofiltration) membranes were screened
and tested for filtration of IL-water solutions
and for removal of impurities, i.e. dissolved
material, during the process.
Model solution (galactoglucomannan, GGM,
representing carbohydrate impurities in the
spinning bath) filtration tests were conducted in
cross-flow mode with a polymeric ultrafiltration
membrane (GM by GE Waters, USA) and a
ceramic nanofiltration membrane (Inopor®nano
by Inopor® GmbH, Germany) (Figure 17). With
the ceramic membrane the flux was slightly
better than that with the GM membrane and
the model compound retention was also better.
The normalized fluxes were more or less the
same at the beginning of testing (around 13 L/
m2
hbar) but remained higher with the ceramic

Figure 15. Fibrillation indexes of unmodified and crosslinked Tencel (commercial lyocell) fibres after
mechanical abrasion test (ball bearing method).
Figure 16. Chemical crosslinking via triazole ring (click chemistry) (a) and chemical crosslinking via triazine
ring (commercial crosslinker) (b).
a)
b)
R1
=H, CH2
COONa or azide group
R2=H, CH2COONa or alkyne group
O
O
N
N
N
O
O
O
O
O
O
O
nOR2
R2
O
R2
O
R2
O
R2
O
R2O
OR2 HO OR2
HO
HO
HO
HO
HO
O
O
O
O
O
O
O
O
O
O
nOH
OH
OH
OH
OH
OH
OH
HO
HO
HO
HO
HO
HO
O
O
O
O
O
O
O
O
O
O
n
n
O
O
O
O
O
O
R1O
R1
O
OR1
OR1
OR1
R1O
R1O
R1
O
OH
OH
OH
OH
OH
OH
OH
HO
HO
HO
HO
O
O
O
OH
HO
HO
HO
O
O
N
Na
N
OH
O
HO
HO
HO
oup
roup
OO
O
O
nOR2
HOO
O
O
O
n
OH
OH
OH
HO
O
O
O
O
O
O
n
nOR1
OH
OH
OH
HO
HO
O
O
O
O
O
O
O
O
O
n
O
O
O
OH
OH OH
OH
OH
HO
HO
HO
HO
HO
HO
HO
HO
O
O
O
O
O
O
O
O
O
O
O
N
Na Na
N
N N
N
N
n
O
OH
OH
OH
OH
OH
HO
HO
HO
HO
HO
HO
HO
HO

Figure 17. Filtration with the GM UF membrane (cross flow)(a); [emim][OAc]] 20w%/H2
O 80w% + 1 g/L
GGM, 7.5 bar, 25°C, initial feed volume 2.0 L, A = 100 cm2
and cross-flow velocity 1.7 m/s.
Filtration with ceramic NF membrane (Inopore®nano TiO2
) (cross-flow)(b); [emim][OAc]] 20w%/H2
O 80 w%
+ ≈1 g/L GGM, 2.0 bar, 25°C, initial feed volume 2.0 L, A = 660 cm2
and cross-flow velocity 0.3 m.
a)
b)
membrane (10 L/m2
hbar with the ceramic
membrane and 7 L/m2
hbar with the polymeric
membrane, see Figure 17). This may be due to
more fouling of the GM membrane during the
filter test.
It is possible to remove polysaccharides
almost entirely and monosaccharides (glucose)
partially (more than 50%) by NF (nanofiltration)
if the IL concentration is high enough (thus IL
is not retained, i.e., the solution can be purified
from organic contaminants). UF (ultrafiltration)
can be used as a preceding step, although it will
not remove smaller molecules.
The filtration trial with real spinning bath solution
was done with an NF 270 membrane in cross-
flow. Sugars in the spinning bath solution were
retained totally, but the IL was also 90% retained.

Reverse osmosis
The reverse osmosis (RO) process can remove
water from the spinning bath solution up to
about 30 wt%. This limit derives from the
osmotic pressure of the IL/water solution,
which would require even higher pressures
to be used to overcome the osmotic pressure
resistance. In this study the highest operating
pressure was 50 bar. A 15 wt% solution of IL
([DBNH][OAc]) in water would give about 40
bar osmotic pressure, theoretically. Pressures
above 50 bar would remove more water, but
would be economically unfeasible due to
energy consumption. A 30 wt% solution would
give about 80 bar osmotic pressure (at 25°C
and if IL is dissociated completely) which is,
however, already close to the recommended
upper limits for RO membranes.
Metal removal by ion-exchange resins
A simple IEX (ion exchange) test to remove
metals from the model spinning bath solution
gave very positive results. SAC (Strong Acid
Cation) exchange resin removed magnesium
almost totally, whereas WAC (Weak Acid Cation)
exchange resin showed significantly inferior
performance.
Recovery of ionic liquids by pervaporation
Several polymeric membranes were tested
for the recycling of ionic liquids used in the
cellulose dissolution process. Pervaporation
(PV) tests with tri-1,5-diazabicyclo propionate
([DBNH][CO2
Et]) showed that PVA-TiO2
and
PVA-PDMS membranes were able to separate
water from the [DBNH][CO2
Et] / water solution.
Also 1-ethyl-3- methylimidazolium acetate
([emim][OAc]]) showed high selectivity to water
permeation with PVA-TiO2
and PVA-PDMS
membranes. The membranes did not, however,
tolerate mDBN-dimethyl phosphate ([mDBN]
[Me2
PO4
]). In summary:
1) [emim][OAc]]
• PVA-TiO2
and PVA-PDMS membranes
from HZG (Helmholtz-Zentrum Geesthacht)
tolerated [emim][OAc]] and were able
to separate water from the [emim]OAc/
water solution (90 wt%/10 wt%) in the PV
experiments.
2) [mDBN][Me2
PO4
]
• None of the seven studied pervaporation
membranes tolerated [mDBN][Me2
PO4
].
3) [DBNH][CO2Et]
• PVA-TiO2
and PVA-PDMS membranes
from HZG and PERVAP 2255-30 membrane
from Sulzer tolerated [DBNH][CO2
Et] in the
preliminary experiments.
• PVA-TiO2
and PVA-PDMS membranes were
able to separate water from the [DBNH]
[CO2
Et] / water solution (90 wt%/10 wt%) in
the PV experiments.
DIL (distillable IL) recovery concepts
Twelve DIL recovery concepts were identified
together with the research groups. Preliminary
production cost estimates of these concepts
were analysed in six scenarios. In summary,
energy and capital costs dominated the costs
in all concepts, while evaporation of water
made up the bulk of the energy costs. The
lowest production costs were achieved in the
concepts where IL was purified by distillation,
followed by concepts with flotation and ion
exchange based purification.
[DBNH][OAc] hydrolysis and recycling
[DBNH][OAc] as the favoured distillable ionic
liquid for Ioncell fibre spinning has a certain
degreeofhydrolyticinstability.Thisisdependent
on temperature and water content. IL hydrolysis
is faster as the water content decreases and
temperature increases. Successive cycles of
pulp dissolution and regeneration in a batch
reactor were undertaken to determine how

many times the IL can be recycled whilst still
maintaining its cellulose-dissolving capability.
It was determined that under batch dissolution,
regeneration and drying conditions the IL could
be reused seven times until the hydrolysis
product reached a level preventing dissolution
(15 mol% hydrolysis product). As compared
to a dynamic process, the regeneration and
recycling conditions under batch dissolution
conditions allow much longer solvent residence
times in contact with hot surfaces. Therefore,
hydrolysis was greatly increased. Nevertheless,
the hydrolysis product needs to be converted
back to ionic liquid, and this requires a recycling
step. This was demonstrated to be possible
by preparing the hydrolysis product and
converting it back to the IL in the presence
of excess DBN and Amberlyst 15 (superacidic
resin). This enabled recovery of [DBNH][OAc]
(Figure 18), avoiding irreversible decomposition
to the amide decomposition product.
Main achievements
• It was proven that membrane filtration
and ion exchange resin processes can be
used as building blocks for solvent recovery
and recycling in novel cellulose processing
techniques that utilize ILs.
- Nanofiltration (NF) membranes were
effective at separating dissolved organics
from IL-water solutions while not retaining IL
([emim][OAc]).
- Reverse osmosis (RO) was able to remove
some water from IL-water solution ([DBNH]
[OAc]).
- Ion Exchange Resin (IEX) was highly
effective at removing metals from IL-water
solution
• Pervaporation was shown to be a sufficient
method for separating water from IL.
• [DBNH][OAc] can be reused 7 times. A
certain degree of hydrolytic instability is
overcome by a recycling step that converts
the hydrolysis products back to the IL.
• The most promising concept for DIL
(distillable IL) recovery was based on
evaporation and distillation technology.
4.7 Techno-economic modelling of Ioncell
fibres
The objectives of the techno-economic
evaluations were to facilitate communication
between researchers and decision-making
companies, to identify central R&D needs during
the course of the programme, and to provide
recommendations for further research. The
techno-economic modelling of Ioncell-F fibres
analysedtheproductionofcellulosicstablefibres
using dissolution and regeneration technology
based on novel ionic liquids developed at the
University of Helsinki. Lyocell staple fibre was
selected as a reference both for the production
concept and end-use market.
Figure 18. Hydrolysis and reconversion to [DBNH][OAc] under suitable experimental conditions.

Figure 19. Block-flow diagram of the IL-based staple fibre production process
Water
Dissolving Pulp
DC 90%
Fibres
Premixing
Thin film evaporation
Filtering
Spinning
Washing
Finishing
Drying
Evaporation
IL recovery
IL Distillation
Impurities
Makeup IL
The techno-economic modelling compared
two ionic liquid based processes with the
commercial NMMO process (Table 5). SPINCELL
process concept included Ioncell-F process and
ionic liquid recycling process using dissolving
pulpandINTEGRATEDprocessconcepthadkraft
pulp as feedstock and the same ionic liquid to
function both in hemicellulose dissolution and
fibre spinning. NMMO refers to the dissolution
solvent used in the process of producing lyocell
fibres from dissolving pulp feedstock.
All studied cases showed good profitability at
a lyocell staple fibre price above 2 000 EUR/t.
Steam consumption in the SPINCELL process
was about 16% higher than the corresponding
NMMO process. IL consumption in the
INTEGRATED process was somewhat higher
compared to SPINCELL. Optimization of ionic
liquid recovery could lead to lower production
costs via energy and IL savings. A block-flow
diagram of the IL-based stable fibre production
process is illustrated in Figure 19.
Of all the techno-economic modelling cases
studied in the FuBio Cellulose programme,
the Ioncell fibres received the highest scores
in the qualitative opportunity assessment.
Process concept Feedstock Solvent
SPINCELL Dissolving pulp [DBNH][OAc]
INTEGRATED Kraft pulp [DBNH][OAc]
NMMO Dissolving pulp NMMO
Table 5. Studied process concepts.

Market opportunities were seen as promising,
with a large global market, increasing demand
for man-made cellulosic fibres and excellent
product qualities recorded. However, technical
feasibility still remains a question with
several uncertainties related to ionic liquid
performance, recyclability and availability.
The strengths and weaknesses of Ioncell fibre
production are summarized in a SWOT analysis
in Figure 20.
Based on the techno-economic assessment, the
key recommendations for further Ioncell fibre
research are related to ionic liquid preparation
and recycling and the impact of various
impurities on the spinning process. In developing
novel ionic liquids, researchers should take into
account the industrial-scale availability and
price of applied reagents. If the reagents are not
commercially available, the complexity of the
chemical synthesis to produce such chemicals
should be evaluated. Whether the same
ionic liquid is suitable for both hemicellulose
dissolution and fibre spinning is likely to define
whether the fibre producer should use kraft
or dissolving pulp as feedstock. The type and
efficiency of the recycling concept has a great
impact on total production costs, and thus
should be studied as a top priority. Similarly, the
maximum concentration of impurities in ionic
liquids should be analysed in order to define
the required level of IL purification, and thus
purification costs.
5. Exploitation plan and impact of
the results
Finland has a long tradition in the pulp and
paper industry. The importance of the wood
sector to the Finnish economy is reflected in the
multitude of university faculties that belong to
the leading institutes of wood technology and
chemistry worldwide. Industrial and academic
cooperation in the Finnish wood sector is truly
outstanding, providing the perfect foundation
for cutting-edge research and innovative
products. With the pressure of globalization
Figure 20. SWOT analysis of textile fibre production via ionic liquids.
Helpful to achieving business success Harmful to achieving business success
Process
related
STRENGTHS
• Excellent fibre product quality
• Large market and growing demand of man-
made cellulosic fibres
WEAKNESSES
• Process concept is not yet demonstrated
Ionic liquid performance and availability
still on a vague basis
• Ionic liquid recyclability remains a
question
• No clear cost advantages foreseen
compared to commercial NMMO process
Business
environment
related
OPPORTUNITIES
• Promising production economics
• Political support for env. sustainable and safe
fibres and production processes is very strong
• Integration to pulp mill seems feasible with
mutual benefits
• Flexible product line with opportunities for
various raw materials and product specs.
• Ionic liquid screening is still at early stage. Even
more promising solvents may be found.
THREATS
• Price competitiveness against lyocell in
niche applications and against viscose,
cotton and other fibres in bulk textile
end-uses
• Market entry may be difficult due to close
market with only few large players

and the accompanying shift of big production
sites towards the southern hemisphere
(especially South America, Indonesia and
China), the need for new specialized products
is urgent. The market for cellulosic fibres is
growing fast due to rapidly industrializing
nations, offering new possibilities for Finnish
wood-based industry. The knowledge created
by this project opens new opportunities to
utilize a broad spectrum of different grade
pulps for fibre spinning, enabling raw material
costs to be saved and energy consumption in
subsequent process steps to be reduced. This
increases the economic feasibility and reduces
the environmental impact of the entire process.
Finnish companies are well positioned to take
a leading role in the biorefinery concept. Stora
Enso has already taken a lead by starting
dissolving pulp production at its Enocell plant
in Uimaharju.
The successful production of textile fibres
from local biomass resources has attracted the
interest of Finnish textile and design companies
and has already led to close collaboration with
the leading Marimekko brand.
The developed generic methodology for the
chemical functionalization of regenerated
fibres showed potential, and is expected to gain
the attention of industrial viscose producers.
However, the method has not yet reached the
exploitation stage and further work is needed to
gather additional data and to understand the full
potential and feasibility of the new technique.
The ability to use membranes in ILs recovery
is important when considering the use of ILs in
cellulose dissolution processes. Pervaporation
offers a potential energy-efficient separation,
purification and recovery technology for
biorefinery that can bring financial savings and
competitiveness to producers. This finding is
also of benefit to other industries using ILs as
solvents.
6. Networking
The research was carried out jointly by
Aalto University, Technical University of
Lappeenranta, University of Helsinki, University
of Oulu and VTT. Table 6 presents both the
research partners and industrial partners and
their roles in the programme.
Aalto University has initiated collaboration
with the Department of Chemical Engineering,
University of Porto, in elongational viscosity of
cellulose-IL solutions. The University of Helsinki
has collaborated with the University of Santiago
in the development of analytics related to
recyclable ionic liquids. The University of Oulu
has collaborated with the Membrane Separation
Processes Group of the Chemistry Faculty at
Nicolaus Copernicus University (Torún, Poland)
in the form of researcher exchanges in the area
of pervaporation theory and research.

Partner Role
Aalto University
- Forest Products Technology (FPT)
- Biotechnology and Chemical Technology (BCT)
FPT: Pulp analyses, rheological characterizations
of spinning dopes. IL fibre spinning. Pulp and fibre
modification.
BCT: Modelling of ionic liquid hydrolysis kinetics
Glocell Quantitative economic modelling
- Separation Technology
Membrane filtrations
Metsä Fibre Industrial tutor. Defining, steering and providing com-
petence for the modelling. Providing industrial view
insight to techno-economic assessments
Pöyry Management Consulting Market study. Economic feasibility modelling. Busi-
ness potential evaluation
Stora Enso Industrial tutor. Providing industrial insight to techno-
economic assessments
- Organic Chemistry
• Preparation of large-scale ionic liquid samples (for
Aalto spinning trials, for UO for pervaporation and
VLE studies, for LUT membrane purification studies)
• Chemical modifications of cellulose in ILs
• Designing optimal dope modification procedures in
cooperation with Aalto
• Understanding IL recyclability in cooperation with
Aalto and VTT
University of Oulu
- Mass and Heat Transfer Process
• Pervaporation studies
• Role of pervaporation in cellulose dissolution and
regeneration processes
UPM-Kymmene Industrial tutor. Providing industrial insight to techno-
VTT • Molar mass analyses
• Techno-economic screening of DIL (distillable ionic
liquid) recovery concepts
• Process modelling, ionic liquid based processes
Table 6. Partner organizations and their roles.

7. Publications and reports
Publications:
García, V., Valkama, H., Sliz, R., King, A.,
Myllylä, R., Kilpeläinen, I., Riitta L. and Keiski,
R.L. Pervaporation recovery of [AMIM]Cl during
wood dissolution; effect of [AMIM]Cl properties
on the membrane performance, Journal of
Membrane Science, 2013, Vol. 444:9-15.
Hauru, L. K. J., Hummel, M., Michud, A. and
Sixta, H. Dry jet-wet spinning of strong cellulose
filaments from ionic liquid solution. Cellulose,
2014. DOI: 10.1007/s10570-014-0414-0
Hummel, M., Michud, A., Tanttu, M., Asaadi,
S., Ma, Y., Hauru, L. K. J., Parviainen, A.,
King, A. W. T., Kilpeläinen, I. and Sixta, H.
Ionic liquids for the production of man-made
cellulosic fibres – opportunities and challenges.
Advances in Polymer Science, 2014, accepted.
Stépán, A. M., King, A. W. T., Kakko, T., Toriz,
G., Kilpeläinen, I. and Gatenholm, P. Fast and
highly efficient acetylation of xylans in ionic
liquid systems. 2013, Cellulose, 20 2813-2824.
Conference proceedings
Asaadi, S., Michud, A., Hummel, M. and Sixta,
H. High tenacity cellulosic fibres from novel
ionic liquid-cellulose solution by dry-jet wet
spinning. Proceedings of the 13th European
Workshop on Lignocellulosics and Pulp, Seville,
Spain, 24th-27th June 2014.
Hauru, L. K. J., Hummel, M. and Sixta, H. Fibre
spinning from ionic liquid dope. Proceedings of
the 12th European Workshop on Lignocellulosics
and Pulp, Espoo, Finland, 27th-30th August
2012, pp. 272-275
Hauru, L. K. J., Hummel, M., Michud, A.,
Asaadi, S. and Sixta H. High-strength (870
mpa) cellulose filament spun from ionic liquid.
Proceedings of the 7th Aachen-Dresden
International Textile Conference, Aachen,
Germany, 28th-29th November 2013.
Hummel, M., Hauru, L. K. J., Michud, A.
and Sixta, H. Mechanistic studies on the
regeneration of cellulose from ionic liquid
solutions. Proceedings of the 12th European
Workshop on Lignocellulosics and Pulp, Espoo,
Finland, 27th-30th August 2012, pp. 284-287.
Michud, A., Arnoul-Jarriault, B., Hummel, M.
and Sixta, H. Influence of molecular mass
distribution on the rheological behaviour of
cellulose/ionic liquid solutions during dry-jet
wet spinning process. Proceedings of the 12th
European Workshop on Lignocellulosics and
Pulp, Espoo, Finland, 27th-30th August 2012, pp.
134-137.
Michud, A., Hummel, M. and Sixta, H. Influence
of molar mass distribution on the final properties
of regenerated fibres from cellulose dissolved in
ionic liquid by dry-jet wet spinning. Proceedings of
the 13th European Workshop on Lignocellulosics
and Pulp, Seville, Spain, 24th-27th June 2014.
Presentations:
Hummel, M., Michud, A. and Sixta, H.
Structure formation of cellulosic material upon
regeneration from ionic liquid solutions. (Oral
presentation at the 243rd American Chemical
Society (ACS) meeting, March 2012, San Diego,
USA).

Hummel, M., Michud, A., Hauru, L. K. J. and
Sixta, H. Ionic liquids as powerful tool to
exploit renewable biomass. (Oral presentation
at the Technoport conference, April 2012,
Trondheim, NO).
Solution state of cellulose in ionic liquids. (Oral
presentation at the 245th American Chemical
Society (ACS) meeting, March 2013, New
Orleans, USA).
Hummel, M., Michud, A., Asaadi, S., Tanttu, M.
and Sixta, H. High tenacity cellulosic fibres via
ionic liquid processing. (Oral presentation at the
FuBio open seminar day, August 2013, Helsinki,
Finland).
Hummel, M., Michud, A., Tanttu, M., Asaadi,
S., Ma, Y., Hauru, L. K. J., Hartikainen, E. and
Sixta, H. Rheological aspects of ionic liquid
based fibre spinning. (Oral presentation for
the Finnish section of the Nordic Rheological
Society, March 2014, Espoo, Finland).
Hummel, M., Michud, A., Roselli, A., Tanttu, M.,
Asaadi, S., Ma, Y., Hauru, L. K. J., Hartikainen,
E. and Sixta, H. Ioncell: From pulp to high-
performance fibres via ionic liquids. (Oral
presentation at the “Journée Scientifique des
GDRs LIPs et Biomatpro: Liquides ioniques et
polymères biosourcés“, April 2014, Sophia-
Antipolis, France).
Michud, A., Hummel, M. and Sixta, H. Dry-
jet wet spinning of cellulose/ionic liquid (IL)
solutions. (Oral presentation at the 245th
American Chemical Society (ACS) meeting,
March 2013, New Orleans, USA).
Michud, A., Hummel, M., Tanttu, M., Ma, Y.,
Asaadi, S. and Sixta, H. IONCELL-F Ionic Liquid
based Fibre Spinning. (Oral presentation at the
FuBio seminar, April 2014, Espoo, Finland).
Mänttäri, M., Keiski, R., Pihlajamäki, A., Nakari,
O., Valkama, H. and Turkki, A., Recovery of
Ionic Liquid by Hybrid Membrane Process (oral),
FuBio Cellulose Seminar: Cellulose activation,
dissolution and fibre regeneration, Åbo
Akademi, 11th March 2013, Turku.
Sixta, H., Hummel, M., Michud, A., Hauru,
L. K. J., Roselli, A., King, A., Kilpeläinen,
I., Froschauer, C. and Schottenberger, H.
Progress in Processing Lignocellulose with
Ionic Liquids. (Oral presentation at The 3rd
International Cellulose Conference, October
2012, Sapporo, Japan).
Sixta, H. Progress in processing lignocellulose
with ionic liquids. Invited lecture at the
University of Leipzig, Institute of Chemical
Technology, July 2013, Leipzig, Germany.
Sixta, H. Progress in processing lignocellulose
with ionic liquids. (Oral presentation at the
FuBio JR2 WP2 summer seminar, August 2013,
Helsinki, Finland).
Sixta, H., Hummel, M., Roselli, A., Asaadi,
S., Hauru, L. and Tanttu, M. Processing
lignocellulosic materials in ionic liquids. (Oral
presentation at the 3rd EPNOE International
Polysaccharide Conference, Nice, France, 21th-
24th October 2013).
Posters:
Extensional rheology of cellulose-ionic liquid
solutions. Extended abstract and poster at the
Nordic Rheology Conference, 6th-8th June 2011,
Helsinki, Finland.

Hummel, M., Michud, A., Hauru, L. K. J. and
Sixta, H. Applicability of various ionic liquids
in dry-jet wet spinning of cellulose solutions
(poster presentation at the 5th International
conference on Ionic Liquids, April 2013,
Vilamoura, Portugal).
Michud, A., Hauru, L. K. J., Hummel M. and
Sixta H. Dry-jet wet spinning of cellulose-
ionic liquid solutions (poster presentation at
the FuBio Cellulose seminar, June 2012, Espoo,
Finland).
Michud, A., Parviainen, A., Hauru, L. K.
J., Mutikainen, I., Kilpeläinen, I., Sixta, H.,
Hummel, M. and King A. W. T. Tailored ionic
liquids for dry jet wet spinning of cellulose
solutions (poster presentation at the FuBio
Cellulose seminar, August 2013, Helsinki,
Finland).
Michud, A. and Rissanen, M. From cellulose
to textile fiber and a ready products (poster
presentation at the SHOK summit seminar, May
2014, Helsinki, Finland).
Nakari, O., Pihlajamäki, A. and Mänttäri, M.
Membranes for Recovery of Water-Ionic Liquid
Solutions (poster), Fubio Cellulose internal
seminar, 12th June 2012, Innopoli 1, Otaniemi,
Espoo.
Nakari, O., Pihlajamäki, A. and Mänttäri, M.
Membranes for Recovery of Water-Ionic Liquid
Solutions (poster), Fubio Cellulose and JR2 joint
seminar, 1st October 2012, Innopoli 2, Otaniemi,
Espoo.
Tanttu, M., Michud, A., Asaadi, S., Ma, Y.,
Hummel, M. and Sixta, H. Textile application
of cellulosic fibres from ionic liquid solution
(poster presentation at the FuBio Cellulose
Textile Company Workshop 31st October 2013,
Espoo, Finland).
Valkama, H., Niemistö, J. and Keiski, R. L.
“Pervaporation in ionic liquid’s recovery: Effect
of 1-Ethyl-3-methylimidazolium acetate on
permeability properties of hydrophilic polymeric
membranes” (poster presentation at the
FuBio Cellulose Programme Internal Seminar
12.6.2012, at the FuBio Programme Seminar
1.10.2012 and at the XXIX EMS Summer School
on Membranes in Nancy, France, 11.7.2012; the
abstract was also published in the abstract
book of the XXIX EMS Summer School 2012).
Theses:
Benoît Arnoul-Jarriault. Influence of the
molecular weight distribution of cellulose on the
rheological properties of cellulose-ionic liquid
solutions (Master’s thesis, 2012, Aalto Univeristy)
Hartikainen, Eeva. Solution state of cellulose
in ionic liquids – A rheological study (Master’s
thesis, 2013, Aalto University).
Saastamoinen, Jouni. Influence of the
solute’s molecular weight distribution on the
spinnability of cellulose-ionic liquid solutions
(Master’s thesis, 2011, Aalto University)
Selg, Christoph. New amidinium, imidazolium
and phosphonium ionic liquids for cellulose
dissolution and modification, (Master’s thesis,
2012, University of Helsinki)

WATER-BASED
DISSOLUTION
AND
REGENERATION
PROCESSES
CONTAC T PE RSON
Marja Rissanen, marja.rissanen@tut.fi
PAR TN E RS
Aalto University
Metsä Fibre
Suominen
Stora Enso
Tampere University of Technology
University of Oulu
UPM-Kymmene

ABSTRACT
The objective was to develop novel, sustainable water-based dissolution and regeneration
processes for the production of a cellulosic staple fibre, and to demonstrate the regenerated fibres
in textile and nonwoven structures. The chosen water-based process was the Biocelsol process.
The dissolution factors involved in the Biocelsol process were studied to generate new knowledge
to support the development of novel up-scalable pre-treatment and dissolution processes. The
studies showed that both mechanical and enzymatic treatments are needed in order to obtain a
spinnable dope. The mechanical treatment opened up the pulp fibre matrix to enzymes, while the
enzymatic treatment reduced the molecular mass of the cellulose for dissolution in sodium zincate.
The novel combined mechanical shredding and enzyme pre-treatment developed decreased
the treatment time and enzyme dosage significantly. Based on the pre-treatment studies, it is
expected that the dissolution process can be up-scaled to the industrial scale.
The regeneration studies focused on trials of spin dope and spin bath additives for controlling
the regeneration process. The maximum tenacity of the novel Biocelsol fibres (19 cN/tex) was
achieved by using both spin dope and spin bath additives. This value was slightly lower compared
to commercial viscose (22 cN/tex), but further optimization of the spin bath could increase the
tenacity value. Enzyme recycling and removal of oligosaccharides from the pre-treatment filtrates
as well as separation of acid and salts from the spin bath were demonstrated by nanofiltration. Two
demonstration products, i.e. spunlaced non-woven sheets and a knitted hat, were manufactured
from Domsjö softwood dissolving pulp. The processing properties of novel Biocelsol fibres were
comparable to commercial viscose fibres.
The novel chemical pre- and post-modification methods were demonstrated to achieve improved
solubility and/or water absorption of regenerated fibres. The fibres regenerated from the pre-
modified (butylated) pulp had slightly lower tenacity (15 cN/tex) compared to fibres from unmodified
pulp. Water absorption, measured in terms of swelling coefficient, was 300% for the butylated
fibres, which was significantly higher than both the unmodified Biocelsol fibres (170%) and
commercial viscose fibres (100%). Post-modification of the regenerated fibres further improved
the swelling coefficient to 500-1200%.
In addition, the cellulose and hemicellulose molecular weight distribution was modelled as a
function of process conditions in order to optimize conventional viscose fibre production.
Keywords:
Biocelsol, modification of cellulose, nanofiltration, pre-treatment of cellulose, spunlaced
nonwovens, modelling of viscose process

1. Work background
Global annual consumption of textile fibres
reached 90 million tonnes in 2012 and is
increasing continually along with population
growth and rising living standards. Cellulose-
based textile fibres (natural and man-made)
have a comfortable texture and high moisture
absorption and are thus used mainly in apparel
and home textiles. Cotton is the main natural
cellulose textile fibre, with an annual production
yield of between 20–27 million tonnes. The total
global area dedicated to cotton cultivation has
barely changed for 90 years, yet yields have
tripled during this period. This has been achieved
through intensive consumption of irrigation
water, chemical pesticides, insecticides, and
fertilizers, and at no small ecological cost. Man-
made cellulosic fibres (MMCF), such as viscose,
modal, cupro, lyocell, acetate, and triacetate,
are made from cellulose dissolving pulp. MMCF
production stood at about 5 million tonnes in
2012, the majority of which was viscose fibre.
Man-made cellulosic staple fibre production
has increased markedly during the past 10
years at an average annual growth rate of 7.5%,
compared to 3.2% for synthetic staple fibres.
Total textile fibre production has increased
5.4% during the same period.
The viscose process was invented as early as
1892. In the process, dissolving pulp is first
treated with caustic soda, then with carbon
disulphide (CS2
), and dissolved in diluted caustic
soda.Thecellulosesolutionisthenspunusingthe
wet spinning method. The process is relatively
lengthy and cellulose undergoes degradation
reactions during the treatment process. The final
fibre quality is strongly dependent on the degree
of polymerization (DP) of cellulose, the degree of
substitution and the by-products present in the
viscose solution. Models developed in the FuBio
Cellulose programme for different viscose steps
can be used for optimization of product quality
or for determining optimal process conditions
for specific product grades.
The viscose process also consumes large
amounts of water and chemicals, of which CS2
is extremely volatile and highly toxic. One of
the most promising sustainable water-based
processes for the manufacture of MMCF is
the Biocelsol process (WO 2009/135875 A1)
developed at Tampere University of Technology.
In this process, chemical pulp is pre-treated
mechanically and enzymatically and then
dissolved directly in sodium zincate (NaOH/
ZnO) solution using a freezing/thawing cycle.
The solution is then regenerated into fibres
using the wet spinning method. The benefits
of the Biocelsol process are the lack of CS2
in
the process and the possibility to use existing
viscose fibre plants for the manufacture of
regenerated fibres.
In order to further develop the Biocelsol
process, deep understanding of the starting
material structure and cellulose modification
and dissolution were needed. Development
and up-scaling of the pre-treatment processes
were also essential in order to decrease
the energy demand and to minimize water
consumption. The coagulation conditions were
adjusted to enhance fibre properties such
as tenacity. Modification tools needed to be
developed in order to modify key regenerated
fibre properties, such as water absorption and
holding capacity for nonwoven applications.
2. Objectives
The main objectives were to develop novel,
sustainable water-based dissolution and
regeneration processes for the production of
cellulosic staple fibres, and to demonstrate
the regenerated fibres in textile and nonwoven
structures. Examination of dissolution factors
involved in the aqueous process aimed to
generate basic understanding of the dissolution
process used in the development of pre-

treatment processes. The modifications, both
pre- and post-modifications, aimed to improve
cellulose dissolution and regeneration as well
as the properties of regenerated fibres, such
as increased water uptake. The cellulose and
hemicellulose molecular weight distribution
modelling as a function of process conditions
was aimed at optimizing conventional viscose
fibre production.
The overall approach was to manufacture textile
products and nonwovens from wood via the
production of chemical pulp, pulp dissolution
and regeneration using the novel Biocelsol
method, and subsequent production of textiles
and nonwovens, as summarized in Figure 1.
The chemical pulp used was Domsjö softwood
sulphite pulp (spruce-pine, viscosity 520 ml/g).
The dissolution factors were studied to support
the development of up-scalable pre-treatment
and dissolution processes for use in the novel
Biocelsol process. The treatment time of
shredding and enzymatic hydrolysis as well
as the enzyme concentration were varied to
evaluate the structural changes in the cellulose
pulp fibre. The pore structure of untreated and
differently treated pulp was studied by means
of solute exclusion, thermoporosimetry, and
different NMR methods. Crystal structures
were evaluated by 13C
CPMAS NMR and wide-
angle X-ray scattering (WAXS) measurements.
The nanoscale structure was examined with
small-angle X-ray scattering (SAXS). X-ray
microtomography was used to observe
structural changes at the micrometre scale.
A novel pulp pre-treatment was developed to
improve alkaline solubility without extensive
reduction in chain length, and to reduce energy
and water consumption. The state-of-art
Biocelsol treatment was used as a starting point.
Several up-scalable mechanical treatments and
enzyme preparations and their combinations
were screened and the applicability of the novel
treatments was evaluated by means of solubility
tests. The solutions were characterized by
optical microscope and by measuring ball
drop viscosity, the quantity of dissolved and
undissolved content, and total cellulose content.
The dissolution process during the freezing/
thawing cycle was examined using an optical
microscope equipped with a cooling stage.
Spinning trials ranging from small scale (5 kg
spinning dope) to large scale (60 kg spinning
dope) were carried out. Factors affecting
Figure 1. The production chain – from wood to textiles.

filterability and degassing of the solution
were studied. The effect of drawing and
other spinning parameters were evaluated.
In addition, the effect of spin dope additives
and the content of the spin bath on the
spinning process and on the fibre properties
were studied. The linear density, mechanical
properties, swelling coefficient, and fibre
morphology were characterized from the
regenerated fibres.
In order to obtain an efficient process system,
the purification and circulation of pre-treatment
process water and the spin bath was evaluated.
Purification of the enzymatic pre-treatment
filtrate from sugars, salts and dissolved
oligosaccharides and separation of acids and
salts from the spin bath were studied by means
of nanofiltration tests.
Chemical pre-modifications of cellulose with
different substituents were studied to improve
the solubility of cellulose. Additionally, the
pre-modification formed tags for a post-
modification step (grafting, crosslinking, etc.).
The spinning properties of pre-modified pulps
were evaluated. TEMPO oxidation as pre-
and post-modification was used to improve
solubility and to create charged groups for
better water absorption. In addition, post-
modification routes, such as grafting and
click-chemistry, were used to improve the
water absorption properties or to functionalize
regenerated fibres.
The Biocelsol fibres were demonstrated in
spunlaced nonwovens and a textile product.
Non-woven sheets were manufactured in lab-
scale pilots from the state-of-art, novel and
modified (with acrylic acid) Biocelsol fibres. The
fibres were carded and the webs hydroentangled
with high-pressure water jets. The processing
properties of the fibres and the water absorption
and tensile properties of non-woven sheets
were compared to commercial viscose fibres.
Finally, a textile demonstration product was
manufactured from the novel Biocelsol fibres.
In order to develop physico-chemical models
for the different steps in the viscose process,
detailed information about related reaction
chemistry, side reactions and component
properties was collected. Modelling of
xanthation, ripening, and dissolution was
carried out. A population balance based method
was used for prediction of the molecular weight
distribution of cellulose during alkali cellulose
ageing. A high-order numerical method capable
of extremely accurate prediction of the integral
properties of the distribution was applied.
The method was also capable of predicting
the actual distribution shape even in complex
states, such as multimodal distributions.
Several scission rate models were proposed
and evaluated against the experimental data.
4. Results
4.1 Development of a novel water-based
cellulose dissolution process
The objective was to develop a novel,
sustainable and techno-economically feasible
water-based dissolution process for wood pulp
cellulose. Dissolution factors were studied
to generate new knowledge to support the
development of pre-treatment and dissolution
processes. The focus was on understanding the
changes in the cellulose fine structure occurring
during mechanical treatment (shredding) and
enzymatic hydrolysis, and understanding the
effect of process chemistry, including additives
and physical parameters, on dissolution.
The Biocelsol process was used as a model
dissolution system.
Dissolution factors
The experimental data showed that the
mechanical treatment opened the pulp fibre

matrix not linearly but stepwise according to
the treatment time. The cellulose structure
broke down or collapsed after a certain amount
of stress (mechanical shredding). The molar
mass and viscosity results showed that a
longer mechanical treatment time increased
the susceptibility of the fibres to subsequent
enzymatic hydrolysis. The solute exclusion
method, used for determination of micro and
macro pores in the fibre wall, showed that the
mechanical treatment has a dominant effect on
fibre swelling. The results were confirmed with
X-ray diffraction, NMR-cryoporosimetry and
thermoporosimetry analysis. The SAXS results
showed a slight loosening of the microfibril
bundles during the mechanical treatment.
According to the WAXS and solid-state NMR
spectroscopy results, the duration of mechanical
treatment did not affect the cellulose crystal
size nor the crystallinity of the samples. The
crystallinity of differently pre-treated cellulose
samples varied between 60-62%, whereas
the crystallinity of untreated pulp was 54%.
The slightly lower crystallinity of untreated
pulp might be due to the dissolution of some
amorphous material during soaking of the pulp
sheets before mechanical shredding.
The aim of enzymatic treatment of
mechanically treated pulp was to decrease the
pulp weight average molecular mass (Mw) and
thus viscosity to a targeted level (e.g. from 520
ml/g to 360 ml/g in 30 minutes). The enzymatic
treatment was found to first rapidly result in
cellulose chain cleavage, leading to decreased
viscosity. Prolonging the treatment time did
not cause further reduction in chain length or
viscosity but merely increased the amount of
carbohydrates dissolved from the pulp. Based
on this result the amount of accessible sites for
enzyme seems to be the limiting factor for the
degree of enzymatic modification. No effect
of enzymatic treatment on the crystal size or
crystallinity of cellulose was observed by the
solid-state NMR spectroscopy studies.
The molar mass and pore size of pre-treated
pulp were found to be the limiting factors for
dissolution in the Biocelsol system. Analyses
of the soluble and insoluble fractions showed
that the soluble fractions had lower molar
mass compared to the original pulp, whereas
the insoluble fraction had the higher molar
mass fraction. The solute exclusion method
and thermoporosimetry studies showed
that the mechanical treatment opened the
structure up for enzymatic treatment, and a
higher enzyme dosage resulted in larger pores
in the pulp. The larger pores most likely allowed
NaOH/ZnO to penetrate more efficiently into
the pulp fibre, thus increasing the quality
of the solution. The relationships between
molar mass and viscosity and solubility in the
Biocelsol systems are shown in Figure 2.
Figure 2. Effect of molecular mass (Mw) on viscosity of cellulose solution (alkaline solubility of the samples
shown as %) (left) and on solubility (right). MX = time of mechanical shredding before enzyme treatment.
E0, E1=commercial enzyme preparations used for treatment.
Figure
2.
Effect
of
a)
Mw
on
viscosity b)
on
solubiltiy
Fb-‐visc. Solub.
%
Mw E1,
1mg/g E1,
0.25
mg/g Mw E0 E1,
1mg/g E1,
0.25
mg/g
405000 405000 41
368431 368431 66
295540 2908 295540 97.347161
263724 484 263724 98.87594
230572 172 230572 99.378971
212931 60 212931 100
404518.5 404518.5 46.300552
301632 301632 49.943442
307629.5 307629.5 58.742995
214167 214167 55.64942
191074 436 191074 88.260604
0

500

1000

1500

2000

2500

3000

3500

0
100000
200000
300000
400000
500000

Drop
ball
viscosity,
s/20cm

Mw,
g/mol

E1,
1mg/g

E1,
0.25
mg/g

99%

88%
98%

97%

0

20

40

60

80

100

120

0
100000
200000
300000
400000
500000

Solubility,
%

Mw,
g/mol

E0

E1,
1mg/g

E1,
0.25
mg/g

M0
M1

M0.5

M0

M2.5

M5

M5

M2.5

M1

M0.5

Figure
2.
Effect
of
a)
Mw
on
viscosity b)
on
solubiltiy
Fb-‐visc. Solub.
%
Mw E1,
1mg/g E1,
0.25
mg/g Mw E0 E1,
1mg/g E1,
0.25
mg/g
405000 405000 41
368431 368431 66
295540 2908 295540 97.347161
263724 484 263724 98.87594
230572 172 230572 99.378971
212931 60 212931 100
404518.5 404518.5 46.300552
301632 301632 49.943442
307629.5 307629.5 58.742995
214167 214167 55.64942
191074 436 191074 88.260604
0

500

1000

1500

2000

2500

3000

3500

0
100000
200000
300000
400000
500000

Drop
ball
viscosity,
s/20cm

Mw,
g/mol

E1,
1mg/g

E1,
0.25
mg/g

99%

88%
98%

97%

0

20

40

60

80

100

120

0
100000
200000
300000
400000
500000

Solubility,
%

Mw,
g/mol

E0

E1,
1mg/g

E1,
0.25
mg/g

M0
M1

M0.5

M0

M2.5

M5

M5

M2.5

M1

M0.5

TheBiocelsolsystemrequiresafreezing/thawing
cycle during dissolution. The mechanically and
enzymatically pre-treated pulp fibres dissolved
through fragmentation without ballooning.
Conversely, the untreated and only mechanically
treated samples dissolved mainly by swelling
and ballooning. Dissolution of the pre-treated
samples started when the temperature reached
0 °C. The untreated pulp started to dissolve at
a much lower temperature (-15 °C) compared
to the mechanically and enzymatically treated
samples, as shown in Figure 3. The most
important process parameters in the freezing/
thawing cycle were the cooling rate, the lowest
temperature of sample, and the duration of the
sample at low temperature.
Pre-treatment process development
Several up-scalable, cost-effective mechanical
methods and commercial enzyme preparations
and their combinations were screened. For
example, a high-intensity homogenizator,
horizontal agitated laboratory pearl mill,
planetary ball mill, and ultrasound treatment
were trialled for mechanical treatment, but
none proved sufficiently effective.
Two novel mechanical treatment devices and
two commercial enzyme preparations were
found to have the most potential for the novel
pre-treatments. The solubility of the novel pre-
treated pulp was better compared to the solubility
of state-of-art treated pulp, as shown in Figure
4. In addition, the novel pre-treatment enabled a
significant decrease in enzyme dosage (0.25 vs. 1
mg/g), higher treatment capacity (200 g/20 min
vs. 70 g/h.), and higher cellulose concentration in
the solution (7wt% vs. 5.5wt%).
The pre-treatment consistency had no clear
effect on dissolution in NaOH/ZnO. However,
if mixing at high consistency could be carried
out at higher speed, the resulting fibre-fibre
interactions might achieve better results. The
best pre-treatment process developed was up-
scaled to 200 g/20 min and the sodium zincate
dissolution process was up-scaled to 60 kg
dope. It is now expected that both processes
can further be up-scaled to the industrial scale.
Preliminary studies of enzyme recycling
showed that some of the enzyme can be
recycled. Two-stage filtration (NF270 and NF90
Figure 3. Microscopy images during dissolution of differently pre-treated pulp in NaOH/ZnO a) untreated, b)
mechanically shredded (5h) and c) mechanically and enzyme treated. Upper row = 0 °C, lower row = -20 °C.

membranes) of the process water from the
enzymatic treatment significantly improved the
removal of dissolved organics and salts. The
treated water had significantly higher purity,
and fouling of the tighter NF90 membrane was
remarkably reduced. The total organic content
of the process water was about 8.2 ppm after
the two-stage NF process (unfiltered process
water over 1400 ppm). The conductivity was
6.2 μS/cm (at 25 °C) (unfiltered process water
52 μS/cm).
4.2 Regeneration of cellulosic fibres
The objective was to develop a commercially
viable cellulose staple fibre spinning process
for a water-based cellulose solution, and to
demonstrate the advanced Biocelsol fibres
(prepared using the developed novel pre-
treatment process) in a textile product.
Regeneration of novel Biocelsol dopes
The regeneration or coagulation of cellulose
should proceed through the gelling process.
If the coagulation process is too fast, the
cellulose solidifies too quickly preventing the
high draw ratios of as-spun fibres and resulting
in low fibre tenacity. Spin dope and spin bath
additives were tested for their effect on slowing
coagulation and increasing the draw ratio.
The combined effect of spin dope and spin bath
additives proved most effective for increasing
fibre tenacity, as shown in Figure 5. A clear
effect on fibre properties was also obtained
when the degree of filtration was altered.
Figure 4. Drop ball time of the alkaline solution as a function of cellulose concentration of the solution
(left), and SCAN viscosity of the pulp (right).
Figure
4
Drop
ball
time
a) b)
Kuviin
muutettu
extruder
-‐>
novel
ja
Baker
Perkins
-‐>
state-‐of-‐art
matalin matalin
alfa,
% kuula,
s/20cm Liuokset alfa,
% kuula,
s/20cm
Novel
mechanical
tr.
SCAN
<220State-‐of-‐art
mechanical
tr.
SCAN
240-‐250Novel
mechanical
tr.
SCAN
250-‐260 SCAN,
ml/g Koodi Novel
mechanical
tr.State-‐of-‐art
mechanical
tr.
6.05 32 200
FBC-‐
TUT254b 6.05 32
6.13 56 250
FBC-‐
TUT268 6.13 56
6.3 44 170
FBC-‐
TUT269 6.15 24
6.49 56 260
FBC-‐
TUT270 6.26 88
6.8 72 220
FBC-‐
TUT271 6.17 48
7 112 250
FBC-‐
TUT133 6.16 128
6.15 24
6.26 88
6.17 48
6.14 32
6.59 60
6.94 116
6.16 128
5.7 108
5.46 40
0

20

40

60

80

100

120

140

100
150
200
250
300

Drop
ball
)me,
s/20cm

SCAN
viscosity,
ml/g

Novel
mechanical
tr.

State-‐of-‐art
mechanical

tr.

1x
with
0.25
mg/g
E1,

6.26%

1x
+
(1x
with
0.25
mg/g
E1)

6.13%

1mg/g
E1,
6.16%

0

20

40

60

80

100

120

140

4.5
5
5.5
6
6.5
7
7.5

Drop
ball
)me,
s/20cm

Cellulose
concentra)on,
%

Novel
mechanical
tr.

SCAN
<220

Novel
mechanical
tr.

SCAN
250-‐260

State-‐of-‐art
mechanical

tr.
SCAN
240-‐250

Figure
4
Drop
ball
time
a) b)
Kuviin
muutettu
extruder
-‐>
novel
ja
Baker
Perkins
-‐>
state-‐of-‐art
matalin matalin
alfa,
% kuula,
s/20cm Liuokset alfa,
% kuula,
s/20cm
Novel
mechanical
tr.
SCAN
<220State-‐of-‐art
mechanical
tr.
SCAN
240-‐250Novel
mechanical
tr.
SCAN
250-‐260 SCAN,
ml/g Koodi Novel
mechanical
tr.State-‐of-‐art
mechanical
tr.
6.05 32 200
FBC-‐
TUT254b 6.05 32
6.13 56 250
FBC-‐
TUT268 6.13 56
6.3 44 170
FBC-‐
TUT269 6.15 24
6.49 56 260
FBC-‐
TUT270 6.26 88
6.8 72 220
FBC-‐
TUT271 6.17 48
7 112 250
FBC-‐
TUT133 6.16 128
6.15 24
6.26 88
6.17 48
6.14 32
6.59 60
6.94 116
6.16 128
5.7 108
5.46 40
0

20

40

60

80

100

120

140

100
150
200
250
300

Drop
ball
)me,
s/20cm

SCAN
viscosity,
ml/g

Novel
mechanical
tr.

State-‐of-‐art
mechanical

tr.

1x
with
0.25
mg/g
E1,

6.26%

1x
+
(1x
with
0.25
mg/g
E1)

6.13%

1mg/g
E1,
6.16%

0

20

40

60

80

100

120

140

4.5
5
5.5
6
6.5
7
7.5

Drop
ball
)me,
s/20cm

Cellulose
concentra)on,
%

Novel
mechanical
tr.

SCAN
<220

Novel
mechanical
tr.

SCAN
250-‐260

State-‐of-‐art
mechanical

tr.
SCAN
240-‐250

Figure 5. Tenacity and elongation of fibres at rupture.
n
dope

d
bath

ditives Filtration
Raw

material
18.80
17.40
11.90
12.00
14.20
15.60
14.90
15.40
1.46
1.45
1.39
1.33
0

5

10

15

20

14
16
18
20
22
24
26
28

Tenacity,
cN/tex

ElongaBon,
%

Reference

Spin
dope
addifve

Spin
bath
addifve

Spin
dope
and
bath

addifves

Filtrafon

A dope additive (alkylpolyamine polyoxyethylene
glycol)wasneededtoremovetheairbubblesfrom
the filtered high-viscosity solutions. In addition,
the dope additive enabled storage of the dope at
ambient temperature for several days.
The fibre tenacity of the advanced Biocelsol
fibres was 19 cN/tex (state-of-art Biocelsol fibre
13 cN/tex). Fibre spinning was scaled-up from
50 g to 3 kg of fibre.
Textile demonstration
The applicability of the advanced Biocelsol
fibres for textile products was demonstrated.
The fibres were regenerated, spin-finished, and
opened at Tampere University of Technology
(TUT). The carding, ring spinning, and plying
processes were performed at the Swedish
School of Textiles (University of Borås, Sweden).
The carding and yarn spinning properties of the
advanced Biocelsol fibres were good due to the
suitable crimping of the fibres. The crimps are
formed during the regeneration stage and no
further fibre crimping or texturization is needed.
The design of the textile product was produced
by Studio Tint Ltd, Finland, and included the
colour and pattern of the knitted textile as well
as the model of a hat with an embroidered
Biocelsol brand mark. The industrial-scale flat-
bed knitting and sewing was done by Nevil
Ltd, Finland. The knitted structure was dyed
at TUT according to a dyeing recipe obtained
from Nanso Ltd, Finland. The Biocelsol mark
was embroidered by Brodeca Ltd, Finland. The
demonstration textile product, a Biocelsol hat,
is shown in Figure 6.
Spin bath recycling was studied using the
two-stage nanofiltration (NF) process. In the
first stage, dissolved oligosaccharides were
removed with the open NF membrane (NP030,
NP010 and NTR-7450). The oligosaccharide
retentions were relatively low (<50%). In the
second stage, salts and acids were separated
with the tight NF membrane (Desal-5 DK and
MPS-36), and only slight separation between
acid and salts was achieved. Other membranes
should therefore be tested to achieve efficient
separation of spin bath compounds.
Figure 6. Biocelsol knitted hat.

4.3 Modifications
Chemical modifications were divided into pre-
modifications, i.e. modification of the pulp
prior to dissolution, and post-modifications,
i.e. modification of the regenerated fibres.
Some of the modification routes used are
suitable for both pre- and post-modification.
The objectives of pre-modification were to
improve cellulose dissolution in NaOH/ZnO and/
or improve the water absorption properties of
fibres regenerated from the modified pulp. The
objectives of post-modification were to improve
the water absorption of regenerated fibre and
to functionalize the obtained fibres. High water
absorption properties are required for nonwoven
products such as wipes, diapers, incontinence
products, and feminine hygiene products.
The modification routes of cellulose (both pulp
and regenerated fibres) are presented in Figure 7.
Routes i, ii, and iv were used for pre-modification
and routes ii, iii, iv, and v for post-modification.
The chemical modification routes were:
i. Functionalization of cellulose with butyl
groups containing reactive double
bonds. These can be further cross-
linked and grafted with hydrophilic
monomers such as acrylic acid (AA) and/or
2-acrylaminomethylpropane sulfonic acid
(AMPS).
ii. Functionalization of cellulose with allyl
groups, followed by grafting or cross-linking
as in route (i).
iii. Grafting of cellulose with hydrophilic
monomers as in route (i).
Figure 7. Chemical modification routes: functionalization with (i) butyl or (ii) allyl groups followed by (iii)
grafting, (iv) TEMPO oxidation, and (v) click chemistry.

iv. Oxidization of cellulose fibres, e.g. TEMPO
oxidization. Cellulose is oxidized to cellulose
derivatives containing carboxylic acid
groups, which can be further modified
for other functionalities or used as cross-
linkable functionalities (pre- and post-
modification).
v. Click chemistry
Modification of pulp and its regeneration into
fibres
For the dissolution and regeneration tests,
3-butoxy-2-hydroxypropyl(butylated,BinFigure
7) cellulose and 3-allyloxy-2-hydroxypropyl
(allylated, A in Figure 7) cellulose samples
were prepared. The allylated sample with
degree of substitution, DSA
, 0.09 had higher
solubility (7wt%) in NaOH/ZnO compared to the
unmodified pulp (6wt%). Butylated pulp also
dissolved well in NaOH/ZnO. Due to the high
surface activity of the pulp, extreme quantities
of air bubbles were formed in the alkaline
solutions, as shown in Figure 8.
The spin dope made from the butylated pulp
regenerated easily into fibres. However, the
spin dope made from the allylated pulp did not
regenerated properly, and the allylated solution
was thus mixed with unmodified solution in
ratios 1/4 and 1/10 prior to spinning.
Tenacity and elongation of regenerated fibres
from the modified pulps and reference pulp
were at the same level. However, the swelling
coefficient of the regenerated fibres from
butylated pulp was huge compared to that of
the other fibres, as shown in Figure 9. Water
absorption values are measured as the swelling
Figure 8. Effect of mixing on the formation of air bubbles in alkaline solution from butylated pulp, a) mixed
with a laboratory mixer and b) mixed by hand.
Figure 9. Water absorption capacity of the regenerated fibres as measured by swelling coefficient value.

coefficient, which indicates the amount of
water that the sample is able to hold under
centrifugation. For example, fibres regenerated
from butylated pulp can absorb three times
their own weight of water, whereas commercial
viscose absorbs only its own weight of water.
TEMPO oxidation increased the carboxyl content
of the pulp, and this could be further increased
with NaClO2
treatment. The solubility of TEMPO-
oxidized pulp fibres in NaOH/ZnO was higher
compared to unmodified pulp. Unfortunately,
the solutions made from TEMPO-oxidized pulp
were not suitable for fibre regeneration.
Modification of regenerated fibres
The post-modifications reported in this chapter
can be performed on all types of regenerated
cellulose fibre. Reference and advanced
Biocelsol fibres and commercial viscose fibres
were used for the trials.
For non-woven trials, advanced Biocelsol
fibres were first allylated and then grafted with
different amounts of acrylic acid to obtain post-
modifiedfibrescontainingpoly(acrylicacid)(PAA)
chains. This route was based on etherification
of cellulose fibres with substituents containing
allyl functionalities, as shown in Figure 7
(ii). Fibres having reactive double bonds can
be grafted in very mild aqueous conditions.
Swelling ratio was improved up to 1200% (Table
1), meaning that 1 g of modified regenerated
fibre can absorb 12 g of water. Water absorption
properties were highest when the samples
were converted into neutralized sodium salt
form. The tenacities of the PAA grafted samples
were slightly lower compared to the unmodified
fibres (8 cN/tex vs. 11 cN/tex) as the grafted PAA
increased the fibre weight (linear density in tex
is defined as mass in grams per 1000 metres of
fibre).
Modification with some anhydrides, such as
maleic anhydrides, gives the fibres reactive
double bonds, and at the same time enables
the hydrophilic-hydrophobic balance of the
fibres to be adjusted. The advantage of
etherification is that ether bonds are rather
stabile in acidic and especially in strong alkaline
conditions. Ester bonds, as in the case of maleic
acid derivatives, can be hydrolysed more easily
Table 1. Mechanical properties and water absorption capacities of the post-modified fibres measured by
swelling coefficient value.
Sample
Water
absorption
capacity %
Tenacity
cN/tex
Elongation %
Reference samples
Biocelsol fibre 140 10.9 21
Lenzing viscose 80 22.5 19
Biocelsol fibres grafted with 32% PAA
Small-scale batch
in acid form 170 8.4 15
in neutralized form 1170 7.6 20
Bench-scale batch for non-woven trials
Biocelsol fibres modified with maleate
in acid form 160 6.0 13

in alkaline conditions. Some fibres with maleate
substituents were prepared, but grafting of
these fibres was not performed. The results are
presented in Table 1.
Two different advanced Biocelsol fibre samples
were TEMPO oxidized. The modification
improved water absorption if the fibres were
washed to Na+
-form. However, oxidative
treatment showed detrimental effects on the
mechanical properties of regenerated fibre, as
shown in Figure 10.
A click reaction can be used for post-
modification of regenerated fibres, as shown
in Figure 7 (v), and for functionalization of
regenerated fibres. The irreversible adsorption
of CMC onto cellulose was combined with click
chemistry (alkyne-azide cycloaddition) for
preparing cross-linked Biocelsol fibres. First,
an azide derivative and an alkyne derivative
were adsorbed on the fibre surface. Next, the
click reaction was executed to bring together
the modified regenerated fibres via crosslinking
reaction with the aim of improving the
mechanical properties of the fibres. However,
no improvement in mechanical properties was
found to have resulted from the crosslinking.
4.4 Nonwovens from Biocelsol fibres
The objective of the non-woven trials was to test
the processing properties of the Biocelsol fibres
developed in the FuBio Cellulose programme,
Figure 10. Effect of TEMPO oxidation on the water absorbency (a) and mechanical properties of regenerated
cellulose fibres (b).
a)
a)
Figure
10
Effect
of
Tempo
SwC,
%
Biocelsol Viscose
REF 176 107
H+ 131 102
Na+ 287 210
0

50

100

150

200

250

300

Biocelsol
Viscose

Water
absorp+on
capacity,%

REF

H+

Na+

b)
Elong,
% Tenacity,

cN/tex
Biocelsol-‐REF 20,47 10,9
Biocelsol-‐Na+ 12,63 8,6
Biocelsol-‐H+ 15,47 9,20
0

5

10

15

20

25

0
5
10
15
20
25

Tenacity,
cN/tex

Elonga+on,
%

Biocelsol-‐REF

Biocelsol-‐Na+

Biocelsol-‐H+

Viscose-‐REF

Viscose-‐Na+

Viscose-‐H+

b)

and to characterize the properties of the
obtained nonwoven sheets. The nonwoven
sheets were manufactured by spunlacing
(carding and hydroentanglement).
The processing properties of state-of-art
Biocelsol and advanced Biocelsol fibres were
comparable to commercial viscose fibres in
carding and hydroentanglement carried out in
pilot line for preparing spunlaced non-woven
(Figure 11). Both 100% Biocelsol nonwoven
sheets and 50% Biocelsol: 50% polyester blend
nonwoven sheets were manufactured. The
basis weights of the samples varied between
43-50 g/m2
. Dry-state thicknesses were 0.5
mm for all non-blended samples, and 0.7 mm
for all blended samples. Wet-state thicknesses
varied between 0.4–0.5 mm for non-blended,
and between 0.6–0.7 mm for blended samples.
The water absorption properties are presented
in Table 2. The absorption capacity (measured
as swelling coefficient, Table 1) of the fibres
does not translate directly into the absorption
capacity of the nonwovens because a lot of the
absorption correlates to the void space in the
nonwoven structure. This can clearly be seen in
the higher absorption of the 50% regenerated
cellulose: 50% polyester blends compared to
the nonwovens of 100% regenerated cellulose
fibre. It is well known that the addition of
synthetic fibres to a viscose mix increases the
dry and wet thickness of the nonwoven, which
is reflected in absorption increase. The highest
absorptive capacity was reached with the 50%
chemically modified Biocelsol: 50% polyester
blend. This was, however, only a 15% increase
over the comparable commercial blend, which
cannot be considered a major improvement.
The mechanical properties of nonwoven sheets
are presented in Figure 12. The tensile strength
values of nonwovens made from Biocelsol fibre
were lower compared to nonwovens made from
commercial viscose. The dry tensile strength of
non-wovens made from 50% modified Biocelsol:
50% polyester blend was relatively good, but
dropped to very low levels when wet.
Figure 11. Pilot line for preparing spunlaced non-
wovens.
Table 2. Water absorption properties of non-woven samples
Property Basis weight
(g/m2
)
Water absorptive
capacity (g/g)
Water absorbency
time (s)Fibre blend
100% Lenzing viscose 43.3 11.8 1.3
100% state-of-art Biocelsol 53.5 9.9 4.9
100% novel Biocelsol 49.5 10.9 1.5
50% viscose : 50% polyester 45.9 12.9 4.6
50% state-of-art Biocelsol : 50% polyester 50.6 12.1 2.4
50% advanced Biocelsol : 50% polyester 49.8 13.0 1.9
50% chem.modified Biocelsol : 50% polyester 44.3 14.8 2.2

4.5 Modelling of hemicellulose and
cellulose chain length distribution
evolution as functions of process
conditions in the viscose process
Physico-chemical models were developed
for the different steps involved in viscose
fibre production (xanthation, ripening and
dissolution). Although a limited amount
of experimental data was available in the
literature for development of the model, the
model provided basic information in accordance
with the literature, such as the relative
concentration of different species, changes in
the degree of substitution of cellulose, change
in reactor pressure, etc. As an example, the
simulated change in degree of substitution
during xanthation is illustrated in Figure 13.
The degree of substitution was defined as the
number of xanthate groups present on one
anhydro-glucose unit of the cellulose chain.
Normally, fresh xanthated cellulose solution
has DS values of 0.5–0.7 after 80–120 minutes
of xanthation and seldom exceeds DS 1. The
simulation results give DS 0.5–0.65 for the
same xanthation period.
The evolution of molecular weight distribution
(MWD) can be modelled by population balance
models. Discretization of MWD into categories
representing a certain DP range considerably
reduced the computational time. A model for
prediction of changes in MWD during alkali-
cellulose aging was developed and different
equations for polymer scission rate were
tested. Promising results were obtained from
the scission rate equation given below.
ScissionRate = k(1-exp[-((DP-1)/c)d
].exp(-a.t)
Values of ‘k’, ‘a’, ‘c’, and ‘d’ are constant and
their values were optimized. ‘DP’ is degree of
polymerization and ‘t’ is ageing period.
Figure 12. Mechanical properties of non-wovens (MD = machine direction, CD = cross direction).
a)
0

20

40

60

80

100

Tensile
strength
MD,

dry

Tensile
strength
CD,

dry

Tensile
strength
MD,

wet

Tensile
strength
CD,

wet

Force
(N)

100%
viscose
100%
state-‐of-‐art
Biocelsol

100%
novel
Biocelsol
50%
state-‐of-‐art
Biocelsol
:
50%
PES

50%
novel
Biocelsol
:
50%
PES

b)
0

20

40

60

80

100

120

140

ElongaSon
MD,
dry
ElongaSon
CD,
dry
ElongaSon
MD,
wet

ElongaSon
CD,
wet

Elonga/on
(max
%)

100%
viscose
100%
state-‐of-‐art
Biocelsol

100%
novel
Biocelsol
50%
viscose
:
50%
PES

50%
state-‐of-‐art
Biocelsol
:
50%
PES
50%
novel
Biocelsol
:
50%
PES

Comparisons of experimental and model
results for different ageing periods are shown
in Figure 14. Continuous lines represent the
experimental MWD, while dots represent the
discretized categories from the model. As
cellulose degradation takes place continuously
throughout the viscose process, a model similar
to ageing could be implemented for the other
Figure 14. Comparison of experimental (lines) and model (dots) MWDs for different ageing periods.
process steps For the model development,
experimentalMWDsfromtheotherstepsshould
be available for optimizing the parameters in
the scission rate equations. The experimental
method generally used for analysis of MWD
does not provide very accurate results for the
lower molecular weight range, which makes
accurate MWD modelling more challenging.
Figure 13. Degree of substitution from simulation (xanthation).
0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0
40
80
120
160

DS
(number/AGU)

Time
(min)

Degree
of
subs6tu6on

3 4 5 6 7
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1 hr
dw/d(logMw)
log Mw
3 4 5 6 7
0.0
0.2
0.4
0.6
0.8
1.0
1.2
4 hr
dw/d(logMw)
log Mw
3 4 5 6 7
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1 hr
dw/d(logMw)
log Mw
3 4 5 6 7
0.0
0.2
0.4
0.6
0.8
1.0
1.2
4 hr
dw/d(logMw)
log Mw
3 4 5 6 7
0.0
0.2
0.4
0.6
0.8
1.0
1.2
2 hr
dw/d(logMw)
log Mw
3 4 5 6 7
0.0
0.2
0.4
0.6
0.8
1.0
1.2
8 hr
dw/d(logMw)
log Mw
3 4 5 6 7
0.0
0.2
0.4
0.6
0.8
1.0
1.2
2 hr
dw/d(logMw)
log Mw
3 4 5 6 7
0.0
0.2
0.4
0.6
0.8
1.0
1.2
8 hr
dw/d(logMw)
log Mw

Population balance based modelling has now
been implemented to predict the evolution
of molecular weight distribution of cellulose
during alkali cellulose ageing. These models
can be used for optimizing viscose process
conditions for a certain end product quality.
the results
The results obtained on the effects of enzymatic
pre-treatment on molecular weight and fibre
surface pores can be utilized in planning pre-
treatments to activate cellulose for dissolution
to advanced Biocelsol or other water-based
systems, as well as for chemical synthesis. The
results regarding combined mechanical and
enzymatic pre-treatments indicate that there is
a potential to decrease the enzyme dosage and
mechanical energy usage and thus enhance the
economic feasibility of mechanical-enzymatic
pre-treatment. It is also expected that both the
pre-treatment and the dissolution process can
be up-scaled to the industrial level. Recycling of
water, enzymes and chemicals in the Biocelsol
process by membrane separation requires
further development.
The results of advanced Biocelsol fibre
regeneration and demonstrations of spunlaced
nonwovens and textile products are promising.
The unmodified and modified Biocelsol fibres
have higher water uptake values compared
to commercial viscose fibres. Applications for
Biocelsol-based fibres should reflect this special
property. However, the mechanical properties
of Biocelsol fibres need further development.
Further modification and improvement of the
mechanical properties could be achieved, for
example, by utilizing the reactive allylic double
bonds of modified and regenerated fibres in
suitable post-crosslinking and post-grafting
techniques.
Replacement of the current viscose process
with the novel water-based dissolution and
regeneration process would reduce the
environmental impact of regenerated fibre
manufacturing. Price competitiveness, however,
remains a key challenge for the implementation
of Biocelsol fibre.
Process models of different viscose process
steps could be used for optimizing process
conditions and product quality. The information
obtained from these models could, in turn,
be used in the evaluation of various process
alternatives during the development of new
processes. The modelling work thus supports
the development of more cost-efficient and
environmentally friendly processes, which
is not only beneficial for industry but also for
the environment and society. New process
understanding gained through modelling could
also be used in teaching at university level.
6. Networking
The research was carried out jointly by research
organizations, Finnish forest cluster companies,
and other companies. Table 3 presents the
research partners and their roles.

Partner Role
Aalto University
- Forest Products Technology (FPT)
- Biotechnology and Chemical
Technology (BCT)
FPT: Characterization of pore structure (thermoporosimetry and solute
exclusion techniques); post-modification of regenerated fibres (click-
chemistry, TEMPO oxidation)
BCT: Physico-chemical modelling of viscose process steps.
Lappeenranta University of
Technology
- Separation Technology
Filtration studies.
Metsä Fibre Industrial tutor. Defining, steering and providing competence for the
viscose modelling.
Suominen Nonwovens Industrial tutor, manufacturing and characterization of spunlaced
nonwovens.
Stora Enso Industrial tutor. Giving industrial insight and steering of the work.
Tampere University of
Technology
-Materials Science
Solubility trials, SCAN viscosity measurements of pulp and regenerated
fibre, preparation of spin dopes, fibre spinning trials, management of
textile demonstration.
- Polymer Chemistry (PC)
- X-ray Physics (XP)
PC&XP: Characterization of fibre pore structure (NMR methods, X-ray
studies)
PC: Characterization of dissolution process.
University of Oulu
-Fibre and Particle Engineering
Experiments of mechanical treatments.
UPM-Kymmene Industrial tutor. Giving industrial insight and steering of the work.
VTT Mechanical and enzymatic treatments, factors affecting cellulose dissolution,
pre-treatment process development; pre- and post-modifications.
Publications:
Pahimanolis, N., Salminen, A., Penttilä, P.,
Korhonen, J., Johansson, L., Ruokolainen, J.,
Serimaa, R. and Seppälä, J. Nanofibrillated
cellulose/carboxymethyl cellulose composite
with improved wet strength. Cellulose 20(3),
1459-1468, 2013.
Grönqvist, S., Hakala, T. K., Kamppuri, T.,
Vehviläinen, M., Hänninen, T., Liitiä, T.,
Maloney, T. and Suurnäkki, A. Fibre porosity
development of dissolving pulp during
mechanical and enzymatic processing.
Cellulose 2014. DOI 10.1007/s10570-014-0352-x.
Rissanen, M., Syrjälä, S., Vehviläinen, M.
and Nousiainen, P. Solubility and solution
rheology of enzymatically treated pulp. Annual
Transactions of the Nordic Rheology Society,
Vol. 19, 303-306, 2011.

Presentations:
Grönqvist, S., Maloney, T., Kamppuri, T.,
Vehviläinen, M., Hakala, T.K., Liitiä, T.,
Hänninen, T., Suurnäkki, A. Activation of
cellulose. Oral presentation in FuBio seminar,
August 25, 2013, Helsinki, Finland.
Kamppuri, T., Vehviläinen, M., Grönqvist, S.
and Rissanen, M. Novel regenerated cellulose
fibres with high water absorption properties.
Oral presentation in Ambience'14 & 10i3m
Conference, Sept 7-9, 2014, Tampere, Finland.
Kamppuri, T., Vehviläinen, M., Grönqvist,
S., Setälä, H., Maloney, T. and Rissanen, M.
Fabrication of wood cellulose – from pulp to
textiles: Biocelsol. Oral presentation in FIBIC
seminar April 15, 2014, Espoo, Finland.
Nousiainen, P., Vehviläinen, M. and Rissanen,
M. Enzymatic Modification of Pulp Cellulose
to Regenerated Fibres and Films via Aqueous
Alkaline Solutions. Oral presentation in The Third
Nordic Wood Biorefinery Conference, March 22-
24, 2011, Stockholm, Sweden.
Rissanen, M. Sustainable Development
in Textiles. Oral presentation in Cristal –
Sustainable development in lifelong learning,
June 6, 2013, Valkeakoski, Finland.
Sixta, H., Nousiainen, P., Vehviläinen, M.
and Rissanen, M. From wood to structural
materials: Regenerated fibres for textiles and
nonwovens. Oral presentation in FuBio seminar,
October 1, 2012, Espoo, Finland.
Vehviläinen, M., Kamppuri, T., Rissanen, M.
and Nousiainen, P. Cellulose regeneration from
aqueous solution. Oral presentation in FuBio
seminar, March 11, 2013, Turku, Finland.
Posters:
Michud, A. and Rissanen, M. From Cellulose
to textile fibre and a ready product. Poster
presentation at SHOK Summit 2014, May 15,
2014, Helsinki, Finland.
Hänninen, T., Kamppuri, T., Vehviläinen, M.,
Grönqvist, S., Hakala, T.K. Dissolution of
TEMPO oxidized pulps in aqueous alkaline
solvents. Poster presentation in FIBIC
seminar, November 20, 2013, Espoo, Finland.
Rajala, S., Kamppuri, T., Vehviläinen, M. and
Setälä, H. Regeneration of modified cellulose
into fibres. Poster presentation in FIBIC seminar,
November 20, 2013, Espoo, Finland.
Waqar A., Kuitunen, S., Alopaeus, V. Modelling
of xanthation kinetics during viscose process.
Poster presentation in FIBIC seminar, November
20, 2013, Espoo, Finland.
Rajala, S., Kamppuri, T., Vehviläinen, M. and
Setälä, H. Regeneration of modified cellulose
into fibres. Poster presentation in FuBio
seminar, August 27, 2013, Helsinki, Finland.
Penttilä, P., Kilpeläinen, P., Tolonen, L.,
Suuronen, J.-P., Sixta, H., Willför, S. and
Serimaa, R. Effects of Pressurized Hot Water
Extraction on the Structure of Birch Sawdust.
Poster presentation. COST FP1105 13.05.2013 -
14.05.2013, Edinburgh, UK.
Penttilä, P., Kilpeläinen, P., Suuronen, J.-P.,
Willför, S. and Serimaa, R. Effectsofpressurised
hot water extraction on the nanoscale structure
of birch sawdust. Poster presentation. Physics
Days, 14–16. 3. 2013, Espoo, Finland.

Maloney, T., Grönqvist, S., Hakala, T.K.,
Hänninen, T., Penttilä, P., Kamppuri, T.,
Vehviläinen, M., Serimaa, R. and Suurnäkki,
A. Pore Analysis of Dissolving Pulps. Poster
presentation at the FuBio open seminar day,
August 2013, Helsinki, Finland.
Rissanen, M., Syrjälä, S. and Vehviläinen, M.
Nousiainen P. Solubility and solution rheology of
enzymatically treated pulp. Poster presentation
at The Nordic Rheology Conference, June 8-10,
2011, Helsinki, Finland.
Theses:
Penttilä, P. Structural Characterization of
Cellulosic Materials Using X-Ray and Neutron
Scattering, PhD thesis, University of Helsinki,
1.11.2013. Report Series in Physics HU-P-D207.
Rajala, S. Regeneration of modified cellulose
into fibres. Master’s thesis. Tampere University
of Technology, 5.6.2013.

PAR TN E RS
Kemira
Metsä Fibre
Stora Enso
Suominen
UPM-Kymmene
CONTAC T PE RSON
Katja Salmenkivi, katja.salmenkivi@poyry.com
TEXTILE VALUE CHAIN
RELATED TO
FUBIO TEXTILE FIBRES

ABSTRACT
The key objective was to provide information on market prospects, needed end-use pro-
perties, and value chain dynamics in order to help Finnish Bioeconomy cluster (FIBIC) part-
ners focus their research investments on the most lucrative market areas. The project
included a market assessment on textile and nonwoven fibres, as well as an analysis of the
value chain structure and value creation in the apparel industry.
As the largest end-user of man-made cellulosic fibres, the apparel industry is a high-po-
tential application area for the FuBio Cellulose fibres Ioncell and Biocelsol. Despite a clear
trend towards more environmentally friendly and natural fibres, the industry is very price
conscious with constant pressure from brand owners to trim the supply chain. Fibres that
offer technical or cost benefits compared to existing products thus offer the greatest po-
tential for success.
Successful entry into the apparel value chain calls for brand owner cooperation as early as
possible in the development process. FIBIC partners should locate end users that are able
and willing to engage in R&D cooperation and provide industry insights regarding potential
applications and needed fibre properties.
Keywords:
apparel, technical textiles, nonwovens, Ioncell, Biocelsol, man-made cellulosic fibres, market
forecast, value chain

Figure 1. Scope of market and techno-economic
analyses related to FuBio textile fibres.
13
jelmatunnukset
1. Work background
The target market for the novel fibres
developed in the FuBio Cellulose programme
(Ioncell and Biocelsol) is the global market for
man-made cellulosic fibres. The increasing
global demand for cellulosic fibres cannot be
met by cotton alone. In addition, there are
serious drawbacks related to the conventional
viscose process, namely the use of extremely
volatile and toxic carbon disulphide (CS2) and
a complex process configuration. The lyocell
process yields fibres with higher strength
properties than other cellulosic fibres with an
environmentally friendlier process. However,
the solvent used in the process, NMMO, has
low thermal stability and there is an increased
risk of runaway reactions. The FuBio Cellulose
programme aimed at developing novel
processes for the manufacture of regenerated
fibres, which yield intrinsic fibre properties
superior to viscose fibres but with significantly
less environmental impact.
Man-made cellulosic fibres are widely used
in the textile and nonwoven markets. In the
FuBio programme, Pöyry analysed cellulosic
fibre consumption in three main segments:
apparel, technical textiles, and nonwovens.
The apparel industry represents over 70%
of all fibre markets and thus this overview is
exclusively focused on the textile fibre market
and apparel value chain (Figure 1).
2. Objectives
The objective of this work was to analyse
the market opportunities for novel cellulose
products in textiles and nonwoven applications
and to analyse the value chain structure and
value creation in the apparel industry.
Market assessments and value chain
analyses were carried out as desktop studies
complemented by external expert interviews
when applicable. All market assessments
were conducted with close guidance from the
industrial partners. The industrial partners
also took part in workshops to define the
frameworks and scope of the studies as well as
in result reviewing and dissemination activities.
4. Results
4.1 Textile fibre markets
Synthetic fibres dominate the textile market
with over 50% of annual fibre consumption.
Cotton is the most widely used natural fibre,
representing approximately a third of all fibre
demand (Figure 2). In 2011, consumption of
synthetic textile fibres was almost 20 times
higher than in 1970. The main reasons for this
dramatic growth are the low production cost
of synthetic fibres, the development of novel
synthetic fibre grades and the limited supply
of cotton fibre. Man-made cellulosic fibres
constitute approximately 5% of global fibre
production.

Polyester
Cotton
Polypropylene
Acrylics
Polyamide
Wool
Silk
Other synthetics
Cellulosics
Viscose
Other cellulosic fibres
Knitted and woven textiles
Nonwoven
Modal
High purity nonwoven
Micro-denier
Flame retardant
Hygiene products
Pöyry Management
Consulting Oy
Figure 2. Global consumption of textile and nonwoven fibres
Although man-made cellulosic fibres
represent only a small share of global fibre
consumption, they have several advantages
compared to cotton and polyester, such as
feel, wearer comfort, softness, smoothness,
moisture absorbency and capability for fibre
modification. The most significant factor
limiting the use of man-made cellulosic fibres
is the price difference compared to cotton
and polyester. Price is typically the strongest
driver of fibre selection, although the effect of
trends and traditions cannot be underrated.
The choice of fibre is always a compromise
between cost and fibre properties. Different
blends are generally used both to reduce
the manufacturing price of the garment and
to modify textile properties such as pilling,
softness, washability, and durability.
The demand for textile fibres is estimated to
grow along with population growth, GDP growth,
and a growing middle class with rising disposable
income, among other trends. The more the
GDP of a nation grows, the more textiles are
consumed.Themajorityoftheprojecteddemand
growth in textile fibres will be realized in the BRIC
(Brazil, Russia, India and China) and booming
Asian countries, to which manufacturing is
also relocating. Today, over half of all man-
made cellulosic fibres are produced in China. It
should be noted that the apparel value chain
as a whole is more energy and labour intensive
than, for example, the chemical or automotive
industries. Due to higher energy prices, higher
labour costs, restricting legislation, numerous
standards, and the complexity of EU policies,
European producers struggle to compete
against developing countries. Fibre innovations
are therefore essential for business survival and
research and development of man-made fibres
is concentrated accordingly in the industrialised
countries. Europe’s long-term opportunity is to
form a hub of specialized fiber production, while
commodities would be produced in countries
such as Indonesia, India or Brazil.
Although man-made cellulosic fibres
represent only a marginal share of today’s

total fibre production, there is one widely-
known but disputed scenario that predicts
strong growth for cellulose-based fibres
(Figure 3). This ‘cellulose gap’ scenario is built
on the estimation that one third of all fibres
should be based on cellulose materials due
to their softness and moisture management
properties. ‘Gap’ refers here to the supply gap
in cellulose-based fibres, to which current
cotton production methods are unlikely to be
able to respond. Cotton production uses over
twenty times more water than viscose, needs
four times more high-grade arable land and
is one of the largest markets for pesticides.
Although viscose fibres have traditionally had
a significant price premium over cotton, this
cellulose gap would provide an opportunity for
man-made cellulosic fibres, including FuBio
Cellulose fibres, as cotton substitutes.
The global textile market is highly dependent
on the general economic situation. In weaker
conditions, apparel is one of the first segments
to suffer. The apparel industry is dominated by
strong brands that dictate each step of the
value chain. This has resulted in an industry
ready to relocate, always seeking the lowest
labour cost countries.
The key trends shaping the apparel industry
include fast-moving fashion trends, mass
consumption, increasing homogeneity of
the sector, growing interest in environmental
sustainability, and increasing demand for
functional textiles. Global fashion trends are
defined by a small number of players (e.g.
Peclers Paris), which the majority of brand
owners follow. Multinational retail chains
supplying low-price, short lifespan products
are among the strongest drivers of mass
consumption and increasing homogeneity of
the apparel industry.
Environmental sustainability and ethical
consumerism have increasing influence in
the textile industry, especially in the mature
Figure 3. Cellulose gap scenario1
.
1. Modified from
‘The Cellulose Gap’ by Gherzi, Feb 2011
0
20
40
60
80
100
120
140
160
1900 1920 1940 1960 1980 2000 2005 2010 2015 2020 2025 2030
Cellulose Gap
MMCF
Cotton
Synthetic
Fibres
Wool
Million tons
Pöyry Management
Consulting Oy

markets. Green products are gaining ground
but remain niche, as the majority of consumers
are unwilling to pay a premium. One means of
justifying premium pricing and ensuring that
products live up to environmental and ethical
standards is labelling. One of the most well-
known labels in the textile industry is the Oeko-
Tex Standard, a globally uniform testing and
certification system for textile raw materials,
intermediate and end products at all stages of
production. The importance of these standards
and labels is emphasized with apparel items
that are in close contact to sensitive skin, such
as underwear and children’s clothing.
Increasing demand for functional apparel is
interlinked with the development of technical
textiles and materials. Functional textiles
include smart features, such as responding to
variations in body temperature and absorbing
heat and body moisture to give the clothing a
comfortable, dry feel. A common example is
the Gore-Tex fabric.
4.2 Apparel value chain
Despite considerably varying end product
requirements, textile supply chains share
numerous close similarities and the apparel
industry as a whole can be described using
a single generic supply chain. The supply
chains all use similar raw materials, which
are further treated processed to gain specific
characteristics. Most differences therefore
occur at the consumer end of the chain. In the
apparel industry, retailing plays a significant
role in distribution, whereas technical and
industrial textiles are typically a business-to-
business market. The textile industry is strongly
connected to the chemical industry throughout
the chain, from fibre to finished product. The
production of man-made fibres is an entirely
chemical process, and many natural fibres
require substantial amounts of pesticides,
fertilizers and other chemicals for cultivation,
harvesting and protection during transport.
Downstream processing also consumes
substantial amounts of chemicals, including
dyes for colouring and different finishes and
preservatives for ready-made garments.
The apparel value chain studied in the FuBio
Cellulose programme starts with pulp and
cellulose fibre producers and ends with
distribution and sales. In this overview, the
beginning of the value chain is limited to viscose
and lyocell fibres. After fibre production the
fibre is spun into yarn, which is woven or knitted
into a fabric and usually treated with finishing
chemicals. Finishing can be also done after
garment manufacturing or in both steps. Each
step in the value chain, from pulp to garment
manufacture, can be performed separately or
be partially or wholly integrated (Figure 4).
Distribution and sales can be roughly divided into
two parts: the ‘traditional apparel value chain’,
where retailers buy ready-made garments from
garment manufacturers with little influence
on the value chain and the products, and the
‘new apparel value chain’, which is controlled by
strong multinational brand owners.
The number of operators increases significantly
towards the end of the apparel value chain
(Figure 5). In the pulp production step there
are only tens of operators, while in the retail
step there are tens of thousands of operators.
Pulp producers are often forward integrated
with fibre producers due to demanding quality
parameters, which makes switching costs
notable resulting in notable switching costs
between suppliers or fibre types. Also, barriers
to entry are substantial in the dissolving pulp
industry, for example, due to high investment
costs of new plants, know-how issues and the
length of the product approval process.
Viscose fibre production is heavily concentrated
in China, with large companies dominating the

Figure 4. Integration of operations within the apparel value chain.
Pulp
producer
(Viscose
staple)
fibre
producer
Spinner
Weaver/
Knitter
Finishing
Garment
manu-
facturing
Distribution and
sales by Brand
owner or Retailer
Integrated from the production of fibre to garment (Toray)
Integrated from yarn spinning to garment production (e.g. Bombay Rayon Fashions, Arvind)
Integrated from yarn spinning to weaving/knitting (e.g. Weiqiao, Alok Industries, Toyobo, Huafu)
Integrated from fabric manufacturing to retail (e.g. American Apparel)
Wholesaling by
brand owner
Cotton,
wool, silk
etc.
producer
Synthetic
fibre
producer
Design, marketing, advertising, supply
chain management
Integrated dissolving pulp and viscose production (e.g. Birla)
Pöyry Management Consulting Oy
• Less than
hundred operators
• Thousands
of operators.
• Focus in
production is
in China and
rest of Asia
• Strong
consolidation
trend
• Thousands of
operators.
• Focus in
production is in
China and rest
of Asia
•Strong
consolidation
trend
• Can be done by
same operator as
fabric or garment
• Low investment
costs make it easy
to establish new
operators if
needed
• Tens of
thousands of
operators
• Focus in
production is in
Asia
• Tens of
thousands of
operators both in
retailing and
brands.
• Some strong
brands are more
significant than
others and drive
the consumption.
Amount of
operators
Fibre
producer
Spinner
Weaver/
Knitter
Finishing
Garment
manu-
facturing
Distribution
and sales
Pulp
producer
• ~30 Operators
Source: Pöyry, Interviews Pöyry Management Consulting Oy
Figure 5. Number of operators in the apparel value chain.

Figure 6. Geographical distribution of key companies in each step of the value chain.
market. In addition, China also accounts for the
bulk share of global viscose demand. The number
ofyarnmanufacturers,weavers andknitters is,on
the other hand, substantially larger compared to
fibre producers. The industry includes hundreds
of large companies and thousands of operators,
which are mainly located in China and the rest
of Asia (Figure 6). Finishing is usually done by
the same company as the weaving or by the
garment manufacturer if finishing is applied to a
ready-made garment. Overall, there is a strong
consolidation trend throughout the fibre industry.
Garment manufacturing has notably more
players than the previous steps in the value
chain. The industry is very labour intensive
and characterized by high cost competition.
Relocating a garment manufacturing company
is relatively cheap and easy, which is one of
the reasons why the industry is in constant
search of new locations with readily available
cheap labour.
There are tens of thousands of operators in
apparel distribution and sales. Consumption
and industry changes are driven by a
handful of a relatively small pool of strong
and influential global brands. Brand owners
rarely own production capacity as it offers
no competitive advantage, and contracts
between garment manufacturers and brand
owners tend to be short-term.
The vast majority of apparel industry companies
are affected by fashion trends driven by strong
consumer brands. These trends are transferred
to yarn and fabric manufacturers through
demand for certain types of products. Trends
rarely impact fibre selection directly, since
brands are unlikely to commit to a specific fibre
type. However, trends can have an indirect
effect on fibre demand. For example, a new
trend for glossy garments will increase demand
for viscose in fibre blends. The freedom to
switch between fibre materials also serves as a
dampener against price volatility. If, for example,
cotton is readily available at a reasonable price,
100% cotton fabrics are used. If the cotton
price soars, production costs can be easily cut
by using blends of lower-cost fibres.
General retailer
Pulp producer
Brand retailer
Garment manufacturer
Spinner
Fibre producer
Weaver

Pulp producer Fibre producer Spinner Weaver/knitter
Garment
manufacturer
Retailer (general)
Retailer
(brand owner)
Key bottlenecks
determining
competitive
position
- Wood price and
availability
- Pulp price
- Price of
competing
fibres
- Competing
yarn producers
- Changes in
material price/
availability
(depending
on flexibility &
specialization)
- Competing
fabric
producers
- Changes
in material
preferences
(depending
on flexibility &
specialization)
- Competing
garment
producers
- Strong
bargaining
and pricing
power of brand
owners
- Competing
retailers
- Competing
brands, private
labels of
retailers
- Threat of
forward
integration of
big garment
manufacturers
Sources of power
& profits
- Cost
competitiveness
- Technology
advantage
- Wood
procurement
- Price and price
variability of
cotton
- Quality
(compared
to synthetic
fibres)
- Lowest cost
producers have
an advantage
- Vertical
integration
to increase
control and
power
- Raw material
flexibility
- Lowest cost
producers have
an advantage
- Differentiation
with high
quality fabrics
- Vertical
integration
- Lowest cost
producers have
an advantage
- Vertical
integration
- Marketing
- Brand and
position of the
retail chain
- Bargaining
power of large
retail chains
- Bargaining &
pricing power
through strong
brand
- Control over
the whole
supply chain
Clock speed SLOW SLOW FAST FAST FAST FAST FAST
(Investment cycle,
raw material price)
(Investment
cycle)
(Labour price,
competitors,
fashion)
(Labour price,
competitors,
fashion)
(Labour price,
competitors,
fashion)
(Category
management,
fashion, colour
industry)
(Fashion trends,
colour industry)
Table 1. Competitive position of operators in the apparel value chain.
The bargaining power of big brands is
overwhelming compared to suppliers, with
practically no switching costs incurred in
changing suppliers. The gap in bargaining
power is not likely to narrow, rather the
opposite, which has an impact on the
profitability of all players in the upstream value
chain. Table 1 summarizes the competitive
position of different operators in the apparel
value chain. The ‘clock speed’ analogy
illustrates how different parts of a value chain
operate with different cadences, or at different
inherent clock speeds. In studying a long value
chain it is not only important to know the speed
of its different parts, but also to understand the
key criteria determining this speed.
Integrated brand owner retailers are the most
profitable players in the value chain, which
emphasizes the strong position of brands in
the apparel industry. Companies with strong
brands have strong bargaining power towards
their suppliers and can sell their products
with higher margins. The high profitability
of integrated brand owner retailers further
illustrates the global success stories of the
rapidly growing ‘fast retailing’ companies such
as Inditex and H&M. Despite fierce competition
in apparel retailing, the big players are able to
run their operations with reasonable margins.
The profitability of non-integrated companies
located in the middle of the value chain
(fibre, yarn, fabric, and garment) is generally
lower than that of companies at the ends of
the chain. Some vertically integrated apparel
manufacturing companies have, however,
reported relatively high profits. Although the
number of fibre producers is significantly
lower than that of apparel manufacturers,
the value created by non-integrated fibre
producers is rather poor. This can be partly
explained by high raw material costs, strong
competition and low switching costs between
different fibre types, resulting in poor pricing
power. Pulp and integrated pulp and fibre
producers, on the other hand, have recorded
solid financial results.

4.4 Conclusions
The apparel industry – the largest end-user of
man-made cellulosic fibres – is an attractive
application area for Ioncell and Biocelsol fibres.
In addition, the proposed ‘cellulose supply
gap’ scenario would have most significance
in the apparel market where there is a clear
trend towards more environmentally friendly
and natural fibres. On the other hand, the
apparel industry is highly price conscious,
with constant pressure from brand owners to
trim the supply chain. The market is dictated
by fashion trends driven by strong brands,
which also make most of the profit in the value
chain. The beginning of the apparel value
chain, especially fibre production, spinning and
weaving, is concentrated in Asia, whereas the
end of the value chain (brand and retailing) is
located mainly in the West.
Operations in the middle of the apparel value
chain are integrated in various ways. In general,
the value chain as a whole is characterized by
heavy competition, especially on price and
quality. Therefore, fibres that offer technical or
cost benefits compared to existing products
offer the greatest potential for success. Overall,
integrated companies seem to have better
profitability than companies focusing on only
one step of the value chain. A major challenge
for many industry operators is the volatile price
and availability of cotton, which has an impact
on the entire market from fibre suppliers to
brand owners. The main bottlenecks for FuBio
Cellulose fibre in the apparel value chain are
related to politics and legislation, product quality,
technology, and the fibre market (Table 2).
Successful entry into the apparel value chain
calls for brand owner cooperation as early as
possible in the development process. FIBIC
partners should locate end users that are willing
and able to engage in active R&D cooperation
and provide industry knowledge and insights
regarding potential applications and needed
fibre properties. Cooperation with Marimekko is
a major step in this direction.
Table 2. Main bottlenecks for FuBio Cellulose fibres in the apparel value chain.
Security of supply for viscose producers in China has to be ensured before you can
build a new cellulose based fibre plant.
India and China can increase their protection of home markets thus making it
impossible to export to these countries.
Politics /
legislation
If the FuBio Cellulose fibre is
expected to be technically
similar to viscose, the quality
barriers are:
A certain capacity is required to be able to get customers.
There are no high switching costs in the value chain for FuBio
Cellulose fibre if the technical quality is similar to viscose.
Product
quality
Technology
The availability of raw material for fibre producer has limited the production (dissolving pulps).
The volatile price and availability of cotton affects the whole fibre market.
Fashion trends can benefit or hinder the demand of the fibre.
Market
Wet strength is not good as for synthetic fibres
Highest yarn uniformity
Resistance to rubbing, abrasion, pilling
No hydrophobic properties

the results
The work provided its industrial partners with
important information on market prospects,
end-use properties, and value chain dynamics
that can be used to evaluate the opportunities
and risks entailed in entering these markets.
It also provided key recommendations and
actions for achieving further development and
market entry.
The results of the different market analyses
were communicated not only to the industrial
partners in question but also to all researchers
in the FuBio Cellulose programme. Based on
the overall results, successful market entry
in the studied value chains requires proactive
cooperation with brand owners, as the brand
owners are the key decision makers in all of the
studied value chains. In 2013, FIBIC organized
a textile value chain workshop in which
selected brand owners took part in evaluating
the industrial importance and applicability of
the FuBio Cellulose results. As a result, new
cooperation was initiated with several brand
owners and FIBIC industrial partners.
According to the industrial partners, the market
analyses provided essential information
regarding the market prospects of the selected
end-product areas of the Fubio Cellulose
programme. The results clarified both the
market potential and barriers for the most
promising products as well as the current
supply and demand situation. The impacts
of the market analyses from the industrial
partners’ viewpoint can be summarized as:
- increasing understanding of the markets
relevant to the programme, and
- increasing understanding of the value chains
relevant to the programme
6. Networking
The market assessments facilitated
communication between the researchers
and industrial partners by bringing the latest
research information into the business arena.
As stated in the research plan approach, all
work tasks were carried out in close cooperation
between the industrial partners, research work
package leaders and individual researchers.
Key results were communicated to industrial
partners and researchers during internal
work package meetings, FIBIC seminars, and
industrial partners’ internal meetings.
Table 3 presents the roles of Pöyry Management
Consulting and industrial partners in this work.
Partner Role
Kemira Industrial tutor. Defining and guiding the market assessments.
Metsä Fibre Industrial tutor. Defining and steering the value chain analysis. Defining and guiding
the market assessments.
Pöyry Management
Consulting
Market assessment. Value chain analysis.
Suominen Industrial tutor. Defining and guiding the market assessments.
Stora Enso Industrial tutor. Defining and guiding the market assessments.
UPM-Kymmene Industrial tutor. Defining and guiding the market assessments.

Posters:
Rouhiainen, J. and Salmenkivi, K. Value chains
and value creation: Case apparel value chain.
Poster presentation at the FuBio seminar,
August 2013, Helsinki, Finland.

NEW
PRODUCTS
CONTAC T PE RSON
Jaakko Hiltunen, jaakko.hiltunen@vtt.fi
PAR TN E RS
Glocell
Metsä Fibre
Stora Enso
Suominen
UPM-Kymmene
Åbo Akademi University

The main focus was to develop novel absorbent cellulose materials for wiping and hygienic
applications and thermoprocessable celluloses for melt-spinning and extrusion coating by
industrially feasible methods. Furthermore, the aim was to produce functional beads from
wood-based cellulose.
A heterogeneous esterification process for producing thermomeltable cellulose esters from
different pulp materials at high pulp consistencies (15-25 wt-%) was demonstrated. The targeted
material properties, including good film forming ability and melt-spinnable formulations, were,
however, not fully obtained, likely due to the inhomogeneity of the starting materials. Melt
extrusion of the best synthesis materials by laboratory-scale twin-screw microcompounder
was successful, indicating that these materials could be utilized, for example, in injection
moulding processes. Commercial cellulose acetate butyrate showed very good spinnability in
the melt spinning process; the mechanical properties were comparable to polypropylene fibres.
The processability of the various commercial cellulose acetates in extrusion was good and
polymer films with good quality were produced.
Absorbent cellulose materials were easily produced at the kilogram scale. The results showed
that drying of chemically modified fibres is challenging when the target is to maintain improved
absorption properties. Hence, the drying process itself can be considered as a bottleneck in
developing novel absorbent materials competitive with currently used superabsorbents and
thus it needs further development. Processing of novel absorbent fibres by foam forming was
feasible and the foam-formed absorbent fibres were suitable for making novel nonwoven
structures by hydroentanglement. The absorbent fibres also provided improved water
absorption and water retention capacities in the evaluated fluff pulp compositions.
Physicochemically and chemically functionalized cellulose beads were prepared using an
environmentally-friendly water-based solvent. HyCellSolv pretreatment was developed for
making cellulose beads from different wood pulps. The method was successfully up-scaled
to meet the demands of the semi-pilot scale bead machine. Functional cellulose beads were
utilized as drug carriers. Drug delivery was studied with physicochemically modified beads,
oxidized anionic beads and CMC-cellulose blended beads. All of the beads demonstrated high
loading capacity and extremely good uniformity. In addition, controlled release from the beads
was recorded with various active pharmaceutical ingredients (APIs). The cellulose beads and
oxidized cellulose beads both showed excellent properties as drug carriers.
ABSTRACT
Keywords:
absorbents, cellulose, cellulose esters, beads, extrusion, fluff, hygiene products, melt spinning,
nonwoven, synthesis

1. Work background
Industrial processes for making cellulose
derivatives are typically rather complicated
multi-stage processes consisting of different
activation, hydrolysis and purification steps,
depending on the nature of the reaction
and the targeted end-product properties. In
addition, use of expensive high-purity pulps
is typically a prerequisite, especially for the
production of thermoprocessable cellulose
materials. For these reasons most cellulose
derivatives are too expensive to compete with
conventional polymers such as polypropylene
(PP), polyamide (PA) or polylactide (PLA).
Simplification of the manufacturing process,
reduced chemical consumption and use of
cheaper raw materials are factors that could
enable significantly lower market prices for
cellulose derivatives. Additionally, general wet-
lay methods for obtaining cellulose filament
material have the problem of low productivity
due to a low spinning rate. Therefore, a spun-lay
process not using an organic agent is necessary
for obtaining low-environmental-load fibres
using cellulose as a raw material. Known
examples of industrial thermoprocessable
cellulose products applicable for melt spinning
are plasticized cellulose acetate and cellulose
acetate butyrate (CAB). However, these fibres
are still mainly produced via spinning from
acetone solution due to the different challenges
associated with their melt spinning.
Highvolumesofabsorbentfibresandnonwovens
are used in producing hygiene products. In
addition, superabsorbents (SAP) are used in
high quantities in applications where high water
absorption and/or water retention capacities are
needed. Crosslinked polyacrylates, which can
typically absorb 40 000–50 000% of distilled
water and 4000-5000% of 0.9% saline solution
by weight, are commonly used as SAPs in
hygiene products. New and innovative personal
care products require increasing amounts of
sustainable absorbent materials in different
application forms. Fluff pulp is currently widely
used as an absorbent in feminine hygiene
products and nappies. The absorption capacity
of fluff pulp cannot, however, compete with that
of commercial SAP materials. If the fluff pulp
absorption capacity could be increased, the use
of SAPs could be reduced respectively. As SAP
materials generally produced from oil-based
polymers are significantly more expensive than
fluff pulp, even partial replacement of SAPs by
fluff pulp could positively affect both the cost
and sustainability of the final product.
Fibre materials applied in disposable
nonwovens are typically synthetic PET and
PP combined with certain cellulose fluff
pulp fibres. Rayon, once a common fibre in
nonwovens, has now been largely replaced
by synthetic fibres. Synthetic fibre blends
are wet-laid along with cellulose for single-
use fabrics. Growing concern regarding the
sustainability of disposables has led to the
creation of new biopolymer-based fibres
that offer more environmentally responsible,
performance-designed alternatives to the
traditional oil-based fibres currently used in
nonwovens manufacturing today. In technical
applications, synthetic fibres are also being
replaced with natural fibre, such as hemp or
coir. The challenge of using cellulose fibres
in nonwovens is their cost compared to
synthetic fibres. Thus, the identification of
suitable end product applications is key when
aiming for increased cellulose incorporation in
nonwovens.
Cellulose and modified cellulose matrixes
can be used as carrier material for specific
functionalities and for their controlled release.
Cellulose beads, for example, can be used for
various chromatographic and ion exchange
purposes. The challenge of functional cellulose

beads is their biodegradability and techno-
economic feasibility. Current solutions based
on synthetic materials are, however, relatively
expensive, which can provide a competitive
edge for cellulose beads. In addition, benefits
can also be found in the ability to produce
biocompatible and very pure cellulose materials
suitable, for example, for medical applications.
2. Objectives
The main objective was to develop i) functional
cellulose bead structures from novel cellulose
starting materials, ii) nonwovens with cellulose
adsorbents aiming at minimum 4 000-5 000%
wateruptakeforhygieneproductsandii)extruded
paper laminates and/or fibres from thermoplastic
cellulose for nonwoven structures. The work
aimedatdemonstratingtheperformanceofnovel
cellulose-based materials in target applications
and facilitating their feasibility evaluation in WP5.
The focus was to develop novel absorbent
cellulose materials for wiping and hygienic
applications and thermoprocessable celluloses
for melt-spinning and extrusion coating by
industrially feasible methods.
Various routes for making thermoprocessable
celluloses and cellulose water-absorbent
materials were evaluated and critical technical
parameters for material development were
identified and investigated. Optimization and
up-scaling of the most potential syntheses
were carried out and the material applicability
for the target end-use applications was
evaluated. Economic evaluations were carried
out to support the material development and to
illustrate the technical feasibility and economic
viability of the optimized materials in the
selected end-use applications.
Thermoprocessable celluloses consisted of
novel cellulose esters and synthesized cellulose
ethers as well as commercial cellulose acetates
(CA) and cellulose acetate butyrates (CAB)
as reference materials. Absorbent cellulose
materials were prepared at laboratory scale
by grafting hydrophilic monomers and/or
allylated xylan to allylated cellulose fibres by
TEMPO oxidation of the fibres or by dissolution
and coagulation of cellulose as bead particles.
Domsjö dissolving pulp was used in most cases
as raw material, but high molecular weight
Borregaard dissolving pulp, never-dried kraft
pulp and Sigma’s commercial α-cellulose were
also evaluated as raw materials. Mechanical,
enzymatic and/or chemical pre-treatments
were used as pulp activation methods to
overcome the low pulp reactivity generally
associated with the Domsjö dissolving grade
pulp. Synthetized products were purified before
material characterizations in order to remove
chemical waste and unreacted reagents.
Water absorption capacities were measured
from air-dried and freeze-dried samples using
standardized methods.
Nonwoven samples were prepared using
two alternative web forming methods. Base
structures were prepared from reference
materials and novel absorbent materials
using air-laid or foam-forming processes
and the applicability of these methods for
making controlled structures was evaluated.
The intermediate structures were combined
with a polyester web by hydroentangling on
a lab-scale pilot line at Suominen Nonwovens
Ltd. targeting 50:50 blends of absorbent
material:polyester. The final samples were
characterized (e.g. mechanical properties) and
their performance was assessed against the

prepared reference materials, which imitated
commercially existing products.
Beside nonwoven structures, novel fluff pulp
compositions with improved water absorption
capacities for use, for example, in nappies
were targeted. Fluff pulp is typically produced
via hammer milling of cellulose web. The
same method was also used to prepare fluff
pulps from foam-formed absorbent sheets. It
was expected that the foam-forming method
would be more suitable for absorbent material
processing than conventional papermaking
technologies. Finally, water absorption
capacities and water retention capacities were
measured from the samples.
Melt extrusion and melt spinning of selected
commercial and novel thermoplastic cellulose
derivatives were carried out with a micro
compounder and laboratory-scale melt
spinning line. The spinnability was studied
by increasing the take-up velocity. The as-
spun filaments were separately hot-drawn in
an oven instead of the heated godet on the
melt spinning line due to the small amount
of polymer used. The fibre properties were
characterized by an optical microscope and
combined linear density and tensile tester
equipment. The target of the small-scale and
pilot-scale melt extrusion trials was to assess
the processability of commercial and novel
thermoprocessable cellulose compared to
currently used synthetic plastics, such as PP,
and to produce continuous film and moulded
structures from novel thermoprocessable
celluloses.
Cellulose bead structures were prepared
by dissolving cellulose in environmentally
friendly NaOH/urea solvent and coagulated
via the sol-gel process in anti- or non-solvent
using dropping or spinning drop atomization
techniques. The bead structure design process
included selection of flocculation media and
flocculation conditions, determination of
the cellulose concentration of the solutions,
functionalization of the bulk and surface of the
beads, and control of micro- and mesopores
and bead shape and size.
Chemical modification of the cellulose beads
can be done either before or after coagulation.
Both methods were applied. Heterogeneous
modification was studied more intensively
due to better stability and higher content of
functional groups.
To evaluate the applicability of cellulose beads
as slow-release drug carriers, the native and
chemically modified beads were loaded with
two model drugs, freely soluble (riboflavin
5`-monophosphate sodium salt, lidocaine
hydrochloride monohydrate) and poorly
soluble active pharmaceutical ingredients,
APIs (griseofulvin and piroxicam). In addition,
anionic beads were loaded with a cationic
drug (Ranitidine HCl). Incorporation of model
drug substances was achieved by immersing
unloaded water-swollen beads in a solution of
the drugs. In-vitro drug release with the loaded
and dried beads was performed according to
the USP paddle method. The drug content of
the beads and the amount of drug released
from them was investigated with a UV/Vis
spectrometer to determine the loading efficacy
and drug release mechanism. The drugs
incorporated in the beads were investigated
in the solid state with field emission scanning
electron microscopy (FE-SEM) and Fourier
transform infrared spectroscopy (FTIR)
to determine the crystallinity of the drug
substances and the reasons for the different
release profiles. The drug distribution in the
beads was studied with a hyperspectral near-
infrared (NIR) imaging device to clarify the drug
release profiles.

Figure 1. Acetate laurate before and after melt-compressing at 200 °C. The formed film was transparent,
homogenous and brittle.
!
4. Results
4.1 Thermoprocessable cellulose
materials
4.1.1 Synthesis and processing of materials
Thermomeltable cellulose esters and ethers
were prepared using several synthetic routes.
The aim was to obtain materials suitable for
applications such as melt spinning and extrusion
coating. Cellulose esters with specific end-
product properties were prepared up to the
maximum theoretical degree of substitution
(DS 3). At least partial melting of materials by
hot compression was observed in most cases,
but the homogeneity of the materials was not
fully comparable to commercial thermomeltable
cellulose acetates and cellulose acetate
butyrates, which were studied as references.
Generally, cellulose hexanoates and cellulose
laurates resulted in ductile melt-compressed
translucent films when the degree of substitution
was over 1.0. An example of the melt-compressed
film is presented in Figure 1. Mixed esters of
cellulose acetate hexanoates and cellulose
acetate laurates typically contained a high
number of acetate groups and only a low number
of long-chain esters (degree of substitution, DS,
0.1-0.5). Figure 2 presents the DSC scans for
the sequentially esterified cellulose hexanoate
acetate sample (DStotal 2.0).
The raw material used for syntheses had a
very significant impact on end product quality;
especially materials made from high cellulose
molecular weight dissolving grade pulp
displayed very poor thermal melting, whereas
esters from commercial α-cellulose were very
homogeneous and resembled the commercial
references. The synthesis product quality
was slightly improved by pre-treatments. The
molecular weight (Mw) and polydispersity (PD)
of the materials were at the same level or lower
than the references. It appears that uneven
distribution of acyl substituents caused by
irregularities in the starting pulp materials and
possibletransglycosylationreactionsmaypartly
explain the synthesis product heterogeneity.
The current methods for determining DS
values are not able to distinguish differences
in substitution between the crystalline and
amorphous regions of cellulose. Especially in
the case of mixed cellulose esters, the order in
which the substituents are added to the pulp
cellulose can affect the end-product properties
remarkably. Regioselectivity is typically
determined by the size of the substituent, and
large acyl groups, such as laurate groups, may
not be able to react with highly crystalline
regions of cellulose due to steric hindrance. It
was observed that sequential and simultaneous
addition of acetate and long-chain fatty acid
substituents resulted in remarkably different

synthesis products. The order in which the acyl
substituents were added was also critical for
the thermal behavior of a sample.
The thermoplasticity and melt processability of
the cellulose esters of hexanoate and laurate
as well as the sequentially esterified celluloses
(Acet-Hex and Hex-Acet) were tested with
a twin-screw microcompounder at 170 and
200 °C. The cellulose ester was fed into the
compounder and mixed for 5 min prior to
forming a homogenous melt. The melt was
extruded through a 2 mm diameter circular die.
For the cellulose esters 170 °C was generally
found to be too low a temperature to obtain
desirable melt formation. Good melt formation
was obtained with all cellulose esters and
when the temperature was raised to 200 °C
the esters exited the microcompounder in rod
form. The cellulose hexanoate and laurate
samples were opaque and broke when bent. The
sequentially esterified celluloses showed better
melt formation than the other cellulose esters.
The sequentially esterified cellulose (Hex-Acet)
resulted in a translucent and ductile rod, whereas
the sequentially esterified cellulose (Acet-Hex)
resulted in a translucent but brittle sample
(Figure 3A and 3B). Interestingly, the sequentially
esterified (Hex-Acet) melt was possible to draw
as a fibre by hand. The rod diameter decreased
from 1.3 mm to a 0.1 mm fibre by drawing (Figure
3C). This indicates that the sequentially esterified
cellulose hexanoate-acetate (Hex-Acet) is
suitable for the melt spinning process. Based
on these observations, sequential esterification
enhanced thermoprocessability, unlike single
esterification.
Etherification of cellulose was not successful
using the method implemented to a high DS
level (highest level achieved DS 0.3). The dry and
hornified pulp sheets may require novel types
of chemical or mechanical activation, such
as strong swelling and/or partial dissolution,
before higher DSs and reaction efficiencies can
be achieved using the etherification method.
Melt spinning of commercial thermoplastic
celluloses
The quality of the synthetized materials
was not sufficient for extrusion coating and
melt spinning processes and, therefore, only
Figure 2. DSC (1st
and 2nd
heatings) thermograms for the cellulose hexanoate acetate before and after
thermal processing with a microcompounder at 200 °C . Thermal processing at 200 °C had no effect on
the glass transition temperature of cellulose hexanoate acetate.

Figure 3. The sequentially esterified celluloses processed at 200 °C: A) cellulose acetate hexanoate and
B) cellulose hexanoate acetate (1.3 mm) and C) hand-drawn cellulose hexanoate acetate fibre (0.1 mm).
commercial materials were used for making
demonstration products. The melt spinning
trials were carried out with a laboratory-scale
melt spinning line. Two commercial cellulose
derivatives, cellulose acetate (Plastiloid CA)
and cellulose acetate butyrate (Sigma Aldrich
CAB Mn 70 000), were used for the spinning
trials. The spinnability of cellulose acetate
was poor. The broad melting point caused a
considerable gas formation at the spinning
temperature (225 °C). The obtained filaments
were weak and the maximum take-up speed
was only 30 m/min. The filaments were thick
(160 μm) due to the slow take-up speed and
low drawing of the filaments. Melt spinning
of CAB was easier. Gas formation was rather
low at the spinning temperature of 220 °C.
The obtained filaments had better spinnability,
and the maximum tested take-up speed was
800 m/min. The spinning trials showed that
the spinning velocity of CAB could be even
higher than 800 m/min. The diameter of the
CAB filaments was about 25 μm, comparable
to commercial textile fibres, but can be further
decreased if needed. The visual appearance
of both cellulose derivatives was typical of
melt-spun fibres. The spinning temperature
had an influence on the mechanical properties
of the CAB. The maximum tenacity value
was 1.2 cN/dtex for fibres spun at 220 °C and
only 0.4 cN/dtex for fibres spun at 240 °C.
The tenacity values can be increased by hot-
drawing. The maximum obtained tenacity
value of subsequently hot-drawn CAB fibre
was 6.4 cN/dtex, which is comparable to
melt-spun polypropylene fibres. The hot-
drawing trial as a spin-drawing process (with
heated godets) was not as successful as the
subsequent process. The tenacity value of
the CAB fibre was only 0.7 cN/dtex, indicating
that optimization of spin-drawing process
conditions (godet temperature, velocity, linear
density) would be needed. In melt spinning, the
requirements for novel cellulose derivatives
are narrow melting temperature, stability at
melting temperature, high molecular weight,
and narrow molecular weight distribution. This
melt spinning work is one of only a few studies
reported in the literature based on cellulose
derivatives.
A B
C

Figure 4. Modelling scope for thermoplastic cellulose esters.
1) Kraft pulp 1
2) Kraft pulp 2
3) Dissolving pulp
4) Sulphite pulp
Cellulose
acetate/butyrate
process
Lignocellulosic
feedstock / Sugars
Melt
spinning
Film
extrusion
Modelling scope
VARIABLES
1. Electricity price
2. Oil price
3. Chemicals, concentration
and price
4. Biomass price
5. Pulp price
6. Hemicellulose price
Thermoplastic
cellulose
structures
(granules)
PP
PLA
Crude oil
Crude oil PE
REFERENCE MATERIALS
Moulding
End-use examples:
4.1.2 Techno-economic modelling of
thermoplastic cellulose
The objective of the modelling was to
evaluate the techno-economic feasibility
of thermoplastic cellulose in melt-spinning
applications. The thermoplastic cellulose
materials of focus were cellulose esters,
more specifically cellulose-acetate-butyrate
(CAB), cellulose-acetate-hexanoate (CAH) and
cellulose-acetate-laureate (CAL). The modelling
scope for material and energy balances
and quantitative modelling was limited to
thermoplastic cellulose granules. Common
melt spinning materials polyethylene (PE),
polypropylene (PP) and polylactic acid (PLA)
were selected as reference products (Figure 4).
The commercial cellulose acetate process
was used as a starting point for the modelled
production concept (Figure 5). For all studied
cellulose esters, raw materials constituted the
largest part of the costs. The total production
costs and the share of raw materials decrease
with lower degree of substitution and with
higher share of acetyl groups of total acyls. In
this analysis, the modelled production costs of
thermoplastic cellulose esters were well above
the market prices of the fossil-based reference
products PE and PP.
The techno-economic modelling task also
included an analysis of how the projected
oil price development would alter the cost
competitiveness of the studied cellulose
esters. Although the correlation with oil price
development is not as strong in the case of
cellulose esters as it is in the case of PP, PE or
even PLA, increasing oil price will also increase
the price of thermoplastic cellulose. Therefore,
increasing oil price is not expected to improve
the cost competitiveness of cellulose esters
considerably.
Although the modelled production costs of
thermoplastic cellulose esters exceed the
prices of commodity polymers, the costs
are not prohibitive. Figure 6 summarizes the
strengths, weaknesses, opportunities and
threats of cellulose esters in melt spinning
applications.

Figure 5. Block-flow diagram of cellulose acetate derivative production processes.
Activation
Esterification
Stopping
Acid Recovery
Hydrolysis
Precipitation
Washing
Press & Drying
Pulp
Butyric/
Hexanoic/
Lauric acid
Acetic acid
Acetic acid
Acetic anhydride
Waste water
Water
Water
CAB/CAH/CAL
Anhydride prod
Figure 6. SWOT analysis of thermoplastic cellulose in direct melt spinning.
Helpful
to
achieving
business
success
Harmful
to
achieving
business
success

Process

related

STRENGTHS

•  Exis%ng
commercial
process
(cellulose-‐acetate)

as
a
pla6orm
for
a
new
product.

•  Would
open
melt
spinning
process
for
cellulosic

materials.

WEAKNESSES

•  Product
is
not
(and
is
unlikely
to
become)
cost

compe%%ve
with
currently
used
melt
spinning

polymers
PE,
PET,
PP,
PLA.

•  Compa%bility/suitability
of
studied
material
for

melt
spinning
process
is
unknown.

Business

environment

related

OPPORTUNITIES

•  Demand
for
melt
spinning
products
is
increasing.

•  Possibili%es
for
improved
product
proper%es,

and
thus,
new
end-‐use
applica%ons.

THREATS

•  High
subs%tu%on
poten%al
from
compe%tors

(PLA,
Bio
PE,
etc.).

•  Nonwoven
industry
is
very
consolidated
with

only
few
players.

4.1.3 Markets and business opportunities for
thermoprocessable cellulose
Market assessment of thermoplastic cellulose
concentrated on five selected application
areas divided into two categories: “short-term
cases”, which focus on large volume end-uses
where market entry is relatively simple, and
“long-term cases”, which represent end-uses
where market entry is more complicated or the
product development time is expected to be
long (Figure 7). The short-term end-uses include
blister and other high-visibility packaging,
shrink sleeve labels and films used in coated
nonwovens, whereas long-term applications
include food contact packaging with strict
regulation requirements and cellulose
nonwovens through direct melt spinning
where the technical material requirements are
challenging to meet.
Generally, the market opportunities for the
product groups studied are lucrative. All
examined markets are growth markets with
an interest in non-food based bioplastics.
However, cost competitiveness remains a key
challenge in all end-use sectors. Thermoplastic
cellulose materials should be aimed at higher
value applications instead of as a substitute for
commodity polymers, such as polyethylene or
polypropylene.
A blister can be defined as a local partition
of a surface layer that causes a raised area
on a flat surface that can hold items. The
three main end-use segments for blister
and other high visibility packaging are food,
pharmaceuticals and consumer goods, such as
toys and tools. Blister and other high-visibility
packaging meets current product marketing
needs extremely well. Being able to see what
you buy is still considered one of the most
important marketing instruments, particularly
in consumer goods. Despite fierce competition
in the packaging sector, blister and other high-
visibility packaging is winning market share
from other packaging solutions.
Shrink sleeve labels are film tubes that are
applied over the head of a container and shrunk
to the container shape using heat, hot air or
steam. A shrunk-on label can be applied just
to the shoulders or to the cap of the container,
or it can cover the entire product to give 100%
promotional area. This possibility is of particular
importance in the food and pharmaceutical
segments, where the amount of compulsory
regulatory information on labels is ever
increasing. Shrink sleeve tubes can be used to
label, for instance, glass and plastic containers,
aluminium cans, contoured packages or chilled
and frozen products. Shrink sleeve labels are
high-profile promotional tools and the fastest
growing labelling category.
The film-coated nonwoven market is extremely
performance oriented. Film is applied on top of
the nonwoven to gain properties unattainable
bythenonwovenorfilmonitsown.Performance
Figure 7. Selected application areas for the
thermoplastic cellulose market assessment.
Shrink sleeve
labels
Food packaging
Film coated
nonwovens
Cellulose
nonwovens
through direct
melt spinning
Blister and other
high visibility
packaging
Approximate
market size
Pöyry Management
Consulting Oy

depends on the chemical formulation, coating
thickness and weight, the number of layers, the
form of the technical textile and the nature of
any pre-treatments. Currently, there are very
few bio-based materials in use in the coated
nonwoven market. Increasing environmental
concerns are generally tackled by reducing
material consumption or replacing harmful
substances, such as PVC.
Food packaging is a promising market for bio-
based materials, but also challenging due
to strict food contact regulations. Material
requirements depend strongly on both the
packaging design and the type of food. For
instance, confectionery boxes have very
different packaging requirements from chilled
ready meals. Changes in the global diet towards
more meat and dairy are having their effect on
the food packaging market. Overall, the market
is strongly driven by consumer behaviour.
Spunlaid nonwovens and bioplastics are the
fastest growing segments with compound
average growth rates of (CAGR) almost 10%
per annum. Annual growth rates of both short-
and long-term end uses are summarized in
Figure 8. Today, cellulosic fibres cannot reach
almost half of the nonwoven market due to
technical incompatibility. Direct melt spinning
makes possible the combination of fibre
production, web-forming and web-bonding
in a continuous single-step process with
much lower production costs and enhanced
efficiency than, for example, in viscose-
based nonwovens. However, the technical
fibre properties are challenging to meet with
cellulose-based thermoplastics.
The value chain analysis looked at the blister
packaging value chain, which starts with the
raw material producer and plastic manufacturer
and continues with converter, brand owner and
retailer. The converter and brand owner can be
horizontally integrated and, in some cases, brand
0% 2% 4% 6% 8% 10%
Spunlaid
nonwovens
Bioplastics,
total
Film coated
nonwovens
Shrink-sleeve
labels
Bilster and other high
vis. Packaging
Packaging,
total
Food
packaging
Annual CAGR
Long‐term
Short‐term
Pöyry Management
Consulting Oy
Figure 8. Annual compound average growth
rates of selected end-use markets.
ownersoutsourcetheirentirepackagingfunction.
The largest companies by turnover are found at
both ends of the value chain, whereas the middle
is characterized by a large number of small
and highly specialized producers. Raw material
producers and brand owners create most value in
the high-visibility and blister packaging sectors.
The magnitude of the captured value varies
between end-use industries.
The blister market is highly price conscious,
and there is little or no willingness to pay a
premium for bio-based packaging in large-
scale applications. New bio-based materials
should be compatible with existing converting
equipment, as there is low interest in developing
and investing in new converting lines. The food,
pharmaceutical and toy industries have strict
laws and regulations, making market entry
more difficult. However, the market is large
and growing with a growing packaging trend
towards more sustainable solutions.

Figure 9. Degree of polymerization as a function of time and temperature. Optical images demonstrate
the dissolution mechanism in diluted CED solution.
4.2 Cellulose beads
4.2.1 Preparation and application of cellulose
beads
Preparation of physicochemically designed
beads and anionic beads
HyCellSolv pretreatment was developed for the
production of cellulose beads from different
wood pulps. Dissolving pulp was pretreated with
acidic ethanol liquor (HyCellSolv-liquor) using
different treatment times and temperatures
(Figure 9). After 2 h at 75 °C the pulp was
soluble in 7% NaOH-12% urea-water so that
the solution was clear without undissolved
fragments. Cellulose was thus dissolved in
water-based solvent without undissolved
fragments after HyCellSolv pretreatment.
By controlling the coagulation kinetics it was
possible to physicochemically functionalize
cellulose beads. A 4-6% cellulose solution was
coagulated dropwise in nitric acid of different
temperatures and concentrations, as well as
in salt water. Physicochemical modification by
controlling the coagulation kinetics provided
the beads with different pore size distributions
and surface areas (Figure 10).
Beads prepared from 5% cellulose solution
in 2 M HNO3
at 25 °C were oxidized by the
TEMPO/NaClO2
/NaClO system. The main
oxidizing component (NaClO2
) had a molar
ratio of ~
1.2 per anhydroglucose unit (AGU) of
cellulose. Oxidation with the TEMPO/NaClO2
/
NaClO system yielded higher charge than with
meta-periodate or blending with CMC and
the beads were also more stable. The highest
charged measured for the oxidized cellulose
beads was 1848 μmol/g.
Beads as drug carriers
The applicability of cellulose beads as slow-
release drug carriers was evaluated by loading
the native and chemically modified beads with
two model drugs, freely-soluble (riboflavin
5`-monophosphate sodium salt, lidocaine
hydrochloride monohydrate) and poorly-
soluble active pharmaceutical ingredients,
APIs (griseofulvin and piroxicam). In addition,
anionic beads were loaded with cationic drug
(Ranitidine HCl). Figure 11 describes the loading

Figure 10. Effect of (A) temperature, (B) acid concentration and (C) cellulose concentration on specific
surface area of the CPD cellulose beads. General coagulation conditions were: 5% cellulose solution
coagulated into 2 M HNO3 at 25 °C.
Figure 11. Unloaded and loaded drugs and their morphology.
Drug loading
Cellulose beads
in drug loading solution
Drying
Empty water
swollen beads
Dried and
loaded beads
Loading and drying at
room temperature
Content
Analysis
UV/Vis
Swollen CBs crushed and
immersed into 10 ml water
solution and stirred for 24h
FE-SEM & FTIR
Field emission scanning electron
microscopy
NIR imaginig
(SPECIM MCT based
Spectral Camera)
UV/Vis
- USP paddle method
- 0.1 N HCI, @ 37, 100 RPM
- 4-20 beads per vessel
Surface
and Interior
Morphology
Drug
distrubition
Drug release
rate studies

Solubility of drug
substances
Drug substance Type of CBs Drug content (%)
Freely soluble drugs RSP Non-ionic CBs T1 12.7
T2 13.0
T3 14.3
LiHCl T1 23.2
T2 26.6
T3 27.3
Sparingly soluble drug Thp T1 3.7
T2 4.2
T3 5.0
Poorly soluble drugs Piroxicam
Griseofulvin
T2 10.8
T2 22.1
Cationic drugs Ran HCl Anionic CBs
Non-ionic CBs
16.1
20.1
Quinine Sulphate Anionic CBs
Non-ionic CBs
3.3
11.8
Table 1. T1, T2 and T3 refer to different cellulose bead types with different physical properties (porosity;
T3>T2>T1). CB=cellulose beads, RSP=riboflavin 5'-phosphate sodium, LiHCl=lidocaine hydrochloride
monohydrate, Thp=anhydrous theophylline, Ran HCl=ranitidine HCl.
procedure and characterization methods for
the cellulose beads and presents FE-SEM
pictures of unloaded and loaded beads.
Drug loading studies were performed with
various different compounds and several
types of CBs (different charge, porosity,
etc.). Table 1 summarizes the loaded drug
substances, cellulose bead types and drug
loading efficacies. Drug loading is dependent
on the concentration of the drug loading
solution, drug choice and the properties of
the beads. Table 1 shows that drug loading
increased with high porosity and anionic
charge of the beads (for cationic drugs).
The release of freely soluble drugs was
controlled with physicochemically designed
beads (Figure 12). In addition, the amount
of drug release was doubled with anionic
cellulose beads (Figure 13). However, the
release profile of poorly soluble APIs could not
be improved with beads due to shrinkage of
the beads during the drying stage.
Cationic Ranitidine hydrochlorine was used as
a model drug in a study of release profiles from
oxidized cellulose beads. The release profiles
were noted to be constant regardless of the
bead charge, ambient pH, or bead swelling
rate. Compared to native cellulose beads,
oxidized cellulose beads could carry twice as
much drug, and the drug was observed to be in
amorphous form. This property could be utilized
for the delivery of poorly soluble substances.
Additionally, the loaded and placebo beads
demonstrated high mass uniformity, indicating
a good capacity for personalized dosing of
patients.
CMC-cellulose beads with a ratio of 2:8 were
prepared using high DS CMC (DS 1.15-1.45). The
total polymer concentration of the solution
was 5%. Reference beads (5% cellulose, no
CMC) and CMC-beads were loaded with three
different model drugs and the release profiles
of drugs and drug-polymer interactions were
studied.

Figure 12. Release profile of RSP-loaded beads.
Figure 13. Cumulative release of Ranitidine HCl from non-oxidized (reference) and oxidized (20-60 °C)
cellulose beads.
Anionic CMC-beads can be used to delay
drug release. Also, higher amounts of poorly
soluble drugs can be incorporated in anionic
CMC-beads. The release profiles showed
an initial “burst” release, mainly due to
unbound drug, followed by a subsequent more
controlled release of bound cationic drugs
from anionic CMC-beads. Also poorly soluble
drugs demonstrated controlled release after
an initial burst. This can be explained by slow
diffusion and solubility.
Cellulose beads and oxidized cellulose beads
have excellent properties as drug carriers.
They demonstrated high mass uniformity
and high loading capacity. Drug release
was constant, regardless of environmental
changes, such as pH.
Adsorption of metal ions on beads
Cellulose beads contain acidic groups, which
were studied by potentiometric titration. The
titration data was evaluated by the FITEQL
software, giving detailed information about the
different acidic groups on the cellulose beads
and modified cellulose beads. Modified cellulose
beads had more than ten times the amount of
acidic groups than cellulose beads (Figure 14).

Figure 14. Potentiometric acid titration for cellulose and cellulose derivative with acidic group.
Figure 15. Preparation of cellulose with anionic cellulose derivatives (left). Concentration of metal ions in
the collected fractions as a function of elution volume for a chromatographic column filled with cellulose
beads (right).
The Domsjö dissolving pulp was treated with
HCl and ethanol to eliminate any lignin residue.
Cellulose with 3-sulpho-2-hydroxypropyl
groups was inserted during preparation of the
cellulose beads. The new modified cellulose
can be used as a cation exchanger, a unique
characteristic that can be used to achieve
better and higher sorption. In these studies,
cellulose beads were used as a stationary
phase in column chromatography in order to
study metal ion affinities. The mechanism is
mainly ion exchange by complexation of metal
ions to the cellulose, which contains carboxylic
groups as a functional group. It was observed
that divalent ions show better sorption than
monovalent ions (Figure 15).
cellulose beads
Cellulose beads are porous spherical cellulose
particles with diameters in the micro- to
millimetre scale. Cellulose beads can be
functionalized by introducing different organic
or inorganic materials to the bead structure.
Depending on the derivatization agent,
cellulose bead properties can range from, for
example, steady drug release to rapid water
absorption.
There are thousands of potential applications
for cellulose beads with such functionalization
capacity. Cellulose beads have been
commercially available for 15-20 years,
but annual production volumes are very
0

2

4

6

8

10

12

14

12
17
22
27
32
37
42

pH

Volume
added
of
NaOH
(mL)

Blank

Cellulose

Cellulose
deriva8ve

Iontosorb

Poten5ometric
5tria5on

Poten8ometric
8tria8ons
were

performed
for
a
cellulose
sample
,
a

cellulose
deriva8ve
with
3-‐sulpho-‐
hydroxypropyl
group
and
a

cellulose
with
carboxylic
func8onal

group
(Iontosorb).

Cellulose
Cellulose
deriva5ve
Iontosorb

lgK
Concentra8on
lg
K
Concentra8on
lg
K
Concentra8on

2.8
112.8
2.7

94.3

3.7
750.6

4.5
6.8
5.4

17.9

4.7
538.2

6.1
2.5
9.6

45.0
8.7
25.4

total

122.1

157.2

1314.2

Table
5.
Protona8on
constants(lgK)
and
concentra8on
(µeq/g)
of
acid
groups
of
cellulose
beads,

cellulose
deriva8ve
and
cellulose
with
carboxylic
func8onal
group.

0

2

4

6

8

12
17
22
27
32
37
42

pH

Volume
added
of
NaOH
(mL)

Blank

Cellulose

Cellulose
deriva8ve

Iontosorb

Poten8ometric
8tria8ons
were

performed
for
a
cellulose
sample
,
a

cellulose
deriva8ve
with
3-‐sulpho-‐
hydroxypropyl
group
and
a

cellulose
with
carboxylic
func8onal

group
(Iontosorb).

Cellulose
Cellulose
deriva5ve
Iontosorb

lgK
Concentra8on
lg
K
Concentra8on
lg
K
Concentra8on

2.8
112.8
2.7

94.3

3.7
750.6

4.5
6.8
5.4

17.9

4.7
538.2

6.1
2.5
9.6

45.0
8.7
25.4

total

122.1

157.2

1314.2

Table
5.
Protona8on
constants(lgK)
and
concentra8on
(µeq/g)
of
acid
groups
of
cellulose
beads,

cellulose
deriva8ve
and
cellulose
with
carboxylic
func8onal
group.

Domsjö Cellulose
1 h to -15°C
Cellulose
beads
Cellulose derivative
Preparation of cellulose beads with acidic cellulose derivatives
Cellulose
+ Urea
+ NaoH
+ Water
Collected in 10%
HNO3
Cellulose
O
OH
SO3
Na
Cellulose O
O
ONa
0

0,5

1

1,5

2

2,5

3

3,5

4

4,5

5

0

0,02

0,04

0,06

0,08

0,1

0,12

0,14

0,16

0,18

0,2

40
60
80
100
120
140
160
180
200
220

C
,

mmol
/L

V,
mL

K

Li

Na

Ba

Ca

Mg

Sr

Cd

Zn

Ni

Mn

pH

pH

Potential end uses
Chromatography Composites Acoustic boards and panels
Metal ion-exchange and water
treatment
Growth medium Smart sponges
Protein immobilization Pollution recovery, e.g. oil Light-adjusting paint
Cosmetics Water damage clean up eInk Lite
Air purification Oil-water emulsion aid Bending/origami sheet
Drug loading and release Plaster Reactive textiles
Ammunition Active food packaging Cellusensors
Dietary/ nutritional supplements Mixing with CMC Cellubots
Fertilizers Absorbents Cellubricks
Feed additive Replacement of charcoal tablets Swallowable perfume
Solid-phase synthesis support
Table 2. Potential end uses for cellulose beads.
small. Cellulose beads are currently
used in niche applications such as ion-
exchange, chelating sorbents, dye-ligand
chromatography, hydrophobic interaction,
affinity chromatography, size exclusion
chromatography, filter material and core-
particles for pellets. In the market assessment,
the objective was to identify potential end-uses
for cellulose beads (both existing and novel)
and to analyse which of these seemed the most
favourable for commercial bead production.
The screening of possible applications was
based primarily on the unique properties
of cellulose beads, i.e. identifying end-uses
where cellulose beads could offer significant
advantages compared to competing solutions.
Keypropertiesforcellulosebeadcompetitiveness
include mechanical stability, narrow particle
size distribution, high chemical resistance and
compatibility with most commonly used solvents,
high temperature stability, high selectivity of
separation, excellent flow properties, chemical
reactivity in derivatization, non-toxicity, high
porosity and large surface area.
In the market assessment reference markets,
key drivers, annual growth rates and competitive
advantagesofcellulosebeadsinover30potential
applications were identified and innovated in the
screening stage (Table 2). These end-uses could
be divided into four main categories: consumer
beads driven by business-to-consumer (B2C)
markets, industrial beads driven by industrial
B2B markets, chemoactive beads, which refer
to laboratory-related end-uses, and “jokers”,
which can represent any end-use but with a
higher degree of unconventional elements.
The majority of the identified bead applications
were related to industrial end uses.
The identified end-uses were prioritized
based on three criteria: (i) market potential
(including reference market size and annual
growth), (ii) margin between reference price
of competing solution and estimated cellulose
bead production costs, and (iii) applicability of
cellulose beads in a given end-use. As a result,
eight potential end uses were ranked as hitting
the “sweet spot” with good market potential,
adequate margin and technical applicability
for the target end-use. These most interesting
applications included active food packaging,
solid-phase synthesis support, composites,
feed additives, cosmetics, growth mediums,
plaster and dietary supplements (Figure 16).

Figure 16. Eight end-uses were ranked in the
“sweet spot”.
Figure 17. Preparation of novel cellulose absorbents by grafting hydrophilic monomers onto allylated
cellulose fibre surface.
The majority of identified end-use opportunities
were completely new applications as opposed
to direct substitutes for existing products. All
in all, even if the required bead properties for
a specific application were achieved, cellulose
beads would still represent only a niche market
for the forest industry. The greatest incentive
for further development therefore most likely
lies outside the forest sector.
4.3 Novel absorbent materials for
hygiene products
4.3.1 Preparation and application of absorbent
cellulose materials
Production of novel absorbent materials
Various novel cellulose absorbent materials
were prepared at the laboratory scale by
grafting hydrophilic monomers and/or allylated
xylan to activated cellulose fibres (Figure 17)
or, alternatively, by TEMPO oxidation of fibres.
In addition, blends of specific materials were
also evaluated. Mechanically or enzymatically
pre-treated dissolving pulps and bleached
never-dried softwood kraft pulps (BNDS) were
mainly used as starting materials. Reaction
efficiencies in cellulose activations were
low, with a typical degree of substitution for
allylated fibres (DSallyl) of 0.05-0.10. TEMPO
oxidation was more efficient with a degree of
substitution for oxidized cellulose (DSoxidized)
of up to 0.2, which is close to the theoretical
maximum DS. Unlike most other oxidative
reactions, TEMPO oxidation is highly selective
to primary alcohol groups. This decreases
the maximum amount of carboxyl groups
introduced to cellulose drastically; however, it
also enables oxidation without disrupting the
crystalline structure of cellulose (Figure 18).
Absorbent materials were freeze-dried
before analysis of their capacity to absorb
water and 0.9 wt-% NaCl solution. The water
uptake values of the cellulose absorbent
materials generally varied between 10-40
g water/g, with the highest values obtained
with TEMPO-oxidized fibres, whose structure
was subsequently mechanically loosened.

Figure 18. Oxidation of surface anhydroglucose units of cellulose nanofibrils by TEMPO oxidation.
Figure 19. Absorbent properties of different structures.
The mechanical loosening was a prerequisite
for high water absorption capacity. Without
mechanical treatment water-absorption values
for TEMPO oxidized pulps were generally lower
than 10 g water/g absorbent. The significant
increase in water uptake capacity by light
mechanical treatment is apparently due to the
increase in available fibre surface. Even if the
chemical composition of the absorbents would
favour very high water sorption, the sorptivity
would remain low if the structure of the material
does not allow access of the water and swelling.
Thewateruptakevaluesofthegraftedfibreswere
maximum 15 g water/g fibre. Mild mechanical
disintegration did not significantly improve the
absorption capacity of the grafted materials.
Drying of absorbent fibres is crucial for
preserving material performance. As freeze-
drying may not be realistic at the industrial
scale, the applicability of foam forming for
the processing and drying of TEMPO-oxidized
absorbent fibres was evaluated. The absorbent
properties of different structure types are
illustrated in Figure 19. Absorption capacity was
lowest for paper-like structures and highest for
porous and bulky freeze-dried structures.
• 100 % oxidized cellulose
• Foam-laid structure
• Air-laid drying
• Film structure
• Softwood & oxidized cellulose
• Foam-laid structure
• Air-laid drying
• Paper structure
• 100 % oxidized cellulose
• Freeze-drying
• Porous and bulky structure
Improved (structural) absorption capasity
TEMPO

oxida,on

Glucose
unit

Oxidized
glucose
unit

Figure 20. Free swelling absorption capacity of cellulose-based materials when processing method was varied.
In addition to drying, further processing of
the material is essential, especially when
considering applications using fluff pulps. In end
product (e.g. nappy) manufacture, a hammer
mill is used for disintegrating the cellulose. In
the present small-scale studies, this process
was simulated by dry blending the materials
with a mixer. The results are shown in Figure
20. Absorption capacity increased in the case
of foam-laid papers (target grammage 80 g/m2
)
when the structure was dry blended.
Novel nonwoven structures
Novel spunlaced nonwoven structures
simulating commercial household wipes (50%
polyester and 50% pulp) were produced in
Suominen pilot line. Both air-laid and foam-
forming technologies were utilized in making
absorbent cellulose sheet structures. Air laying
was suitable only for processing fluff pulp
fibres, whereas foam forming was applicable
also for the production of sheet structures
from novel absorbent materials, among which
TEMPO-oxidized fibres were identified as the
most promising novel absorbent materials.
Foam-formed handsheets with different
ratios of softwood kraft pulp and TEMPO-
oxidized pulp were prepared. The aim was
to determine the optimum pulp composition
for producing nonwoven base structures.
The target structure needed to have as high
water absorption capacity as possible. Foam-
formed sheets (80 g/m2
) containing 80%
softwood kraft pulp fibre and 20% TEMPO-
oxidized fibre were used. The foam-formed
layers were combined with a polyester web
by hydroentangling, targeting 50:50 blends of
pulp:polyester. As a reference, a pulp:polyester
composition, using tissue sheets for the pulp,
was produced on the same pilot line.
The foam-formed structures provided a
much stronger pulp:polyester nonwoven than
the tissue-derived reference pulp. Also, the
decrease in wet strength was less for the foam-
formed product, even though all samples had
lower wet than dry strengths. The absorption
capacity was only marginally better with the
foam-formed nonwoven compared to the
reference. Both pulp-containing products had
clearly lower absorption capacity than the 100%
polyester nonwoven. The absorption capacity
of the fibres does not translate directly into the
absorption capacity of the nonwovens, as a lot of
the absorption is attributable to the void space
Dry-‐blended
Absortion Absortion Grammage
[g/g] [g/g] [g/m2]
%
/
Consistency
0.5% 27

/
Consistency
2.0% 28 18
ation
4

Softwood
50% 16 19 80

Softwood
70% 20 25 80
27

28

4

16

20

18

19

25

0

2

4

6

8

10

12

14

16

18

20

22

24

26

28

30

32

34

Tempo
oxidized

cellulose

100%
/

Consistency
0.5%

Tempo
oxidized

cellulose
100%
/

Consistency
2.0%

Tempo
oxidized

cellulose
100%

Tempo
oxidized

cellulose
50%
/

SoEwood
50%

Tempo
oxidized

cellulose
30%
/

SoEwood
70%

AbsorpIon
[g/g]

Freeze-‐dried
Air-‐laid
dried;
not
dry
blended
Air-‐laid
dried;
dry
blended

Target
grammage
80
g/m2

in the nonwoven structure. The pulp-containing
products make a much denser product
compared to the 100% polyester nonwoven,
which can be seen from the thickness of the
products at a given g/m2
. Clearly, it is not enough
to improve the absorption of the pulp alone; in
addition, the pulp needs to be in a favourable
nonwoven structure.
Improved fluff pulp
The applicability of TEMPO-oxidized pulp
for improving fluff pulp water absorption
properties was evaluated. Besides water
absorption capacity, also water retention
capacity, which is highly important for many
hygiene products, was characterized. Different
commercial superabsorbents were used as
reference materials. Foam-formed and air-dried
TEMPO-oxidized handsheets were defibrated
together with fluff pulp using a hammer mill.
Alternatively, TEMPO-oxidized fibres were
applied onto a fluff pulp web using a semi-pilot
scale coating device. The oxidized cellulose
containing fluff reel was also defibrated by a
hammer mill before material evaluation.
The results of the gravity-based analysis of
the absorption capacity of the absorbent
cellulose materials are given in Figures 21
and 22. According to the results, TEMPO-
oxidized pulp did not improve the free swelling
absorption capacity of fluff pulp and the
material performance was not comparable
with commercial superabsorbents. However,
gravity-based analysis may not be the most
suitable method for determining absorption
capacity.
Correspondingly, the increase in absorption
capacity under load was enhanced by 40%
when 100% fluff was compared to 100%
TEMPO. A range of different SAP grades are
available, and when comparing the absorption
under load of the TEMPO-based solution to the
commercial bio-based SAP, the difference was
only 25%, as seen in Figure 22.
The applicability of the novel fluff pulp
material in a hygiene product application was
demonstrated at Delipap Oy. The target was
to demonstrate the potential of bio-based
absorbent material – oxidized cellulose – in a
product application. The reference material
was conventional fluff pulp used in different
kinds of hygiene products. Foam coating was
used as the coating method and the Surface
Figure 21. Free swelling absorption capacity of absorbent cellulose materials.

0.9%
NaCl-‐liuos
[g/g] [%] [g/g]
Fluff
100% 22
Tempo
100% -‐100 19
/m2

Fluff
reel

&
Foam
coated
Tempo
2.5% 22 0
Fluff

50%
/
Tempo
50% 22 0
Fluff

50%
/
SAP
50% 33 50
Commercial
SAP
/
Fubio 44 100
ORB
T7061Commercial
SAP
BASF
HYSORB
T7061 40 82 40
Commercial
Bio-‐based
SAP 30 36 30
22
22
22

33

44

19

40

30

0

5

10

15

20

25

30

35

40

45

50

Fluff
100%
Tempo
100%

Fluff
reel

&

Foam
coated

Tempo
2.5%

Fluff

50%
/

Tempo
50%

Fluff

50%
/

SAP
50%

Commercial

SAP
/
Fubio

Commercial

SAP
BASF

HYSORB

T7061

Commercial

Bio-‐based

SAP

AbsorpJon
[g/g]

Figure 22. Absorption under load (0.3 psi) of absorbent cellulose materials.
Treatment Concept (Sutco) as the research
environment for the manufacture of oxidized
cellulose containing fluff reel. The product
demonstration was carried out on Delipap’s
hygiene products production line. The
demonstrated bio-based absorbent material
was a fluff reel containing 10% oxidized
cellulose. The product demonstration was an
anatomically shaped panty liner. The results of
the production-scale demonstration indicated
improved absorption capacity in the case of the
bio-based absorbent material.
4.3.2 Techno-economic modelling of bio-
based absorbents
The objective of the modelling task was to
evaluate the techno-economic feasibility of
bio-based and fully biodegradable absorbent
materials as replacements for fossil-based
superabsorbents and to increase the use
of cellulose materials in absorbent hygiene
products (Figure 23).
The main cost factor in bio-based absorbent
production was feedstock pulp, followed by
energy and chemicals. Chemical and catalyst
recycling rates and doses were the most
uncertain process parameters. Successful
chemical recycling could have a major
impact on the production economics. The
competitiveness of the studied absorbents
compared to commercial superabsorbents is
fully dependent on the absorption capacity,
which at the time of the modelling case was
inferior to commercial SAPs. The lower the
absorbent capacity, the higher volumes are
required. Because in the modelling scope
bio-based absorbents were defined as direct
substitutes for SAPs, the quantitative modelling
showed poor economic feasibility.
In the qualitative opportunity assessment, the
technicalavailability,politicalandhealth-related
feasibility, as well as the bio-based absorbents’
compatibility with the forest industry value
chain were all very positive, but challenges
arose from the absorbent markets and
technical feasibility. The technical feasibility of
the recycling processes, purification and drying
were still a major question mark. On a general
29
20
16
29
20 10

14

17

23

34

16

29

20

0

4

8

12

16

20

24

28

32

36

40

Fluff
100%
Tempo
100%
Fluff
reel

&

Foam
coated

Tempo
2.5%

Fluff

50%
/

Tempo
50%

Fluff

50%
/

SAP
50%

Commercial

SAP
/
Fubio

Commercial

SAP
BASF

HYSORB

T7061

Commercial

Bio-‐based

SAP

AbsorpKon
[g/g]

+
40%

+
60%

"Tempo
vs.
Biobased
-‐
25%"

Figure 23. Block-flow diagram of bio-based
absorbent production.
level, the hygiene industry is a challenging
market that is difficult to penetrate due to
the dominance of a handful of strong brand
owners, the large number of major absorbent
producers that are developing their own bio-
based absorbents, a limited willingness to pay
bio-premiums in the bulk nappy market, and
difficulty in demonstrating and communicating
the environmental benefits. The pros and cons
of bio-based absorbents in baby nappies are
summarized in Figure 24.
Figure 24. SWOT analysis of bio-based absorbents in baby nappies.
Helpful
to
achieving
business
success
Harmful
to
achieving
business
success

Process

related

STRENGTHS

•  Bio
and
non-‐food
based
raw
material,
increase
of

bio-‐based
content
in
nappies

•  Biodegradability

•  Growing
demand
of
all
absorbent
hygiene

products

•  Rela;vely
simple
process
with
high
yields

WEAKNESSES

•  Low
absorp;on
proper;es

•  Overall
performance
s;ll
unclear:
absorbent

capability
in
absorbent
core,
mixing

performance,
bulkiness
and
ability
to
distribute

liquids
to
prevent
SAP
gel
blocking

•  Unknown
recycling
and
drying
process

Business

environment

related

OPPORTUNITIES

•  Increasing
demand
of
eco-‐nappies

•  Possibili;es
in
other
end-‐uses,
especially
as

“improved
ﬂuﬀ”

THREATS

•  Bio-‐based
SAP
already
entering
the
market

•  Development
of
pulpless
core
and
other

technology
breakthroughs
such
as
nanocellulose

•  Contaminants
of
WP3
products
s;ll
unclear

•  Absorp;on
proper;es
remain
low

•  Recovery
process

proves
to
be
too
expensive

Figure 25. Market volume of hygiene absorbent products.
bio-based absorbents
Hygiene absorbent products dominate
the absorbent market. Today, over 80% of
fluff pulp and over 90% of superabsorbent
polymers (SAPs) are consumed in baby
nappies, training pants, feminine hygiene and
adult incontinence products (Figure 25). Fluff
pulp can also be found in airlaid, spunlaced and
other nonwovens, whereas there are numerous
specialty end uses for superabsorbents,
for example in agriculture, cable wraps and
packaging.
Two in three mothers in the US view disposable
nappiesasa“necessaryevil”andhaveexpressed
concerns about the environment, but not at
the expense of convenience. Only a marginal
consumer group uses reusable cloth nappies
in the West, despite the fact that absorbent
hygiene products alone make up as much as
2-3% of all municipal solid waste in Europe1
.
Overall demand for absorbent hygiene
products is driven by population growth, GDP
development, urbanization, ageing population,
growing middle-class and increasing time
pressure. For instance, in Japan, sales of
adult incontinence products have already
1 Edana Sustainability Report: Baby Diapers and Incontinence
Products (2005)
exceeded sales of baby nappies. Increasing
time pressure has led to the development of,
for example, pull-up training pants and heavy
adult incontinence products that can absorb
more than five litres of liquid.
There are four key trends shaping today’s nappy
development: sustainability, convenience,
demand for ultra-thin products and increasing
concern for product safety. Consumers
have gradually started to demand more
environmentally friendly products. Currently,
there is a vast number of different solutions
on the market including thinner, lighter and
more efficient nappies with less raw materials;
combinations of re-usable cloth nappies with
disposable absorbent pads; nappies with bio-
based materials and reduced carbon footprint;
and a variety of partly biodegradable nappies.
Whether the emphasis is on bio-based content
or biodegradability depends strongly on the
region and regional end-of-life solutions.
At present, there are no 100% bio-based or
biodegradable nappies on the market due to
the lack of substitutes for several fossil-based
components.
Almost all new hygiene product launches focus
on convenience. Examples include pull-up
training pants, body-conforming stretchable
products, adult briefs with flexible waist belts,
Fluff pulp
4.2 million tons
SAP
1.5 million tons
3.4 million tons
1.4 million tons
Adult incontinence
products
2.0 million tons
Hygiene absorbent
products
8.2 million tons
To other applications
0.1 million tons
To other applications
0.8 million tons
Nappies
4.6 million tons
Feminine hygiene
products
1.7 million tons

and light incontinence products designed for
everyday use. Ultra-thin products are win-win
solutions not only for consumers and brand
owners, but also for retailers. These lightweight
products are more comfortable, more
convenient to wear, require less space, bring
savings in logistic costs and reduce waste. For
retailers, ultra-thin products provide reductions
in valuable shelf and warehouse space.
In recent years, product safety has become a
top priority in the hygiene industry. Families are
concerned about, for instance, chemical safety
and possible traces of contaminants. Many
eco-branded products are therefore marketed
as “chemical free”, “containing less chemicals”,
or “certified free from harmful chemicals”.
Two such chemicals commonly perceived as a
possible threat or allergen are TBT (tributyl tin)
and latex.
The evolution of nappy composition in Figure
26 demonstrates how the introduction of
more efficient and lower cost superabsorbents
has resulted in lighter nappies with enhanced
performance, more superabsorbents and less
fluff pulp material. A typical modern nappy has
roughly the same amount of fluff pulp and SAP,
each representing about a third of the nappy
weight. In the past few years, the drive towards
ultra-thin products has led to a completely new
nappy design, the “pulpless” nappy. Pampers
Drymax nappies are one example of such a
product, with an absorbent core consisting
of SAP between nonwoven sheets instead of
bulky fluff pulp.
The nappy value chain consists of component
producers (such as fluff, SAP, nonwoven
and adhesive producers), converters, brand
owners and retailers. There is a great deal
of horizontal integration in the value chain:
many brand owners are backward integrated
to convert their own products, and more and
more retailers are launching their own nappy
brands. There are hundreds of operators in
the nappy industry, and yet innovation and
product development are led by only a few
multinational converter/brand owners. Two
leading brand owners, Procter & Gamble and
Kimberly Clark, represent together more than
half of all nappy sales.
Environmentally friendly nappies are still a
very small, but growing, segment. In addition
to leading brand owners having their own
Figure 26. Evolution of nappy composition 1987-20112
.
0
10
20
30
40
50
60
70
1987 1995 2005 2011
Average nappy composition [g/pad]
Other
Adhesives
Elastic back ear
Tape
PP
LDPE
SAP
Fluff pulp
Mass [g/nappy]
2
Modified from Edana Sustainability Report (2011)

eco-brands sold at a premium price, there are
smaller players focusing on online retail with
a significant market share of the eco-nappies
segment. Unfortunately, “greenwashing” is a
major issue in the hygiene industry, and thus
all environmental claims should be supported
by, for example, LCAs.
Nappy performance and absorbency are more
than the sum of the individual materials. In
addition to superabsorbents, each absorbent
core needs other materials (typically fluff
pulp) to distribute liquid into the structure
and preventing gel blocking. Because water
molecules are attracted to superabsorbents by
electrical charges, the absorbency is strongly
affected by electrolyte concentration. When
salinity is the most important factor reducing
the absorbency of superabsorbent polymers,
pressure has a similar role to fluff pulp. Hence
superabsorbent polymers are not only needed
to increase nappy absorption, but also to hold
the liquids under pressure.
the results
Heterogeneous synthetic routes for producing
thermomeltable cellulose esters from different
pulp materials at high pulp consistencies (15-25
wt-%) were demonstrated. Etherification, which
is typically carried out in aqueous conditions,
would be economically attractive and
industrially easy to adopt, but the experiments
showed that more basic research on cellulose
reactivity and synthesis development will be
needed before sufficiently high degrees of
substitution providing thermal melting of an
end-product can be obtained. The esterification
route was more efficient than etherification,
and thermomeltable cellulose esters were
obtained from various pre-treated and non-
treated cellulose pulps. The main raw material
was a dissolving pulp that can be regarded
as a significantly cheaper raw material than
typical high-purity dissolving pulps (acetate-
grade pulps) industrially used in cellulose ester
production.Thetargetedpropertiesforcellulose
esters, including good film-forming abilities
and melt-spinnable formulations were not fully
obtained, probably due to the inhomogeneity
of the materials and the hot-pressed (by static
laboratory press) film structures, which always
contained some visible clods. Melt extrusion
of the materials was, however, successful
indicating that these materials could be utilized
e.g. in injection moulding processes. The
degree of substitution (DS) of the materials
was theoretically sufficiently high to provide
completely homogeneous melts and it can
be speculated that the uneven distribution of
the ester substituents may be at least partly
explained by the raw material quality and
heterogeneous reaction conditions. The results
show the importance of raw material quality in
producing thermoprocessable materials.
The melt spinning studies provided new
insights regarding the property requirements
of novel cellulose derivatives. Melt spinning
offers a more economical and efficient method
compared to dry or wet spinning, which both
need a polymer solvent and a solvent recovery
system for the spinning line. The only main
environmental impact of melt spinning is the
energy required for extruder heating and
running the machine.
Cellulose absorbent materials can be easily
produced at kilogram scale, for example
by TEMPO oxidation. The results showed
that drying of chemically modified fibres
is challenging when the objective is to
maintain improved absorption properties.
The drying process can thus be considered
as a bottleneck in developing novel absorbent
materials competitive with currently used
superabsorbents, and the technology therefore
needs further development.

The preparation of air-laid nonwovens provided
new information on the requirements of the
different materials (fluff pulps and SAP) and
how to optimize the process parameters,
and highlighted key air-laying technology
development targets.
Cellulose dissolution is generally challenging
and untreated pulps cannot be properly
dissolved in aqueous alkaline solutions for
cellulose bead production. Controlling the
degree of polymerization and primary cell
wall rupturing enables the use of weaker
environmentally friendly solvents. Additionally,
an opened structure increases the penetration
of derivatizing reagents. Understanding the
roles of the different factors involved in the
preparation of cellulose gel-based products
enables the design of cellulose beads for
multiple purposes. Oxidation post-processing,
blending and physicochemical design during
bead coagulation are tools that can be utilized
to target certain functionalities. The knowledge
accumulated on process parameters and
the control of basic properties enables well-
established methods to be readily modified
for other functionalities, such as protein/
enzyme immobilization. More research is,
however, required to harness the full potential
of cellulose beads.
Pharmaceutical companies are increasingly
using more sophisticated excipients and blends
in order to defend against generic competition.
Greater use of so-called functional excipients
– which go beyond the traditional role of
excipients as a carrier for active pharmaceutical
ingredients (APIs) – is one of the key drivers
for growth in the excipients market. The
commercialization of cellulose beads might
thus be more feasibly pursued via excipient
manufacturers rather than pharmaceutical
companies. Both commercialization routes
should, however, be explored.

6. Networking
The research was carried out jointly by industrial and research partners. Table 3 presents the
research partners and their roles in this topic.
Partner Role
Glocell Qvantitative economic modelling
Metsä Fibre Industrial tutor. Providing industrial view insight to techno-
Pöyry Management Consulting Market study. Economic feasibility modelling. Business potential
evaluation
Stora Enso Industrial tutor. Steering of work related to thermoplastic
celluloses, material supply. Providing industrial view insight to
techno-economic and market assessments
Suominen Industrial tutor. Preparation and testing of nonwovens, steering
of experimental work. Providing industrial view insight to techno-
economic and market assessments
Materials Science
Extrusion coating and melt spinning of thermoplastic cellulose;
mechanical processing of fibres into nonwoven structures
Organic Chemistry
Research adviser
UPM-Kymmene Industrial tutor. Development of absorbent fibre materials,
steering of experimental work related to hygiene products.
Providing industrial view insight to techno-economic and market
assessments
VTT Syntheses and testing of thermoplastic celluloses and cellulose
absorbent material
Åbo Akademi
Fibre and Cellulose Technology (FCT)
Analytical Chemistry (AC)
Pharmaceutical Sciences (PS)
FCT: Dissolution of cellulose in water-based systems, preparation
and functionalization of cellulose beads, tailoring of beads for
applications in different value chains.
PS: Beads as drug carriers
AC: Chemical analyses
Table 3. Partner organizations and their roles

Publications
Gericke, M., Trygg, J. and Fardim, P. Functional
Cellulose Beads: Preparation, Characterization,
and Applications, Chem. Rev. 113, 2013:4812-
4836.
Trygg, J., Fardim, P., Yildir, E., Kolakovic, R.
and Sandler, N. 2014. Anionic cellulose beads
for drug encapsulation and release. Cellulose
21(3)2014:1945-1955.
Trygg, J., Gericke, M. and Fardim, P. 10.
Functional Cellulose Microspheres, in Popa,
V. (Ed.) Pulp Production and Processing: From
Papermaking to High-Tech Products, Smithers
Rapra Technology, 2013.
Trygg, J., Fardim, P., Gericke, M., Mäkilä, E.
and Salonen, J. Physicochemical design of the
morphology and ultrastructure of cellulose
beads. Carbohydr. Polym. 93, 2013:291-299.
Trygg, J. and Fardim, P. 2011. Enhancement of
cellulose dissolution in water-based solvent
via ethanol–hydrochloric acid pretreatment.
Cellulose 18, 2011:987-994.
Yildir, E., Kolakovic, R., Genina, N., Trygg, J.,
Gericke, M., Hanski, L., Ehlers, H., Rantanen, J.,
Tenho, M., Vuorela, P., Fardim, P. and Sandler,
N. Tailored beads made of dissolved cellulose
- Investigation of their drug release properties.
Int.J.Pharm. 456, 2013:417–423.
Presentations
Setälä, H. 2012. Novel materials based on wood
polysaccharides. BiPoCon 2012 conference,
May 27-31, 2012, Siófok, Hungary.
Setälä, H. 2012. Cellulose absorbents. FuBio
Cellulose seminar, 1st October 2012, Espoo,
Finland.
Posters
Rissanen, M., Wikström, L. and Lahti, J. 2012.
Commercial thermoplastic celluloses in melt
spinning and extrusion coating, FuBio Cellulose
seminar, 1st October 2012, Espoo, Finland.
Setälä, H. 2012. The use and preparation
of fibrous celluloses with 1-allyloxy-2-
hydroxypropyl substituents, 3rd International
Cellulose conference, Nov 8-13, 2012, Sapporo,
Japan.
Trygg, J., Kuzmanovski, G. and Fardim, P.
Up-scaling of cellulose beads manufacturing.
Poster presentation in FuBio seminar, August
27, 2013, Espoo, Finland.
Kolakovic, R., Redant, H., Trygg, J., Gericke,
M., Fardim, P. and Sandler, N. Porous cellulose
beads in drug delivery – comparison of anionic
and nonionic systems. Poster presentation in
FuBio seminar, August 27, 2013, Espoo, Finland.
Arroyo, J., Trygg, J. and Fardim, P. Targeted
applications of modified cellulose beads:
Chromatographic column and drug release.
Poster presentation in FuBio seminar, October
2012, Espoo, Finland.
Rissanen, M., Lahti, J. and Wikström, L.
Thermoplastic celluloses in extrusion coating
and melt spinning. Poster presentation in FuBio
seminar, October 2012, Espoo, Finland.
Theses
Redant, H. Cellulose beads in drug delivery –
comparison of anionic and non-ionic systems,
M.Sc. Thesis, Åbo Akademi University 2013.

CATIONIC
CELLULOSE BASED
CHEMICALS
CONTAC T PE RSON
Jonni Ahlgren, jonni.ahlgren@kemira.com
PAR TN E RS
Glocell
Kemira
Metsä Fibre
Stora Enso
University of Oulu
UPM-Kymmene

ABSTRACT
The high molecular weight biopolymers such as cellulose become more and more important
when alternatives for synthetic polymer raw materials for water soluble chemicals are
considered. Synthesis of uncharged derivatives such as hydroxyethyl cellulose and anionic
derivatives such as carboxymethyl cellulose are currently used in different commercial
applications e.g. as rheology modifiers and as process additives. On the other hand the
products from cationic derivatives of cellulose are practically non-existent. Cellulose
was used here as a raw material in the production of flocculating agents for paper and
wastewater treatment applications. Cationic water-soluble polymers and cationic nano-
scale particles were targeted.
It was shown that a water-soluble derivative can be made only if sufficient charge density
is achieved (about DS 0.5). Three interesting reaction routes to a cationic cellulose product
were identified, each of which requires further development towards commercialization.
It was shown that, in addition to dissolving pulp, also ordinary kraft pulp can be used as a
raw material for polymer synthesis. Hemicelluloses need not necessarily be removed, and
the cationic product quality is better if the pulp is not heat-dried before use. Softwood
performed better than hardwood, although hardwood also showed good properties.
The same cationic cellulose polymer is not suitable for all applications. Sludge dewatering
prefers high charge density, whereas retention and other flocculation requires high molecular
weight. The cationic particle performed relatively well in both applications. The cationic
cellulose derivatives as such did not perform as well as polyacrylamide in sludge dewatering
or in flocculation. In flocculation under high shear and in fixing applications, however, certain
cellulose derivatives exceeded the performance of the polyacrylamide reference.
Keywords:
cationic cellulose, cationic particle, cellulose betainate, CST, dissolving pulp, FBRM, GTAC,
kraft pulp, market analysis, never-dried kraft pulp, techno-economic modelling

1. Work background
Cationic flocculating polymers have an
important role in many industrial and municipal
applications, such as in papermaking as
a retention aid, in different wastewater
treatments as a flocculant, and in sludge
dewatering. Increasing environmental concern
limits the use of synthetic cationic flocculating
polymers, and alternatives to them are needed.
One good option for this is cellulose.
Wood cellulose has one of the highest molecular
weights of all natural polymers, i.e. biopolymers.
Cellulose is also widely available, being the most
abundant annually renewable biomass on the
planet. Cellulose also has an important role in
many industrial processes, such as paper- and
boardmaking and fibre production.
Cellulose in its raw state is not water soluble
and has no cationic charge and thus requires
modification before it can be used as a cationic
flocculating polymer.
Cellulose derivatization to water-soluble
products has long been known. Uncharged
water-soluble derivatives such as hydroxyethyl
cellulose, hydroxypropyl cellulose and methyl
cellulose, and anionically charged derivatives
such as carboxymethyl cellulose are used
widely in different applications, for example,
as rheology modifiers and process additives.
Cationic derivatives are much less known,
and their commercial utilization is currently
negligible.
The cationic polymer market is worth close to $4
billion, with global production (excluding starch in
papermaking) at around 1 million tonnes a year.
It has been previously proven that cellulose can
be made water soluble by introducing either
non-ionic or anionic polar groups. If cationic
groups could also be introduced to the cellulose
molecule, water solubility would be achieved
and its use as a flocculating polymer would
become possible.
This has been demonstrated in the TEKES
project, Novel Cellulose Chemicals in
Papermaking – CelPlus, where cationic
water-soluble derivatives were made from
cellulose. The derivatives were also tested in
papermaking applications such as retention,
dewatering and pitch control. Optimization of
the manufacturing process was not included
in the scope of CelPlus. Thus, identification of
the optimum manufacturing method and its
optimization were the main goals of the current
research.
Although the main goal was to obtain a water-
soluble cationic cellulose derivative, this
approach was also questioned. An alternative
approach involving the production of a cationic
derivative of nano-scale cellulosic particles was
investigated to determine whether an efficient
cationic flocculant needs to be a water-soluble
polymer. Both approaches were tested in
different applications at laboratory scale.
The economic feasibility of the selected
processes was also evaluated in order
to compare how successful the cellulose
derivatives would be commercially compared
to synthetic cationic polymers.

2. Objectives
A new process for producing a cationized,
water-soluble, cellulose-based polyelectrolyte
chemical product was to be developed, verified
at initial pilot scale and modelled economically.
The polymeric product was to be tested and
benchmarked in selected applications as a paper
and/or water processing chemical.
Cationized cellulose nano-scale particles were
also to be developed in order to compare them
to water-soluble derivatives.
The research was divided into five Tasks:
• Activation and molecular weight control
• Evaluation of routes for water-soluble
cationic cellulose
• Development of routes for water-soluble
cationic cellulose
• Cationic particles from reactive milling
• Application testing
In ‘Activation and molecular weight control’
the aim was to improve the reactivity of wood
pulp cellulose. In ‘Evaluation of routes for
water-soluble cationic cellulose’ the aim was
to evaluate different routes for synthesizing
cationic cellulose derivatives for flocculants.
In ‘Development of routes for water-soluble
cationic cellulose’ the objective was to identify
the two most potential synthesis routes and
optimize them with the aim of upscaling
one of the routes to the pilot scale. ‘Cationic
particles from reactive milling’ was devoted to
cationic nano-scale particle synthesis. Most
of the product characterization work of the
samples and application testing were done in
‘Application testing’.
4. Results
4.1 Used pulps
Different native and pretreated cellulosic pulps
were used as starting material during the
research. The aim was to find a pulp for chemical
synthesis having high reactivity and resulting
in a cationic polymeric end product of as high
as possible molecular weight. In the cationic
nano-scale particles the molecular weight of the
cellulose did not play as important a role.
The used pulps and some of their properties are
listed in Table 1.
4.2 Activation
The aim of the activation research was to
investigate means of improving the chemical
reactivity of cellulose and to evaluate how
the improved activation leading to reactivity
improvement could best be characterized. For
the latter, the most straightforward method was
considered to be a chemical derivatization itself.
As a model derivatization reaction, a
simple carboxymethylation using sodium
monochloroacetate, CMC synthesis, was
selected. Using this method it was concluded
that there was no significant difference
in reactivity whether the cellulose was
disintegrated in dry or wet form, and that
isopropanol was the best performing additive
in the reaction media.
Different energy sources during the pulp
activation and CMC reaction stages were also
studied. Use of microwaves gave a better reaction
result than conventional heating, whereas
reaction with ultrawaves performed the worst.
Three different heating techniques, conventional
heating, microwaves and ultrasound, were
thereafter used for CMC reaction with Borregaard
and Domsjö dissolving pulp. There was no clear

Cellulose starting material Abbreviation Mw (kDa)
Bahia dissolving pulp Bahia 360
Blue Bear Ultra Ether, Borregaard dissolving pulp Borregaard 1700
Domsjö dissolving pulp after mechanical treatment Domsjö 410
Domsjö dissolving pulp after acetylation DAc 550
Domsjö dissolving pulp after standard Biocelsol mechanical and enzyma-
tic pre-treatment (see Chapter “Water based dissolution and regeneration
processes”)
DENz nd
Domsjö dissolving pulp after novel mechanical and enzymatic treatment
(advanced Biocelsol pre-treatment, see Chapter “Water based dissolution
and regeneration processes”)
Dext 160
Enoalfa, Enocell dissolving pulp, Stora Enso Enoalfa nd
Bleached birch kraft pulp, Kaskinen Birch nd
Never-dried hemi-poor hardwood kraft pulp (birch) – freeze dried before use HWNDHP nd
Never-dried softwood kraft pulp (hemicelluloses not removed) – freeze
dried before use
SWND nd
Never-dried hemi-poor softwood kraft pulp – freeze dried before use
when used in DMAc/LiCl or HC DIT reaction systems
SWNDHP 880
Acetylated softwood never-dried hemi-poor kraft pulp Ac-SWNDHP nd
Butylated softwood never-dried hemi-poor kraft pulp B-SWNDHP nd
Thermomechanical pulp TMP nd
Microcrystalline cellulose MCC nd
Methylated cellulose MeC nd
Micro- and nanofibrillated cellulose MFC nd
Sigma α-cellulose 670
Table 1. Used cellulose and cellulose pulps. nd=not determined.
difference between the reactivity of the two
dissolving pulps despite their different origin and
cellulose degree of polymerization (DP) when the
same energy source was used.
Different activation parameters were further
studied using the actual cationization, GTAC
synthesis (see Figure 1). Of the energy sources
studied, microwaves gave again the best
reaction results. Disintegration, especially wet
disintegration, had a higher effect on GTAC
than CMC reaction. It was also shown that
CMC and GTAC synthesis have different NaOH
concentration optimums. No better additive
than isopropanol was found.
Chemical pre-treatment methods, such as
acetylation, hydroxypropylation or methylation,
did not give satisfactory results in increasing the
reactivity of cellulose in cationization. Moreover,
methylation seemed to disturb the cationization
reaction, probably due to competing for the
same active hydroxyl sites. It was considered
that in certain cases hydroxypropylation could
help improve solubility if it is done after the
cationization step.
Freezing of alkaline water treated cellulose to
-40 °C with and without ZnO as an additive was
also tested as a pre-treatment method, but
was not found to improve cationization.

Figure 1. The first six selected preparation methods for cationized cellulose derivatives: (1) Williamson
etherification, (2) glycidyl route (GTAC), (3) Mannich routes, (4) C6 activation, (5) Michael route, and (6)
grafting methods.
Reaction route Utilized for pulps
Cellulose betainate Borregaard, Domsjö, Bahia
Cationization of cellulose acrylate with 3-methylimidazolium propionate chloride Borregaard, Domsjö
Cationization of cellulose acrylate with diethylamine Borregaard
Cationization of cellulose 2-methylpropanoyl bromide with 1-methylimidazole Borregaard
Cationization of low DS nitrocellulose Borregaard
Mannich reactions with cellulose carbamate Domsjö
Cellulose esterification with aromatic, tertiary amine group containing acid
halide and its quaternization with methyl iodide
Table 2. Some homogenous system reaction routes.
4.3 Water-soluble derivatives
4.3.1 Reaction route screening
In the first phase several alternative reaction
routes to cationic, water-soluble cellulose
derivatives were screened. Figure 1 and Table 2
summarize the approaches tested.
The GTAC method was studied in several
reaction systems, both in homogeneous
and heterogeneous systems, and their
reaction efficiencies were compared. The
other cationizations were made only in the
homogeneous systems. Table 3 lists the used
reaction systems.
The reaction routes with highest potential were
GTAC modification in a DMAc/LiCl homogenous
system and in a HC DIT heterogeneous
system, cationization of cellulose acrylate,
and cationization of cellulose betainate. These
reaction routes were further optimized.

Abbreviation Reaction system
Aq Aqueous system with cosolvent (typically 10-50% cosolvent) and 5-8 wt-% of cellulose (het-
erogeneous). With GTAC.
Biocelsol 5.5% NaOH/1.3% ZnO (homogeneous), 5-8 wt-% of cellulose. With GTAC.
DMAc/LiCl System with 5% lithium chloride in dimethylacetamide. Typically 1-5 wt-% of cellulose depending
on the cellulose type was dissolved in DMAc/LiCl yielding a homogeneous solution. With GTAC.
HC DIT High-consistency DIT or other reactor system typically with 20-70 wt-% of cellulose (hetero-
geneous) without any cosolvents. With GTAC.
two-phase E.g. water-toluene (heterogeneous) with 2-15 wt-% of cellulose. With GTAC.
MIPCl Cationization of cellulose acrylate with 3-methylimidazolium propionate chloride
NClB Cellulose betainate (N-chloro-betainate)
Other UH The other routes studied.
Table 3. Used reaction systems for water-soluble derivatives.
4.3.2 Reaction efficiency of the GTAC routes
Reaction efficiencies were calculated from
reacted GTAC amount, and they were evaluated
for different GTAC routes. The results are shown
in Figure 2, where the calculated reaction
efficiencies are plotted against the DS achieved.
The plot forms three straight lines when the
dominating factor is the used GTAC amount.
This is seen in Figure 3, where the parameter
‘Reaction efficiency per DS’ is plotted against
the used GTAC amount. The best reaction
efficiencies were obtained with the HC DIT
route, thus in high consistency systems. When
the best reaction efficiencies from Figure 2 are
plotted against the achieved DS, the relationship
between achievable reaction efficiency and DS
is obtained (Figure 4).
There is a clear correlation between high DS
and good reaction efficiency. It seems, however,
that with the reaction systems used it is difficult
to reach over 50% reaction efficiency. If a large
amount of GTAC reactant is lost, which is the
case when the reaction efficiency is below
50%, the manufacturing costs are high, making
GTAC cationization uneconomical. Thus, means
of achieving a higher GTAC utilization rate need
to be studied further.

Figure 2. Reaction efficiency calculated from reacted GTAC against achieved DS. Classified by reaction
system (a) and cellulose type (b).
a)
b)
0

10

20

30

40

50

60

0,0
0,5
1,0
1,5
2,0

Reac%on
eﬃciency
(RE),
%

DS

Aq

Biocelsol

DMAc/LiCl

HC
DIT

two-‐phase

0

10

20

30

40

50

60

0,0
0,5
1,0
1,5
2,0

Reac%on
eﬃciency
(RE),
%

DS

Borregaard

Domsjö

DAc

DENz

Dext

SWNDHP

HWNDHP

SWND

Enoalfa

Ac-‐SWNDHP

MeC

B-‐SWNDHP

MCC

Figure 4. The best achieved reaction efficiencies against the DS values achieved (based on maximum
points from Figure 2).
35

40

45

50

55

60

0,9
1
1,1
1,2
1,3
1,4
1,5
1,6
1,7
1,8
1,9

Max
reac(on
eﬃciency
(RE)
based
on
GTAC,
%

DS
achieved

Figure 3. Parameter ‘Reaction efficiency per DS’ against used GTAC amount.
0

20

40

60

80

100

120

140

0,0
1,0
2,0
3,0
4,0
5,0
6,0

Reac%on
eﬃciency
RE
per
DS
achieved

GTAC
used,
mol/AGU

Aq

Biocelsol

DMAc/LiCl

HC
DIT

two-‐phase

Figure 5. Charge density at pH 4 against DS. Classified by reaction system (a) and by cellulose type (b).
4.3.3 Sample characterization
For flocculating polymers, the key parameters are:
• Solubility
• Molecular weight (chain length)
• Charge density
Thesolubilityofthecationiccellulosederivatives
was characterized by measuring the turbidity
of a 1% polymer solution. The effect of polymer
chain length was characterized by measuring
the viscosity of a 2% polymer solution.
Because the polyelectrolyte charge affects
the solution viscosity, salt viscosity was used
as the main parameter to describe the effect
of molecular weight. The charge density of the
polymer was characterized by polyelectrolyte
titration at pH's 4 and 7.5. The charge density
measured at lower pH was mostly used in the
evaluations because the ester derivatives did
not give reliable results due to decomposition
by hydrolysis at higher pH.
In the results, samples are classified based on
both the reaction route used and the cellulose
starting material.
Figure 5 shows the relationship between
measured DS and measured charge density
at pH 4. The charge density of most of the
derivatives is lower than expected based on the
DS values. In some cases the charge density
is, however, higher than expected, especially
with some of the DMAc/LiCl samples. The DS
of the samples was measured partly based on
sample nitrogen content and partly by using
a)
b)
0,0

0,5

1,0

1,5

2,0

2,5

3,0

3,5

4,0

4,5

5,0

0,0
0,5
1,0
1,5
2,0

Charge
density
at
pH
4,
meq/g

DS

Aq

Biocelsol

DMAc/LiCl

HC
DIT

two-‐phase

MIPCl

NClB

Other
UH

theoreEcal

0,0

0,5

1,0

1,5

2,0

2,5

3,0

3,5

4,0

4,5

5,0

0,0
0,5
1,0
1,5
2,0

Charge
density
at
pH
4,
meq/g

DS

Borregaard

Domsjö

DAc

DENz

Dext

SWNDHP

HWNDHP

SWND

Enoalfa

Ac-‐SWNDHP

MeC

Bahia

theoreHcal

NMR. Further investigation is clearly required
in order to clarify the relationship between DS
and charge density.
The solubility behaviour of the samples is
presented in Figure 6. Polymer solubility
improves and thus solution turbidity decreases
with increasing charge density. There is no
exact definition of good solubility, but, for
example, when targeting solutions with
turbidity lower than 100 NTU, a charge density
of 2 meq/g of polymer or higher is required.
No significant difference between the reaction
systems or cellulose types is found, although
the HC DIT reaction system seems to give
somewhat lower and cellulose betainate
(NClB) and 3-methylimidazolium propionate
chloride (MIPCl) somewhat higher solubility
compared to the majority of the samples. The
highest charge densities were achieved using
the DMAc/LiCl reaction system and Domsjö or
extruded Domsjö (Dext) pulps.
The highest viscosities were achieved using
cellulose betainate, 3-methylimidazolium
propionate chloride (MIPCl), HC DIT or DMAc/
LiCl reaction routes, and Borregaard pulp,
SWNDHP or Domsjö pulp (Figure 7). When the
salt viscosity correlating with polymer chain
length is plotted against its charge density in
the area of proper solubility (charge density
>2 meq/g), the figure reveals a clear trend of
reducing viscosity with higher charge density
(Figure 7). This indicates that a very high
molecular weight product with very high charge
density cannot be obtained.
Figure 6. Turbidity against charge density at pH 4. Classified by reaction system (a) and by cellulose type (b).
a)
b)
1

10

100

1000

10000

0,0
0,5
1,0
1,5
2,0
2,5
3,0
3,5
4,0
4,5
5,0

Turbidity
at
1
%,
NTU

Charge
density
at
pH
4,
meq/g

Aq

DMAc/LiCl

HC
DIT

two-‐phase

MIPCl

NClB

Other
UH

1

10

100

1000

10000

0,0
0,5
1,0
1,5
2,0
2,5
3,0
3,5
4,0
4,5
5,0

Turbidity
at
1
%,
NTU

Charge
density
at
pH
4,
meq/g

Borregaard

Domsjö

DAc

Dext

SWNDHP

HWNDHP

SWND

Enoalfa

Ac-‐SWNDHP

MeC

Bahia

Figure 7. Salt viscosity vs. charge density, when charge density is >2 meq/g. Classified by reaction system
(a) and by cellulose type (b).
4.4 Cationic particles
Two main cationization routes were used to
produce cationic nano-scale particles, see
Table 4.
Cationization was tested both before and after
cellulose comminution. Comminution after
cellulose cationization was considered to be
better. In addition, commercial microfibrillated
and nanofibrillated celluloses were tested in
cationization.
The main problem with the cationic nano-
scale particles was the product suspension
concentration, which remained <0.5% after
the comminution stage in a high intensity
homogenizer.
Concentration of the suspension was also
tested, and promising results were obtained by
concentrating the suspension to 30-100% with
no significant loss in performance efficiency. It
was also found that drying without performance
loss can be achieved if the particles have a
higher charge density.
a)
b)
1

10

100

1000

10000

2,0
2,5
3,0
3,5
4,0
4,5

Salt
viscosity
at
2
%,
mPas

Charge
density
at
pH
4,
meq/g

Aq

Biocelsol

DMAc/LiCl

HC
DIT

two-‐phase

MIPCl

NClB

Other
UH

1

10

100

1000

10000

2,0
2,5
3,0
3,5
4,0
4,5

Salt
viscosity
at
2
%,
mPas

Charge
density
at
pH
4,
meq/g

Borregaard

Domsjö

DAc

DENz

Dext

SWNDHP

HWNDHP

SWND

Enoalfa

Ac-‐SWNDHP

MeC

Bahia

4.5 Performance evaluation
Product performance was evaluated using
laboratory-scale methods. Sludge dewatering
was tested using a CST (capillary suction time)
method where the speed of water drainage
from a sludge into a standardized piece of board
is measured; the shorter the drainage time, the
better the dewatering capacity. Flocculation
efficiency, describing, for example, retention on
the paper machine, was tested using the FBRM
(focused beam reflectance measurement)
method, which detects the particle or floc
size of the suspension dynamically, with shear
forces induced in the suspension controlled by
mixing speed and time of mixing; the bigger floc
size, the better the flocculation efficiency. Pitch
control by fixing was tested by measuring how
much turbidity in a mechanical pulp suspension
water is removed by adding the fixing agent;
the higher the removal percentage, the better
the fixing performance.
The relationship between CST time and charge
density of the sample is shown in Figure 8,
which clearly reveals charge density to be the
dominating factor in CST performance. The
best performing samples are those with the
highest charge density.
Some additional conclusions can also be drawn.
Certain products, such as the cationic nano-
scale particles and derivatives made via the
methylimidazolium propionate chloride (MIPCl)
route, perform better in sludge dewatering
than expected based on their charge density.
On the other hand, the HC DIT made samples
seem to perform worse than expected based on
their charge density. One possible explanation
for this is that because the HC DIT route is a
heterogeneous system, the charge created
during the modification is not evenly distributed.
In addition, pulp type was found to have less
of an impact on performance than expected;
the Dext and Domsjö pulps were among the
best performers with the DMAc/LiCl route, but
well-performing samples were also made using
Borregaard pulp.
None of the cellulose derivatives had as good
performance in CST as the reference polyacryl-
amide, although their performance was not
far from the reference (Figure 8). There are
indications that sludge dewatering performance
ofthederivativesimproveswithhighermolecular
weight, but this requires further confirmation.
Typical FBRM curves are presented in Figure
9. When a flocculant is dosed, the floc size
increases rapidly. When shearing is induced,
i.e. mixing is continued, the floc size starts to
decrease. With some cellulose derivatives it was
found that even though the initial floc size was
not as big as the reference, the decrease in floc
size due to shearing was not as severe as the
reference. Moreover, after a certain shearing
level the floc size of the cellulose derivative was
larger than with the reference (Figure 9).
Abbreviation Reaction system Maximum charge density achieved at
pH 4, meq/g
AminoG Periodate oxidation of cellulose to
dialdehyde cellulose and subsequent
cationization using aminoguanidine
hydrochloride
2.3
Girard’s T Periodate oxidation of cellulose to
dialdehyde cellulose and subsequent
cationization using Girard’s reagent T
1.2
Table 4. Cationic particle reaction systems and achieved charge densities.

Figure 8. CST times vs. charge density at 8 kg/t dosage. Classified by reaction system (a) and by cellulose
type (b). Reference polyacrylamide, Fennopol K506, gave a typical CST time of <10 s at 6-8 kg/t dosage.
The sludge is a municipal digested sludge, pH 7-7.5 (all CST tests). Tests were performed at different times
using different sludges from the same source.
a)
b)
0

20

40

60

80

100

120

140

160

180

200

220

240

260

0,0
0,5
1,0
1,5
2,0
2,5
3,0
3,5
4,0
4,5
5,0

CST
%me
at
8
kg/t,
s

Charge
density
at
pH
4,
meq/g

Aq

Biocelsol

DMAc/LiCl

HC
DIT

two-‐phase

MIPCl

NClB

Other
UH

AminoG

Girard's
T

0

20

40

60

80

100

120

140

160

180

200

220

240

260

0,0
0,5
1,0
1,5
2,0
2,5
3,0
3,5
4,0
4,5
5,0

CST
%me
at
8
kg/t,
s

Charge
density
at
pH
4,
meq/g

Borregaard

Domsjö

DAc

DENz

Dext

SWNDHP

HWNDHP

SWND

Enoalfa

Ac-‐SWNDHP

Bahia

Birch

(Kaskinen)

MFC

Figure 9. Curves measured using the FBRM method. Fennopol K3400R is the polyacrylamide reference.
UH-FBC-WP4-I is a sample made from Borregaard pulp using the cellulose betainate route. Dosing as kg
active / t dry. See other details in Figure 10.
The dependence of flocculating efficiency on
salt viscosity remains an anomaly, see Figure 10.
In some cases high salt viscosity of the product
clearly gave better flocculation efficiency. This
was the case for samples made using cellulose
betainate (NClB), MIPCl or HC DIT reaction
systems and mostly Borregaard, SWNDHP or
Enoalfa pulps. However, the same pulps also
resulted in samples that did not show good
flocculating efficiency despite having high salt
viscosity. The reason for this odd behaviour is
still unclear.
 
The cationic particles also performed
well in the flocculation tests. The better
performing cationic particle, made using the
aminoguanidine route, gave a maximum mean
floc size of 24-27 µm at 12 kg/t dosage.
The cationic particles had an odd effect on floc
strength, see Figure 11. Floc size increase was
minimal, but shear resistance was high, even
increasing with higher shear levels. In high
shearing systems the cationic particle thus
gave as good, or even better, final flocculation
efficiency than the reference polyacrylamide.
The fixing performance of selected samples
is presented in Figure 12. While some samples
performed better than the reference, worse
performance was found in many cases. The
cationic nano-scale particles performed clearly
worse in fixing than many other samples.
Some properties of the samples used in the
fixing tests are presented in Table 5. The best
performing samples in the fixing tests clearly
belong to the group with the highest charge
densities. Interestingly, the fixing performance
of many of the cellulose derivatives with a
clearly lower charge density than the reference
was comparable or better than that of the
polyamine reference having a charge density of
about 7 meq/g.
10

15

20

25

30

35

40

45

50

15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
95
100
105
110
115
120

Mean
ﬂoc
size,
µm

Mixing
1me,
s

Fennopol
K3400R_4kg/t

UH-‐FBC-‐WP4-‐I_12kg/t

UH-‐FBC-‐WP4-‐I_8kg/t

UH-‐FBC-‐WP4-‐I
4
kg/t

Figure 10. Maximum mean floc sizes at dosing 12 kg/t from the FBRM experiments against salt viscosity.
Classified by reaction system (a) and by cellulose type (b). Tests were made at different times using the
same type of mechanical pulp from two different sources. Furnish suspension: 60% groundwood, 40%
PCC, pH 7.5, mixing speed 1500 rpm (all FBRM tests).
a) 10

15

20

25

30

35

1
10
100
1000
10000

Max
mean
ﬂoc
size
at
12
kg/t,
µm

Salt
viscosity
at
2
%,
mPas

Aq

Biocelsol

DMAc/LiCl

HC
DIT

two-‐phase

MIPCl

NClB

Other
UH

b)
10

15

20

25

30

35

1
10
100
1000
10000

Max
mean
ﬂoc
size
at
12
kg/t,
µm

Salt
viscosity
at
2
%,
mPas

Borregaard

Domsjö

DAc

DENz

Dext

SWNDHP

HWNDHP

SWND

Enoalfa

Ac-‐SWNDHP

Bahia

Figure 12. Fixing performance of selected samples. Groundwood pulp, pH 6.9. Dosing as g active / t dry
pulp. Polyamine is the reference polymer.
20

30

40

50

60

70

80

90

100

600
800
1000
1200
1400
1600
1800
2000

Turbidity
removal
(%)

Dosing,
g/t

VTT-‐312

VTT-‐311

UH-‐FBC-‐
WP4-‐VII

UH-‐1-‐1

Polyamine

VTT-‐313

VTT-‐320

UH-‐FBC-‐
WP4-‐VI

AGDAC11

Groundwood
pulp

Figure 11. Curves measured using the FBRM method. Fennopol K3400R is the polyacrylamide reference.
AGDAC11 is a cationic nano-scale particle sample.
10

15

20

25

30

35

40

45

50

15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
95
100
105
110
115
120

Mean
ﬂoc
size,
µm

Mixing
1me,
s

Fennopol
K3400R
4
kg/t

AGDAC11_4
kg/t

AGDAC11_17
kg/t

4.6 Summary of the results
• The best pulp activation method found was
wet disintegration. Chemical pre-treatment
or freezing did not increase pulp reactivity.
Use of microwaves during the activation
and reaction stages gave better results than
conventional heating.
• The best reaction routes identified for
cationization were GTAC synthesis, the
cellulose betainate route, and cationic
particles. All three routes have their benefits
and limitations, but none of them alone
result in sufficient cationization.
• High-consistency reaction systems were the
best systems identified.
• Good solubility of a cationic cellulose
derivative requires a charge density of about
2 meq/g (DS about 0.5).
• When the charge density of the cationic
end product is higher, the molecular weight
tends to be lower.
• Different applications require different
polymer properties.
• Sludge dewatering performance (CST):
- Prefers high charge density, except with
cationic particles
- Best reaction systems: MIPCl and other
Michael routes, NClB and cationic particles
- Best starting celluloses: no significant
differences, only SWNDHP gave poorer
results
• Flocculation performance (FBRM):
- Prefers high molecular weight, except
with cationic particles where cellulose DP is
irrelevant
- Best reaction systems: NClB, HC DIT, MIPCl
and other Michael routes, cationic particles
- Best starting celluloses: Borregaard,
SWNDHP and Enoalfa
- Cellulose derivatives did not give as big
floc size as the reference polyacrylamide,
but flocs were more shear resistant
• In pitch control by fixing, some derivatives
performed better than the reference.
• Normal kraft pulp can be used,
hemicelluloses have no significant effect on
performance.
Table 5. Some properties of the samples mentioned in the figures above. Nd=not determined.
Code Cellulose Reaction
system
Viscosity 2%,
mPas
Salt viscosity
2%, mPas
Turbidity 1%
(NTU)
Charge pH 4
meq/g
AGDAC11 Birch
(Kaskinen)
Amino-
guanidine
nd nd nd 2.3
UH-1-1 Domsjö NClB 394 nd 4 1.9
UH-FBC-WP4-I Borregaard NClB 55000 4515 18 2.6
UH-FBC-WP4-VI Domsjö MIPCl 21 11 10 3.5
UH-FBC-WP4-VII Bahia NClB 48 31 4 3.6
VTT-311 Dext DMAc/LiCl,
GTAC
25 16 11 4.4
GTAC
22 14 15 4.3
GTAC
85 38 31 3.1
VTT-320 Domsjö DMAc/LiCl,
GTAC
66 34 22 4.5

4.7 Markets and business opportunities
for cationic water-soluble cellulose
About 20% of water-soluble polymers bear a
cationic charge, while the remaining 80% consist
of anionic and neutral polymers. Key properties
of cationic polymers are molecular weight
and charge density, which vary significantly
depending on the end-use application. The
cationic cellulose developed in the FuBio
Cellulose programme has a molecular weight
and charge density suitable for coagulation
and flocculation applications. The key end-use
sectors for cationic cellulose therefore include
the water treatment, pulp and paper, oil, mining,
cosmetics and textile industries.
The most widely used cationic polymer is
polyacrylamide (PAM), followed by quaternary
ammonium polymers and polyamines. Less
than 10% of cationic polymers are currently
based on natural materials such as chitosan.
Large-scale manufacturers of bio-based
cationic chemicals are currently few in number,
but there is significant research interest in this
area. Because the FuBio Cellulose products
were developed specifically as coagulants
and flocculants, other potential water-soluble
polymers, such as cationic starch, were
excluded from the market analysis.
Water treatment and pulp and paper are the
largest end-use markets, which together
account for over 80% of annual cationic
polymer consumption. Key applications in these
sectors include coagulation and flocculation in
raw water and wastewater treatment, sludge
dewatering, and retention aid in pulp and
paper processes. Different applications require
different polymer properties. For example,
sludge dewatering requires a high molecular
weight with linear, branched or cross-linked
structures, whereas coagulant polymers have
much lower molecular weight but very high
cationic charge.
Demand for cationic water-soluble cellulose
in water treatment applications is driven by
limited water availability, changes in water
use, increasing quality requirements, types
of pollutants, trade-offs between various
chemical compounds, government policies,
and the emerging bio-based economy. Water
scarcity is the key driver behind all water-
related businesses and has driven both public
and private sectors to focus on water recycling,
reutilization and minimization of discharge
water – all of which increase the demand for
water treatment chemicals.
Geographical location, seasons, and water
end-use have a major impact on the type and
amount of coagulants and flocculants required.
Demand for cationic polymers, in particular,
is growing alongside increasing energy and
resource efficiency targets. Municipal and
many industrial wastewaters bearing impurities
with high anionic charge can be effectively
neutralized with cationic flocculants. The
required molecular weight and charge density
of the applied cationic polymer depends on the
types of pollutants contained in the wastewater.
Selecting the optimum coagulant and flocculant
combination includes trade-offs among various
chemicals. For instance, the ratio of metal salts
to cationic polymers, or the ratio between
different cationics may be altered based on
chemical price changes to optimize overall
cost efficiency. In most applications, the right
coagulant and flocculant combination needs to
be confirmed by on-site sedimentation tests.
Government policies also have a strong
influence on the demand development of
cationic polymers. Dosage volumes depend
on the required purity levels (e.g. COD, BOD5
and phosphorus), which vary in different
administrative regions. There are planned
legislative restrictions for cationic polymers in
Spain, Germany and Sweden driven by monomer
residues, biodegradability and sustainability,

Figure 13. Potential of cationic cellulose in cosmetics and pulp and paper applications.
Market size
Growth potential of
the end‐use segment
Growth potential
Unit value
Capability and
willingness
to pay
Regulative
restrictions
Legislative
environment of the
end‐use segment
Threat of new
technologies
Threat of subsitution
chemicals
Technical substitution
potential
Market size
Growth potential of
the end‐use segment
Growth potential
Unit value
Capability and
willingness
to pay
Regulative
restrictions
Legislative
environment of the
end‐use segment
Threat of new
technologies
Threat of subsitution
chemicals
Technical substitution
potential
Cosmetics Pulp & Paper
respectively. These restrictions would limit or
even ban the use of the most common cationic
polymer, c-PAM, and thus, could open new
opportunities for cationic cellulose. In addition,
any actions supporting the creation of a bio-
based economy will support the adoption of
alternative bio-based materials.
In order to identify the potential of cationic
cellulose in different end-use sectors, a variety
of factors were analysed, including market
size, growth of the end-use segment, growth
of cationic chemical use in that segment, unit
value, and capability and willingness to pay
for bio-based or biodegradable products.
However, the most crucial factors affecting
the potential for cationic cellulose were the
technical substitution potential, unit value and
regulatory environment.
According to the analysis, cosmetic
applications seemed the most promising end-
use for cellulose-based cationic chemicals.
Although the market size of cationic chemicals
in cosmetics is small, both the cosmetic
industry and the cationic chemicals used in
hair care products (mainly polyquaternium) are
estimated to grow at a respectable rate of over
3% per annum. There is a clear demand for
bio-based raw materials in cosmetic products
and the industry has both the capability and
willingness to pay a premium for specific
products. However, the product cycle in
cosmetics is short and polymer quantities small.
Cosmetic applications do not require very high
molecular weight or high cationic charge, and
therefore the technical substitution potential of
cationic cellulose is very high.
The pulp and paper industry was also an
interesting end-use for cationic cellulose. Pulp
and paper production is the second largest
market for cationic chemicals in general. Both
the end-use market and the use of cationic
chemicals are slightly growing, although
mostly in emerging markets. The key factors
determining the potential of cationic cellulose
in cosmetics and pulp and paper applications
are summarized in Figure 13.
4.8 Techno-economic modelling of
cationic cellulose
The techno-economic analysis examined the
production of cationic water-soluble cellulose
production for water purification applications.
The process concept of carboxymethylcellulose
(CMC) was used as a general reference for the
cationic cellulose process concept. Five different
process variations were evaluated: (i) aqueous
media with GTAC as reagent, (ii) organic media
with GTAC as reagent, (iii) reactive dissolving
with chloro-betainyl chloride as reagent, (iv)
DMAc-LiCl as organic media and (v) a high-
consistency process. Figure 14 shows a block-
flow diagram of the studied process concept.

Based on the techno-economic analyses,
the high-consistency process seems to be
the most promising of the five concepts. The
cationic derivatization agent GTAC was the
biggest production cost factor and thus had
a major impact on the economic feasibility.
As a result, further research should focus
on reducing GTAC consumption either by
increasing reaction efficiency or by improving
chemical recovery and recycling.
Cationic water-soluble polymers are performance
chemicals, and their performance thus defines
their potential selling price. Product functionality
and application testing should be a top priority
of future research and product development.
Monomer residues, biodegradability and
sustainability are the key driving forces
behind planned legislative restrictions on
cationic polymers in Spain, Germany and
Sweden, respectively. Product development
should also focus on exploiting these
unique opportunities for cationic cellulose.
The strengths and weaknesses of cationic
Figure 14. Block-flow diagram of cationic water-
soluble cellulose production.
NaOH
water
DMAc
LiCi
GTAC
HCI
Cellulose
Basification
Cationization
Neutralization
Filtration
Washing
Cationic cellulose
IPA
Drying
Solvent
recovery
water-soluble cellulose in water treatment
applications are summarized in a SWOT
analysis in Figure 15.
the results
Use of cellulose pulp as a raw material for
the production of cationic flocculants was
shown to have good potential for industrial
utilization. The ready availability of cellulose,
as the largest annually renewable biomass on
the planet, further underscores the potential
of this raw material.
Cellulose derivatives are normally produced
using a dissolving pulp. The present study,
however, showed that normal kraft pulp can be
used irrespective of hemicellulose removal; only
a minimal impact on final product performance
was observed when hemicelluloses were not
removed. This is a key finding, as it affects the
raw material price remarkably.
A major impact on end product properties was
found if the starting cellulose raw material
was not heat dried before use. This calls for
integrating cationic derivatization of cellulose
close to pulp manufacturing. Although never-
dried pulp performed better, normal heat-
dried pulp can also be used.
The preferred pulp is softwood. Hardwood pulp
gives a higher molecular weight end product, but
reacts less readily. The hemicellulose content of
hardwood is also higher than that of softwood.
Although the majority of the present findings
require further confirmation, the results at this
stage are very promising.
Several potential reaction types could be
developed, although none of them can be
utilized directly.

GTAC synthesis lacks reaction efficiency,
which increases reactant consumption and
thus manufacturing costs. The best reaction
efficiency achieved was 55%, and any
significant further increase on this is considered
unlikely. One means of making the GTAC route
more viable is to find a way to regenerate and
circulate the extra reactant. This offers a very
interesting avenue for further research.
The GTAC synthesis results show how important
the processing consistency or concentration is
during the reaction. The more solvents or other
media that are needed, the higher the recycling
costs. Low processing concentration is the
weakest link in the cellulose betainate route
developed. The end product properties were
good, but the processing cost became high due
to too low concentration during processing.
The process economics would be dramatically
improved if the processing could be done in
a high concentration system, such as in an
extruder, kneader or such. This requires further
study, with a focus on high viscous processing.
Uncertainties also remained regarding the
cellulose betainate product, namely the
stability of the dry product and, because they
are esters, the need for special attention in the
application systems. Thus, further development
of cellulose betainate is also required.
The third promising technique found is the use
of nano-sized cationic particles. The particles
performed comparably to soluble polymers
and, in some cases, even better. However, the
studied route has two weak aspects. One is
cationization through the aldehyde oxidation
route, which presents a challenge regarding
chemical recycling. Another weak point is the
product concentration. After the comminution
stage the product concentration is below 0.5%.
Good progress was made in the concentrating
studies, but the drying method used, freeze
drying, is technically undesirable. While simple
thermal drying is not effective, methods such
as fluid bed drying or spray drying deserve
further study. Due to the large amounts of
water removed, the drying technique used
must be combined with mechanical dewatering
in order to become economically feasible.
Figure 15. SWOT analysis of cationic water-soluble cellulose.
Helpful to achieving business success Helpful to achieving business success
Process
related
STRENGTHS
• Growing demand for both bio-based water
treatment chemicals and for cationic
polymers as a whole
• Legislation may support development of
bio-based cationic chemicals
• Economic feasibility seems attainable
WEAKNESSES
• Reagent represents too high share of total
production costs.
• DIT reactor can operate at high consistency
but processing high viscosity material
streams may be challenging.
• Expensive and harmful reagent needed
with a risk of harmful residues
Business
environment
related
OPPORTUNITIES
• Bio-based replacement for c-PAM
• New end-use for dissolved cellulose
• New business opportunities for FIBIC
• CMC production is already existing, same
analogy could be used here
• Potential biodegradability
THREATS
• Product quality cannot reach c-PAM
• Full sustainability assessment (cradle-
to-grave) may not show significant
improvements to c-PAM
• Bio-based monomer development

Thus although none of the processes is ready
as such, there are several options for further
development, either each route separately or by
combining the best parts from each one. Other
interesting reaction routes, which remained
outside the scope of the present study, should
also be examined. In addition, the Michael type
addition reactions, to which the MIPCl route
also belongs, gave very interesting and well-
performing samples, although more work is
required to find a substitute for trifluoroacetic
acid used in it.
Although the derivatives did not generally
performaswellasthereferencepolyacrylamide,
their performance matched the reference
when combined with polyacrylamides. In some
applications, such as high shear condition
flocculation and pitch control by fixing, some of
them performed even better than the reference
products.
Flocculating applications that do not require very
high charge densities may be more attractive in
the first instance, as the reaction efficiency with
these was the highest. However, this depends
heavily on the chosen reaction route.
Efficient raw material utilization and low-cost
and low-toxic reactants are the key issues in
the successful cationization of cellulose. The
ultimatesolutionforproducingcationiccellulose
may be based on one of these processes, or
be a combination of several of them. The final
success of cellulose cationization will depend
not only on process efficiency, but also on how
raw material prices develop compared to the
raw material prices of synthetic polymers.
6. Networking
The research was carried out jointly and
exclusively by the programme partners, see
Table 7.
Partner Role
Glocell Qvantitative economic modelling.
Kemira Steering of overall work. Sample characterization and application
testing. Defining, steering and providing competence for the modelling.
Providing industrial insight to techno-economic assessments.
Metsä Fibre Industrial tutor. Providing industrial insight to techno-economic
assessments.
Pöyry Management Consulting Market study. Economic feasibility modelling. Business potential
evaluation.
Stora Enso Industrial tutor. Providing industrial insight to techno-economic
assessments.
• Organic Chemistry
Synthesis development of water-soluble polymers. New routes.
University of Oulu
• Fibre and Particle Engineering
Synthesis development of cationic particles.
UPM-Kymmene Industrial tutor. Providing industrial insight to techno-economic
assessments.
VTT Synthesis development of water-soluble polymers. GTAC routes.
Techno-economic modelling.

Publications
Liimatainen, H, Suopajärvi, T, Sirviö, JA,
Hormi, O. and Niinimäki, J. Fabrication of
Cationic Cellulosic Nanofibrils through Aqueous
Quaternization Pretreatment and Their Use in
Colloid Aggregation. Carbohydrate Polymers
103, 2014:187-192.
Presentations
Ahlgren, J., Jääskeläinen, H., Kurkinen, S.,
Rouhiainen, J., Salmenkivi, K., and Hult Mori,
E-L. New products: The market potential for
cationic cellulose chemicals. FuBio Cellulose
Seminar, Espoo, June 12, 2012.
Ahlgren, J. Cationic chemicals. FuBio Seminar,
Espoo, October 1, 2012.
Posters
Karisalmi, K. and Kyllönen, L. Activation
studies in cellulose derivatization. FuBio
seminar, Espoo, August 27, 2013.
Kavakka, J., Sievänen, K., Labaf, S.,
Lagerblom, L., Kilpeläinen, I., Karisalmi, K.
and Ahlgren, J. Towards Cationic Cellulose:
Reactive Dissolution Approach. FuBio Seminar,
Espoo, June 12, 2012.
Kavakka, J., Sievänen, K., Labaf, S., Lagerblom,
L., and Kilpeläinen, I. Towards Cationic
Cellulose: Reaction Dissolution Approach.
FuBio Seminar, Espoo, October 1, 2012.
Liimatainen H, Sirviö J, Niinimäki J and Hormi
O. Cationic cellulose particles as flocculation
agents. FuBio Programme Seminar, Espoo, June
12th, 2012.
Sievänen, K., Kavakka, J., Fiskari, J., Vainio, P.,
Karisalmi, K. and Kilpeläinen, I. 2013. Synthesis
of Cationic Cellulose derivative for Wastewater
Treatment. FuBio Programme seminar, Espoo,
August 27, 2013.
Vuoti, S., Setälä, H. and Karisalmi, K.
2013. Cellulose cationization in water. FuBio
Programme seminar, Espoo, October 22th, 2013.
ABBREVIATIONS
Pulp types: see Table 1
Reaction routes: see Table 3
• CMC = carboxymethyl cellulose
• CST = capillary suction time; a sludge dewatering testing method
• DP = degree of polymerization; corresponds to molecular weight
• DS = degree of substitution; corresponds to charge density
• FBRM = focused beam reflectance measurement; a dynamic floc size measuring method
• GTAC = glycidyltrimethylammonium chloride; a cationization reagent
• NMR = nuclear magnetic resonance spectroscopy
• PCC = precipitated calcium carbonate

The FuBio Cellulose programme focuses on promoting selected novel
value chains starting from wood derived cellulose. The specific target of
the programme is to develop novel sustainable processes for production
of staple fibres, new cellulose based materials and water treatment
chemicals. The programme provides knowledge and capabilities
supporting the new value chains based on wood cellulose products.
www.fibic.fi

FIBIC FuBio Cellulose programme report

More Related Content

What's hot (13)

Viewers also liked (19)

Similar to FIBIC FuBio Cellulose programme report (20)

More from Finnish Bioeconomy Cluster FIBIC Oy (17)

Recently uploaded (20)

FIBIC FuBio Cellulose programme report