THE USE OF MEMBRANE FILTRATION TO
IMPROVE THE PROPERTIES OF EXTRACTED
WOOD COMPONENTS
Mikaela Helander
Doctoral Thesis
Kungliga Tekniska Högskolan, Stockholm 2014
AKADEMISK AVHANDLING
Som med tillstånd av Kungliga Tekniska Högskolan i Stockholm framlägges till
offentlig granskning för avläggande av teknisk doktorsexamen fredagen den 25
april 2014, kl. 10.00 i sal F3, Lindstedtsvägen 26, KTH, Stockholm. Avhandlingen
försvaras på engelska. Fakultetsopponent: Professor Arthur J. Ragauskas, Georgia
Institute of Technology.
Supervisor
Professor Mikael E. Lindström
Copyright © 2014 Mikaela Helander
All rights reserved
Paper I © 2013, BioResources
Paper II © 2014, Submitted to Journal of Applied Polymer Science
Paper III © 2014, Manuscript
Paper IV © 2012, Current Organic Chemistry
Paper V © 2012, BioResources
TRITA-CHE Report 2014:9
ISSN 1654-1081
ISBN 978-91-7595-046-4
To all the kind people along the way.
“Kindness is more important than wisdom, and the recognition of this is the beginning of wisdom.”
Theodore Isaac Rubin
ABSTRACT
The forest is a large and important natural resource in Sweden, and
approximately 70% of the country’s land area is woodland. Wood is an excellent
raw material for the replacement of oil-based products because it is renewable,
biodegradable and carbon neutral. Furthermore, the forest industry is searching
for new processes and methods to utilise by-product streams in a so-called
integrated biorefinery. A key to the success of producing new products from
wood could be pure and homogenous raw materials. Because wood contains a
large variety of components with different characteristics and sizes, cross-flow
filtration (CFF) will be a key separation technique to obtain homogenous and
pure raw materials in the biorefinery concept.
Different wood material components have been studied in this thesis. The first
part of this work focuses on kraft lignin. Kraft lignin is interesting because
approximately 3.5-4 million metric tonnes are produced in Sweden annually (~7
million tonnes of kraft pulp/year in 2012), and today it is mainly used as fuel. The
second part of this thesis deals with materials in hot water extract (i.e.,
galactoglucomannan, but also other components). These extracted materials are
interesting because similar materials are extracted in thermomechanical pulping
process by-product streams. Finally, nanocellulose has been studied since it is an
interesting raw material for future applications.
Through CFF, kraft lignin from black liquor has been fractionated into raw
material samples with more homogeneous characteristics such as molecular
weight, aromatic hydroxyl groups and thermal properties. From dynamic
mechanical analysis, low molecular weight fractions were found to have the
highest degree softening. To precipitate low molecular weight fractions (<1000 Da
cut-off) into a convenient solid, lower temperatures than for high molecular
weight fractions were needed. To produce low molecular weight lignin (<5000 Da
cut-off) from re-dissolved LignoBoost lignin, lower lignin concentrations and
higher pH and ionic strength were found to increase the permeate flux at the
tested conditions.
Nanocellulose has been produced by a novel process called nanopulping and
has subsequently been size fractionated by CFF to obtain more homogenous
nanocellulose.
CFF and adsorption chromatography can be used to isolate dissolved wood
components, yielding several upgraded products: lignin, lignin-carbohydrate
complexes, and galactoglucomannan.
SAMMANFATTNING
Skogen är en viktig naturtillgång för Sverige och omkring 70% av Sveriges
landarea består av skog. Då ved är ett förnybart, biologiskt nedbrytbart och
koldioxidneutralt material skulle det kunna vara ett utmärkt råmaterial för att
ersätta oljebaserade produkter. Samtidigt söker skogsindustrin nya processer och
metoder för att utnyttja biprodukter från ved i så kallade integrerade
bioraffinaderier. En nyckel till framgång, för att ta fram nya förädlade
slutprodukter av ved, kan vara användandet av rena och homogena
vedkomponenter. Eftersom ved innehåller en stor variation av komponenter med
olika egenskaper och storlekar kan tvärströmsfiltrering (CFF) därmed vara en
viktig separationsteknik för att erhålla homogena och rena råmaterial i
bioraffinaderi-konceptet.
I denna avhandling har olika materialkomponenter från ved studerats. Den
första delen av arbetet behandlar sulfatlignin. Sulfatlignin är intressant då det
produceras cirka 3.5-4 miljoner ton sulfatlignin årligen i Sverige (baserat på ~ 7
miljoner ton sulfatmassa/år 2012), och i dag används sulfatligninet främst som
bränsle. Den andra delen av denna avhandling handlar om vedmaterial från
varmvatten-extrakt (dvs. främst hemicellulosan galaktoglukomannan men även
andra komponenter). Dessa extraherade vedkomponenter är intressanta då
liknande material extraheras som biprodukter i den termomekaniska
massatillverkningen. Slutligen har nanocellulosa studerats då råmaterialet har
intressanta egenskaper för framtida material.
I det praktiska arbetet har CFF använts för att fraktionera sulfatlignin från
svartlut till homogenare råmaterial med lägre molekylviktsfördelning, vilket
resulterade i enhetligare mängder av aromatiska hydroxylgrupper samt
homogenare termiska egenskaper. Dessutom uppvisade lågmolekylära
ligninfraktioner de lägsta glastransitionstemperaturerna, som lägst 70°C. För att
fälla lågmolekylära ligninfraktioner (<1000 Da cut-off) till en hanterbar ligninsuspension, krävdes lägre temperaturer än för de högmolekylära fraktionerna.
Vid produktion av lågmolekylärt sulfatlignin (<5000 Da cut-off) från återupplöst
LignoBoost lignin, gynnades permeatfluxet av lägre ligninkoncentrationer, högre
pH och jonstyrka vid de testade förhållandena.
Nanocellulosa har tagits fram av en ny process som kallas nanopulping, och
därefter storleksfraktioneras med CFF för att erhålla homogenare nanocellulosa.
CFF och adsorptionskromatografi har använts för att isolera upplösta
träkomponenter i flera uppgraderade produkter: lignin, lignin-kolhydratkomplex
och galaktoglukomannan.
LIST OF PAPERS
This thesis is a summary of the following papers:
I.
II.
III.
IV.
V.
"Fractionation of Technical Lignin: Molecular Mass and pH Effects"
Mikaela Helander, Hans Theliander, Martin Lawoko, Gunnar
Henriksson, Liming Zhang and Mikael E. Lindström, BioResources,
2013, 8 (2), 2270-2282.
“Tailoring the molecular and thermo-mechanical properties of kraft
lignin by ultrafiltration” Olena Sevastyanova, Mikaela Helander, Sudip
Chowdhury, Heiko Lange, Helena Wedin, Liming Zhang, Monica Ek,
John F. Kadla, Claudia Crestini and Mikael E. Lindström, Journal of
Applied Polymer Science, Submitted.
“Parameters effecting cross-flow filtration of LignoBoost kraft lignin”
Mikaela Helander, Tuve Mattsson, Hans Theliander, Mikael Lindström,
2014, Manuscript.
"A Novel Nano Cellulose Preparation Method and Size Fraction by
Cross Flow Ultra-Filtration" Hongli Zhu, Mikaela Helander, Carl Moser,
Adam Ståhlkranz, Daniel Söderberg, Gunnar Henriksson and Mikael E.
Lindström, Current Organic Chemistry, 2012, 16, 1871-1875.
"Separation of Galactoglucomannans, Lignin, and Lignin-Carbohydrate
Complexes from Hot-Water-Extracted Norway Spruce by Cross-Flow
Filtration and Adsorption Chromatography" Niklas Westerberg,
Hampus Sunner, Mikaela Helander, Gunnar Henriksson, Martin
Lawoko and Anders Rasmuson, BioResources, 2012, 7 (4), 4501-4516.
Author´s contribution to the appended papers:
I.
Principal author. Took part in outlining the experiments and performed
all experimental work at KTH and most of the preparation of the
manuscript.
II.
Second author. Took part in outlining some of the experiments,
performed some of the experiments and some of the preparation of the
manuscript. Thermal analyses were performed at the University of British
Colombia, Vancouver, Canada.
III.
Principal author. Took part in outlining the experiments, performed all
experimental work and some of the preparation of the manuscript.
IV.
Principal author/equal contributing author. Took part in outlining the
experiments and performed some of the experiments and some of the
preparation of the manuscript.
V.
Third author. Took part in outlining the filtration part and its analysis of
the experiment and performed the filtration part and its analysis of the
experiments and some of the preparation of the manuscript.
This thesis also contains unpublished results.
Other related materials:
Wallenberg Wood Science Center, Theme 1 (2014/2015): Demonstrator, Manuscript.
Helander, M., Theliander, H., Lawoko, M., Henriksson, G., Zhang L., and Lindström, M.
E. (2012): Lignin for new materials - molar mass and pH effects, 243rd National
Meeting of the American Chemical Society & Exposition (“Spring 2012 National
Meeting and Exposition”), Division “Cellulose and Renewable Materials, Session
“Lignin as Renewable Raw Material”, Mars 25-29, San Diego, California.
Helander, M., Sevastyanova, O., Chowdhury, S., Lange, H., Zhang, L., Wedin, H.,
Crestini, C., Ek, M., Kadla, J. F. and Lindström, M. E. (2013): Evaluation of physicochemical properties and prediction of spinning parameters for high-quality
lignins produced by ultrafiltration of industrial kraft pulping liquor, 217th
International Symposium on Wood, Fiber and Pulping Chemistry, June 12-14,
Vancouver, Canada.
Helander, M. and Lindström, M. (2013): Cross-flow filtration for fractionation of
wood components, Marcus Wallenberg price awards, September 23-24,
Stockholm, Sweden.
Helander, M. (2014): Cross-flow filtration for fractionation of wood components,
Ekmandagarna, January 28-30, Stockholm, Sweden.
Helander, M. (2011): Lignin particles for New materials, SPCI (Svenska Cellulosaoch Pappers Ingenjörsföreningen), May 17-19, Stockholm, Sweden.
WALLENBERG WOOD SCIENCE
CENTER - WWSC
This thesis is part of the research initiated by the Knut and Alice Wallenberg
Foundation through the formation of a Swedish research centre, named the
Wallenberg Wood Science Center, at the Chalmers University of Technology in
Gothenburg and the Royal Institute of Technology in Stockholm. The mission and
vision of the WWSC is to create new materials from trees and to develop the
processes to produce these materials. During the initial five years, the centre was
divided into five themes. This thesis was part of Theme 1 – “From wood chips to
material components”. This theme’s focus is on raw materials, and is the first step
towards the production of new products that are dependent on the development
of efficient and economically feasible novel processes for the separation of wood
components.
TABLE OF CONTENTS
1.
INTRODUCTION ..................................................................................................... 1
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
1.9
1.10
BACKGROUND ....................................................................................................... 1
CROSS-FLOW FILTRATION ..................................................................................... 2
.......................................................................................... 5
EXTRACTION/PRODUCTION OF KRAFT LIGNIN ........................................................ 6
FRACTIONATION OF KRAFT BLACK LIQUOR AND KRAFT LIGNIN ............................ 8
NANOCELLULOSE ................................................................................................ 10
PRODUCTION OF NANOCELLULOSE ...................................................................... 10
ENERGY-EFFICIENT METODS FOR NANOCELLULOSE ............................................ 11
HOT WATER EXTRACT FROM SOFTWOOD ............................................................. 13
FRACTIONATION OF HOT WATER EXTRACT FROM SOFTWOOD.............................. 14
KRAFT BLACK LIQUOR
2.
AIM OF INVESTIGATION ................................................................................... 15
3.
EXPERIMENTAL ................................................................................................... 16
3.1
MATERIALS ......................................................................................................... 16




Black liquor................................................................................................ 16
Kraft lignin ................................................................................................. 16
Kraft pulp ................................................................................................... 16
Softwood .................................................................................................... 16
MEMBRANES AND FRACTIONATION ..................................................................... 17
3.2
4.
RESULTS AND DISCUSSION .............................................................................. 18
4.1
PRODUCTION OF LIGNIN FRACTIONS (PAPERS I-III) .............................................. 18








Precipitation and filtration of weak black liquor ....................................... 19
Precipitation and filtration of <1000 Da cut-off permeate ........................ 19
Composition of precipitated lignin ............................................................ 20
Elemental analysis for precipitated lignin ................................................. 21
Yield for <1000 Da cut-off precipitated lignin .......................................... 23
Molecular weight data of precipitated lignin............................................. 24
Chemical characteristics and thermal properties of precipitated lignin ... 25
Cross-flow filtration of dissolved LignoBoost lignin ................................. 29
FRACTIONATION OF NANOCELLULOSE (PAPER IV) ............................................... 37
Energy demand for nanopulping................................................................ 37
Effect of different pretreatment methods .................................................... 38
Fractionation by cross-flow filtration ........................................................ 39
4.2



4.3
FRACTIONATION OF HOT WATER EXTRACT FROM SOFTWOOD (PAPER V) ............. 40




Yield for membrane fractionated hot water extract ................................... 40
Composition of membrane fractionated hot water extract ......................... 41
Molecular weight distribution for fractionated hot water extract ............. 43
Separation by chromatography of fractionated hot water extract ............. 44
5.
CONCLUDING REMARKS .................................................................................. 46
6.
NOTATIONS ........................................................................................................... 49
7.
ACKNOWLEDGEMENTS .................................................................................... 50
8.
REFERENCES ........................................................................................................ 51
Introduction
1. INTRODUCTION
1.1 BACKGROUND
The forest industry is one of the largest industries in Sweden and contributes a
large portion of Sweden’s export income (Skogsindustrin 2012). Today, however,
the pulp and paper industry face challenges due to competition from fast-growing
trees, digital reading and industrial investments in competing cost-effective
countries. For the Swedish forest industry to remain one of the largest, it is
necessary to be at the forefront in regard to new processes and products from
wood. Therefore, much research has been focused on converting the forest
industry’s by-product streams into new products of high value in a so-called
integrated biorefinery and also on finding new processes for wood components.
For example, in the pulping process, some liquid process streams contain
extracted wood components as by-products, which are not being used to generate
high value application products. Intensive research has been performed on
thermomechanical process waters that mainly contain galactoglucomannan
(Willför et al. 2008) and black liquor (Antonsson 2008, Norberg et al. 2012).
Although promising, they have not yet reached industrial levels for high value
application products. A common feature of these process streams is that they
contain
inhomogeneous
wood
components
(i.e.,
different
components,
characteristics and sizes), and this inhomogeneity may limit opportunities to
create new high value application materials. Separation of these process streams
1
Introduction
into more homogenous and pure fractions could increase the opportunities for raw
materials with different properties to be used in new materials or for chemical
modifications (Stewart 2008).
1.2 CROSS-FLOW FILTRATION
Cross-flow filtration (CFF) is a dynamic filtration technique that can separate
particles and molecules in liquids having different sizes and forms. In the
separation, a semipermeable membrane is used. Depending on the pore size of the
membrane (cut-off) the process is divided into different levels of filtration:
microfiltration, ultrafiltration, nanofiltration, and reverse osmosis (see Figure 1).
Bacteria Protein Polysaccharide Salt
Water
Microfiltration
> 0.1 µm
Ultrafiltration
10-100 nm
Nanofiltration
> 1 nm
Reverse osmosis
< 1 nm
Figure 1. Adapted from Nordin (2008). Different levels of filtration. Pore sizes given in
µm and nm are rough values (Ripperger et al. 2009).
In CFF, the liquid sample (i.e., the feed) flows parallel to the filter media. By
introducing a transmembrane pressure (TMP) across the membrane, the liquid is
forced towards the membrane, and materials smaller than the membrane’s pores
will pass through the membrane as a permeate (see Figure 2). The velocity of the
cross-flow, parallel to the filter media, is relatively high, and the flow is turbulent,
creating dynamic mechanisms that reduce the build-up of particles and molecules
2
Introduction
on the membrane surface. However, there will be an increased concentration of
material closer to the membrane surface due to the retained particles; this
phenomenon is called concentration polarisation.
Tank
Feed
Retentate
Membrane
Pump
Membrane
Retentate
Feed
Permeate
(filtrate)
Permeate
Figure 2. Illustration of a filtration unit (left picture) and a cross-section of a tubular
membrane separation (right picture).
The TMP is the driving force for separation (see Equation 1), where Pf is the feed
pressure, Pr is the retentate pressure and Pp is the permeate pressure.
Equation 1:
(
)
The flow velocity though the membrane is defined by Equation 2, where v is the
flow velocity in m/s, Q is the feed flow in m3/s, and A is the membrane area in m2.
Equation 2:
The filtration volume reduction and volume reduction factor can be calculated
using Equation 3 and Equation 4, respectively. Where VR is the volume reduction,
3
Introduction
VRF is the volume reduction factor, Vp is the volume of the permeate, Vif is the
initial feed volume, and Vr is the volume of the retentate.
Equation 3:
Equation 4:
The rejection coefficient is a measure of how efficient the separation between the
material in the feed is versus the material in the permeate (see Equation 5), where
R is the rejection coefficient, cp is the concentration in the permeate, and cf is the
concentration in the feed.
Equation 5:
The permate flux can be calculated according to Equation 6, here J is the flux, Vp
is the volume of the permeate, h is the filtration time in hour, and m2 is the area of
the membrane.
Equation 6:
4
Introduction
1.3 KRAFT BLACK LIQUOR
In the kraft process (also known as kraft pulping or sulphate process), wood chips
are digested in an alkaline cooking liquor called white liquor, which contains the
cooking chemicals sulphide ions and hydroxide ions. The chemicals in white
liquor dissolve wood components at pH > 13 and temperatures between 150 and
170°C (in the so called primary separation). From the digester, a mixture of
liberated cellulose fibres and a liquor containing the dissolved wood components
and spent cooing chemicals (weak black liquor) are obtained. In weak black
liquor (i.e., before the evaporators) the dry content is approximately 15-20%
(Theliander 2009). The main component degraded and dissolved into black liquor
is lignin, which represents approximately 30-45% of the dry material. Other
materials in black liquor are degraded carbohydrate acids (25-35% hydroxyl
acids, 5% acetic acid and 3% formic acid), 3-5% extractives (fatty acids and
resins), hemicellulose, 3-5% sulphur, and 17-20% sodium. The inorganics in
black liquor are sodium hydroxide (NaOH), sodium sulphide (Na2S), sodium
carbonate (Na2CO3), sodium sulphate (Na2SO4), sodium thiosulphate (Na2S2O3),
and sodium chloride (NaCl) (Grace 1989). Other elements are present in small
amounts (< 0.1%), for example, magnesium (Mg), manganese (Mn), aluminium
(Al), silicon (Si), iron (Fe), and phosphorus (P) (Theliander 2009). Lignin is a
macromolecule that is fragmented and extracted from wood chips during the
pulping process. Hence, the lignin found in black liquor has a broad molecularweight distribution, as for example reported by Mörck et al. (1986). The chemical
composition in the fractions is also inhomogeneous; it has previously been shown
that high molecular weight lignin contains increased amounts of aliphatic groups,
while low molecular weight lignins contain an increased amount of phenolic
hydroxyl groups (Wada et al. 1962). Today, approximately 3.5-4 million metric
tonnes of non-homogenous lignin is being produced each year by the kraft
5
Introduction
pulping process in Sweden, from ~7 million tonnes kraft pulp/year
(Skogsindustrin 2012). This lignin is mainly burned, generating heat, in the
process of recovering pulping chemicals in the recovery boiler. A modern pulp
mill, however, has an energy surplus. Furthermore, in many cases, the recovery
boilers are the so called bottleneck. Consequently, by extracting a part of the
lignin, it is possible to de-bottleneck the recovery boiler, resulting in a pulp
production increase and, at the same time, obtaining a new product, the lignin,
from the kraft process.
1.4 EXTRACTION/PRODUCTION OF KRAFT LIGNIN
To extract lignin from black liquor, the solubility of lignin molecules can be
utilised. Because the pH of kraft liquors is ~13-14, the phenolic hydroxyl groups
in these liquors are almost completely deprotonated. By lowering the pH of the
black liquor with acid, the phenolic hydroxyl groups in lignin can be protonated
(Alén et al. 1979, Passinen 1968). The level of protonation depends on the pKa of
the lignin structures. Softwood lignin mainly consists of coniferyl alcohol. The
pKa value of coniferyl alcohol at room temperature is approximately 10.2
(Norgren and Lindström 2000). There are also some carboxylic groups with lower
pKa values (Ragnar et al. 2000); however, the carboxylic acids are often in small
amounts (Granata and Argyropoulos 1995, Norgren and Lindström 2000).
Furthermore, lignin will coagulate depending on the conditions (Sundin 2000).
The pKa depends mainly on the temperature and ionic strength (Ragnar et al.
2000) and on the molecular weight (Norgren and Lindström 2000). The dry
content of black liquor also influences the efficiency of precipitation (Alén et al.
1979 and 1985, Lin 1992).
In 1872, Tessié du Motay patented a method in which carbon dioxide was
injected into hot liquors obtained by boiling woody fibre, also known as spent
6
Introduction
liquor from the soda process. The aim of this method was to precipitate
“impurities” and consequently obtain a pure liquor that could be recaustified and
reused (Tessié du Motay 1872). Other researchers have also studied the use of
carbon dioxide to precipitate material from black liquors (Rinman 1911, Scott
1940, Reboulet 1941, Tomlinson and Tomlinson Jr. 1946 and 1948, Pollak et al.
1949). In 1910, William J. Hough patented a method in which lignin and resin
were precipitated by acidification of spent liquors from the alkaline pulping
process using an acid, such as sulphuric acid. To improve the dead-end filtration
(i.e., cake filtration, a filtration process were liquid passes through the membrane
and the particles retained building up a filter cake), the precipitated solution was
filtered at a high temperature (Hough 1910).
A problem that was recently solved by the LignoBoost process is the
displacement washing of precipitated lignin to obtain a purified lignin without
plugging the filter cake or incurring large material losses. After precipitation, the
material is filtered and the filter cake is re-dispersed at a low pH (2-4). The
suspension that is formed can then be easily re-filtered and displacement washed
(Öhman and Theliander 2006, Öhman et al. 2007). The products of this process
are lignin-lean black liquor that is returned to the pulping process and highly pure
lignin that can be used for alternative applications. For example, lignin with 0.21% ash content and 1.5-2.8% carbohydrates can be produced by this process using
softwood lignin (Tomani 2013, Öhman et al. 2007), a chemical structure
characterization has also been reported (Nagy et al. 2010). The LignoBoost
process has been a great success with regard to potentially decreasing the load on
recovery boilers, which, as previously mentioned, are usually bottlenecks in the
kraft pulping process. The LignoBoost process is, however, quite new, and no
publications to date have reported any industrial established high value
applications.
7
Introduction
Another method for the extraction of kraft lignin from black liquor, referred to as
“liquid lignin”, has been developed (Lake and Blackburn 2011). In this technique,
a phase separation occurs, and decantation is used instead of filtration (Velez and
Thies 2013). Lignin extracted with this process is reported to have an ash content
of approximately 1-2% (Velez and Thies 2013) and a carbohydrate content of
approximately 3%.
1.5 FRACTIONATION OF KRAFT BLACK LIQUOR AND
KRAFT LIGNIN
Over the years, different techniques have been used to fractionate lignin,
including acidification (Wada et al. 1962), Soxhlet extraction (Lindberg et al.
1964), gel permeation chromatography (Mörck et al. 1982), organic solvents
(Mörck et al. 1986, Yoshida et al. 1987) and ultrafiltration (Lin and Detroit 1981).
Cross-flow ultrafiltration of kraft black liquor was studied in the mid-1980s, and
various parameters were adjusted in an attempt to optimise the filtration for
concentration and purification of high molecular weight lignin (Woerner and
McCarthy 1984). Several studies have examined the fractionation of softwood
black liquor using ultrafiltration (Forss and Fuhrmann 1976, Woerner and
McCarthy 1984, Griggs et al. 1985, Alén et al. 1986, Drouin and Desrochers
1988, Uloth and Wearing 1989, Wallberg et al. 2003, Keyoumu et al. 2004, Rojas
et al. 2006, Brodin et al. 2009).
It has been demonstrated that when lignin is fractionated from kraft black liquor
by precipitation or filtration, the phenolic hydroxyl contents are increased at lower
molecular weights (Wada et al. 1962, Griggs et al. 1985, Lin 1992). These
phenolic structures have interesting properties that could be used in various
8
Introduction
applications, for example they are possible sites for modification and can be used
as an antioxidant (Weber and Weber 2010, Pouteau 2003). A few studies have
examined the possibility of separating lignin from the 1000 Da fraction of kraft
black liquor by filtration (Keyoumu et al. 2004, Rojas et al. 2006, Elegir et al.
2007, Antonsson et al. 2008, Niemi et al. 2011). Some researchers have focused
on utilising the higher content of phenolic hydroxyl groups in low molecular
weight lignin to produce new materials (Antonsson et al. 2008); however, a
convenient method for extracting the low molecular weight lignin is yet to be
found.
Little has been reported on CFF of extracted and re-dissolved kraft lignin,
although a systematic study could lead to better process understanding and
facilitate further process improvements. Woerner and McCarthy (1987) reported
studies on filtration of re-dissolved Indulin AT. As mentioned previously, the
novel LignoBoost method has gained commercial status for extracting lignin from
black liquor. The molecular size distribution of this extracted lignin is, however,
rather broad, which is possibly a drawback for certain applications (Stewart 2008,
Vishtal and Kraslawski 2011). Hence, it could be of interest to enhance the
properties of LignoBoost lignin via membrane fractionation to obtain a more
homogeneous lignin.
There have been investigations that attempt to use unfractionated and fractionated
lignin for the production of carbon fibres (Kadla et al. 2002, Braun et al. 2005,
Kubo et al. 2005, Norberg et al. 2012) and investigations into the functional
groups and thermal properties of lignin (Rojas et al. 2006, Brodin et al. 2009,
Toledano et al. 2010), but a convenient method that could predict spinning
properties a priori has not been established. For lignin-derived products that will
be processed at elevated temperatures, it could be advantageous to alter the
9
Introduction
thermal properties of kraft lignin to achieve the desired performance. By
combining an industrially feasible technique such as CFF with kraft lignin, a
systematic study on the behaviour of fractionated kraft lignin for the production of
materials with tailored thermomechanical properties would be of interest.
1.6 NANOCELLULOSE
Cellulose is interesting as a sustainable material due to its good mechanical
properties, biodegradability and renewable origin. The traditional ways to utilise
cellulose have been as “fibres”, that is more or less intact cell walls with lengths
from approximately 0.5 mm to several centimetres, in e.g. paper and textiles.
During the last few decades, however, there has been interest in producing and
using cellulose as nanosized fibres. This nanocellulose is interesting due to its
high surface to volume ratio, light-weight and high specific strength and can be
useable in, composites, foams and films (Eichhorn et al. 2010). Nanocellulose has
been divided into three types: microfibrillated cellulose (MFC), nanocrystalline
cellulose (NCC), and bacterial nanocellulose (BNC). MFC and NCC reported
with average diameters of 5-60 nm and 5-60 nm, respectively, and lengths of
several micrometres and 100-250 nm (from plant), respectively (Klemm et al.
2011). However, today other abbreviations are commonly used. Over the past ~10
years the numbers of articles and patents concerning nanofibrillar cellulose have
increased (Siró and Plackett 2010).
1.7 PRODUCTION OF NANOCELLULOSE
In the beginning of the 1980s, the first work on MFC from wood was published
(Herrick et al. 1983, Turbak et al. 1983), and in 1982 and 1983, patents were
10
Introduction
issued (U.S. 4,341,807, U.S. 4,374,702, U.S. 4,378,381). These works involved a
method wherein pulp was treated in several passes through a homogeniser with a
high pressure drop (Herrick et al. 1983, Turbak et al. 1983). Since then, other
mechanical treatments have been investigated. Super-grinding is a method in
which natural fibres pass through a specially designed grinder in cycles to
produce nanocellulose (Taniguchi and Okamura 1998). Grinding was also used in
a study by Abe et al. (2007), where nanocellulose was produced from chemically
treated wood powder. Grinding has also been used after homogenisation to
produce nanocellulose from kraft pulp fibres (Iwamoto et al. 2005). By refining
kraft pulp to different degrees before homogenisation, Nakagaito et al. (2004)
showed that nanocellulose with different properties can be created. In 2004,
Zimmerman et al. reported the use of a microfluidizer for the microfibrillation of
sulphite pulp. Cryocrushing was a method developed in 2005 by Chakraborty et
al., in which wood pulp was refined and placed in liquid nitrogen and then
crushed with a mortar. Beads have also been used to fibrillate cellulose particles
(Ishikawa and Ide 1993). Similarly, another apparatus with spherical elements
was addressed for the production of nanocellulose (Curtolo and Eksteen 2006).
Furthermore, ultrasonic treatment of natural fibres in water produces
nanocellulose, according to Zhao et al. (2007). In 2011, Chen et al. chemically
treated sieved wood powder before ultrasonication and obtained cellulose
nanofibres of 5-20 nm in width.
1.8 ENERGY-EFFICIENT METODS FOR NANOCELLULOSE
The challenge with producing nanocellulose commercially has been the high
energy demand to obtain a homogenous product. The energy input needed to
produce nanocellulose can be between 20000 and 30000 kWh/tonne (Siró and
11
Introduction
Plackett 2010). Different attempts have been made to reduce the energy required
for nanocellulose production.
Various pretreatments have been tested, such as oxidation and enzymatic
treatments (Saito et al. 2006, Henriksson et al. 2007, Pääkkö et al. 2007). In 2006,
Saito et al. reported a new way to produce nanocellulose using 2,2,6,6tetramethylpiperidine-1-oxyl (TEMPO) radical oxidation on never-dried cellulose
(i.e., produced cellulose pulp that is never dried after production). By introducing
a negative charge on the cellulose, swelling of the cellulose was increased and it
was possible to produce nanocellulose by mechanical treatment. In 2007,
Henriksson et al. showed that enzymatic pre-treatment with endoglucanase
enhanced the separation of nanofibrills produced by mechanical treatment.
Compared to acid hydrolysis as a pretreatment, enzymatic hydrolysis resulted in a
higher
nanocellulose
aspect ratio.
Enzymatic
pretreatment
to produce
nanocellulose was also performed by Pääkkö et al. (2007). By the use of sulphite
and dissolving pulp with different pretreatments prior to homogenisation in a
microfluidizer, Ankerfors and Lindström produced nanocellulose with an energy
demand of 500-2300 kWh/tonne (Ankerfors 2012), which is approximately
tenfold lower than other nanocellulose production methods (as suggested by Siró
and Plackett 2010).
Although repeated
homogenisation produces
a
relatively homogeneous
nanocellulose, other more energy efficient nanocellulose preparation methods
might produce homogeneous material. Recently, Madani et al. (2011) presented a
study where nanocellulose was separated into different fractions using a
hydrocyclone, pressurised screening or centrifugation in a fluid. Researchers have
attempted different ways of producing low-cost homogenous nanocellulose, but
12
Introduction
further technical progress to increase the numbers of methods available to
produce nanocellulose would be interesting.
1.9 HOT WATER EXTRACT FROM SOFTWOOD
Hot water treatment of wood chips or sawdust dissolves different wood
components, especially hemicelluloses, into water. The degree of hemicellulose
dissolution is higher at higher temperatures. Acetyl groups are easily cleaved off
at these conditions, forming acetic acid in the extract and consequently lowering
the liquid pH to 3-4 (Brasch and Free 1965). These acidic conditions cause
hemicellulose degradation by autohydrolysis. Hence, degraded hemicelluloses can
be found in thermomechanical pulping process liquids, prehydrolysis liquids and
steam explosion extracts. These soluble hemicelluloses are considered a byproduct in the thermomechanical process liquid and could be used for high value
application products, such as hydrogels (Lindblad et al. 2001) or gas barrier films
(Hartman et al. 2006a,b). Applications for such hemicelluloses are also for the
production of bioethanol (Ragauskas et al. 2006).
The main hemicellulose in softwood is galactoglucomannan (GGM), which
constitutes approximately 15-20% of softwood (Teleman 2009). With softwood
as the feedstock, the main component dissolved during hot water extraction is
GGM, and approximately 10% of GGM has been shown to be soluble in
thermomechanical pulping liquid (Thoron et al. 1994); the high solubility of
GGM is attributed to the high degree of acetylation and its comparatively low
molar mass (Hartman et al. 2006b). Other hemicelluloses and lignin, are also
degraded and dissolved to a lower degree. Under low pH conditions pseudo-lignin
might be formed and affect the Klason lignin values (Sannigrahi et al. 2011).
Persson et al. (2010) reported the following amounts of wood components from a
13
Introduction
liquid stream in the thermomechanical process; 11 kg hemicellulose and 8 kg of
aromatic compounds (values per tonne of produced pulp).
1.10 FRACTIONATION OF HOT WATER EXTRACT FROM
SOFTWOOD
There have been several studies on the isolation of hemicellulose from
thermomechanical process liquids and hot water extracts of softwood. Person et
al. (2010) and Andersson et al. (2007) studied the use of filtration to fractionate
hemicellulose in thermomechanical process liquids, resulting in a maximum
hemicellulose purity of approximately 80%. Willför et al. (2003) separated
different components present in thermomechanical pulping process liquids by
ultrafiltration and resin adsorption, resulting in a GGM with 95% purity. Song et
al. studied extraction of GGM by pressurised hot water extraction (2008).
However, little work has been reported regarding the combination of membrane
filtration and two-step chromatography for the isolation of all dissolved wood
components in upgraded fractions that consist of lignin, lignin-carbohydrate
complexes (LCC) (i.e., molecules containing both lignin and carbohydrate), and
GGM.
14
Aim of Investigation
2. AIM OF INVESTIGATION
The objective of this thesis followed the main goal of the Wallenberg Wood
Science Center and its subdivision “From Wood Chips to Material Components”,
which this work was a part of. The aim of this thesis is to find and study new
interesting raw materials and processes from softwood through the use of crossflow filtration (CFF).
The focus on the utilisation of CFF to produce more homogenous raw materials is
essential because this technique could be critical for the production of high value
applications from wood components.
Lignin from existing processes, black liquor and re-dissolved LignoBoost lignin,
have been studied for potential CFF fractionation to produce products with
specific properties. Nanocellulose and hemicellulose have also been used in
combination with cross-flow filtration for the production of more homogenous
raw materials
15
Experimental
3. EXPERIMENTAL
Detailed descriptions of the experimental methods are given in the related papers.
3.1 MATERIALS
 Black liquor
In papers I and II, industrial weak black liquor (pH >13, dry content ~17 %) from
kraft pulping of softwood was generously supplied by the Billerud Gruvön pulp
and paper mill, Sweden.
 Kraft lignin
In paper III, LignoBoost lignin (lignin extracted from softwood kraft black liquor)
was kindly supplied by LignoBoost Demo AB demonstration plant in
Bäckhammar, Sweden.
 Kraft pulp
In paper IV, oxygen delignified softwood kraft pulp with kappa number 10.0 was
kindly supplied by SCA Östrand Pulp Mill in Sundsvall, Sweden.
 Softwood
In paper V, sawdust of Norwegian spruce wood (from a Swedish sawmill) was
used as material for pressurised hot water extraction at 163°C for 60 minutes.
16
Experimental
3.2 MEMBRANES AND FRACTIONATION
Paper I: fractionation of weak kraft black liquor was performed with a ceramic
membrane with 1 kDa cut-off (Novasep, France), to produce 1 kDa permeate to
use for the main aim, i.e., extraction of low molecular lignin.
Paper II: fractionation of weak kraft black liquor was performed with ceramic
membranes of 1 and 5 kDa cut-off (Novasep, France) and 10 kDa regenerated
cellulose (Merck Millipore, USA) membranes, for use in the following steps of
precipitation to obtain lignins of different molecular weights.
Paper III: cross-flow filtration was used to study fractionation of re-dissolved
LignoBoost lignin. The system consists of a 10 L tank, a heating element, a gear
pump, a flow meter and a Valisette membrane unit with ceramic membranes
(TAMI Industries, Nyons, France). At a cross-flow velocity of 1m/s and a TMP of
3 bar, 5 kDa cut-off single channel membranes with a diameter of 0.6 cm were
used.
Paper IV: fractionation of nanocellulose was performed with ceramic membranes
with 0.45 µm and 0.8 µm cut-off (Novasep, France), to obtain homogenous
nanocellulose.
Paper V: fractionation of hot water extract was performed with ceramic
membranes with 1 and 5 kDa cut-off (Novasep, France) to obtain a size
fractionated and purified sample for use in the next step of adsorption
chromatography.
17
Results and Discussion
4. RESULTS AND DISCUSSION
4.1 PRODUCTION OF LIGNIN FRACTIONS (PAPERS I-III)
In this study, cross-flow filtration (CFF) fractionation of lignin in weak black
liquor, and fractionation of re-dissolved LignoBoost lignin for the production of
lignin with low molecular-weight dispersity were investigated.
Through CFF with several different membrane cut-offs, lignin fractions with
specific properties could be produced from weak black liquor. In total, seven
samples were obtained (including the unfractionated sample, i.e., reference): > 10
Da, > 5 kDa, 5-10 kDa, 1-5 kDa, 0-5 kDa, and 0-1 kDa. See Figure 3.
Furthermore, CFF fractionation of re-dissolved LignoBoost lignin was seemingly
affected by the different conditions chosen in the trials.
Figure 3. Scheme of obtained CFF fractionations from weak black liquor.
18
Results and Discussion
 Precipitation and filtration of weak black liquor
To extract solid material from the weak black liquor samples (see Figure 3), the
unfractionated sample and all fractionated samples, except for the 0-1 kDa sample
(see below), were precipitated by adding sulphuric acid (H2SO4 6 M) to pH 9 at
~70°C. The precipitated solids were separated from the liquid, using cake
filtration (i.e., a filtration process were liquid passes through the membrane and
the particles retained building up a filter cake). The filterability of the precipitated
solids was improved by increasing the ionic strength, either by evaporating water
or adding salt, which enhanced lignin coagulation. The electrical charge that is
formed by ionised hydrophilic groups forms a colloidal lignin, the stability of
which is affected by ionic strength (Passinen 1968). Furthermore, Rinman (1911)
reported that the precipitation is improved by increasing salt concentration,
effectively increasing the ionic strength of the solution.
 Precipitation and filtration of <1000 Da cut-off permeate
The precipitation mentioned in the previous paragraph could be performed at
~70°C for all samples except for the low molecular weight lignin 0-1 kDa. During
precipitation of the 0-1 kDa sample, fairly large, rigid particles/agglomerates of
lignin material were formed at ~70°C, whereas lowering the pH at ambient
temperature resulted in the formation of fine particles of precipitated lignin. This
phenomenon was believed to be caused by a lower glass transition temperature
(Tg). At temperatures above Tg, an amorphous polymer material becomes softer
due to segment movements in the polymer chain. Yoshida et al. (1987) reported
Tg temperatures of 32°C and 62°C for weight-average molecular weights lignin of
620 Da and 1300 Da, respectively. Researchers have reported that condensation
reactions could take place when the chain mobility of lignin increases (Guigo et
al. 2009, Cui et al. 2013). Similar agglomerate phenomena have also been
19
Results and Discussion
reported by other research groups, but not for low molecular weight lignin. In
those studies, higher temperatures were reported to melt the precipitate or form
large tacky clumps (Rinman 1911, Uloth and Wearing 1989). It was suggested
that precipitation should be performed just below the coagulation temperature
(Merewether 1962). Lignin in black liquor fractions of 15-0 kDa and 15-1 kDa
has also been reported to clump at higher temperatures, explained by higher
density coagulation (Wallmo and Theliander 2009). Our predictions about lower
Tg for the 1000 kDa in Paper I were supported by Tg measurements in Paper II,
furthermore, this fraction contained a higher amount of aromatic hydroxyl groups
(as shown and discussed later in this thesis).
 Composition of precipitated lignin
Chemical analysis of the unfractionated kraft lignin (i.e., reference) and the
fractionated lignin samples was performed by Klason lignin analysis (TAPPI 222
om-02), sugar analysis (SCAN-CM 71:09), acid soluble lignin analysis (according
to Dence 1992), and ash content analysis (ISO 1762). For more details, see Papers
I-II.
From the chemical composition (Table 1), it can be seen that Klason lignin
content is quite high for all samples (roughly ~95%), while acid soluble lignin
content increases with decreasing molecular weight. In general, the lignin
fractions are quite low in carbohydrate content, which is in agreement with
previous study by Öhman et al. (2007). The largest amount of carbohydrates is
found in higher molecular weight lignin. Because the 5-10 kDa fraction only
contains 0.3% carbohydrates, it is concluded that most of the carbohydrates are
found in the >10 kDa fraction. These carbohydrates might be in the form of
lignin-carbohydrate complexes (LCCs). It has previously been reported that most
20
Results and Discussion
of the lignin present in wood, especially in the pulp, exists in the form of LCCs
with various types of hemicelluloses (Lawoko et al. 2005, Balakshin et al. 2011).
Griggs et al. (1985) also suggested that some xylan was covalently bonded to
kraft lignin. Finally, the ash content was quite low in all samples, indicating good
washing of the samples.
Table 1. Composition of reference and fractionated lignin samples (wt%).
Sample
(cut-off kDa)
Ref.
> 10
>5
5-10
1-5
0-5
0-1
Ref. (Paper I)
0-1 (Paper I)
Klason lignin
Acid-soluble
lignin
Carbohydrates
Ash
%
96.9
94.4
94.9
93.8
97.8
94.3
97.2
93.7
98.2
%
2.0
1.0
1.2
2.9
3.7
5.6
5.7
2.0
3.7
%
1.7
5.2
3.4
0.3
0.2
0.3
0.1
2.7
0.2
%
0.1
0.2
0.5
n/a
0.2
1.1
2.2
0.1
0.9
 Elemental analysis for precipitated lignin
Elemental analysis was performed to investigate the amount of sulphur in the
unfractionated (i.e., reference) and the fractionated solid samples (external service
provided by the Elemental Analysis Unit of the Santiago de Compostela
University, Spain). For more details, see Papers I-II.
From the elemental analysis (Table 2), the unfractionated kraft lignin sample (i.e.,
reference) contained approximately 2% sulphur, a mean of the two reference
values in Table 2. Interestingly, the results show increasing sulphur content with
decreasing molecular weight, despite the fact that the samples were thoroughly
21
Results and Discussion
washed with acidic water, ~pH 2. Low molecular weight lignin seems to be the
main material that contributes to the sulphur content of the lignin sample.
Due to the high amount of sulphur in the lignin, extraction with organic solvents
— toluene and pentane — was performed in an attempt to remove sulphur. A
large amount of sulphur was extracted from the lignin samples, but some sulphur
was retained, especially in the low molecular weight fraction. An explanation for
this could be the reaction mechanism responsible for the lignin dissolution in kraft
cooking, where sulphur is incorporated into lignin via thiirane (episulphide)
formation (Gierer 1980), which might have remained covalently bound.
Therefore, small lignin fragments in the samples could contain a higher ratio of
organically bound sulphur. Other researchers have also reported higher sulphur
content in low molecular weight lignin, but not the same high content. Wallmo
and Theliander (2009) obtained fractions from ultrafiltration of 0-15 and 1-15
kDa black liquor, which contained 4% and 3.2% sulphur, respectively, while the
original sample contained 2.9% sulphur. However, if all of the sulphur in the
reference samples (Table 2) was bonded to only low molecular weight lignin, the
sulphur value for fraction 0-1 kDa would still be too high for all of it being bound
to lignin (based on calculations from the lignin yield in the reference and 0-1 kDa
fraction, Table 3). Another explanation might be that low molecular weight lignin
behaves differently during the washing and filtering stage, and hence, a lesser
amount of sulphur is removed during the washing stage. However, it is not clear if
this sulphur only comes from the cooking chemicals or is also incorporated into
the lignin molecules during kraft cooking. The reason for the high sulphur content
observed is not known at this time and needs to be investigated further.
22
Results and Discussion
Table 2. Elemental analyses of reference and fractionated lignin samples (wt%).
Sample
(cut-off kDa)
Before extraction (%)
After extraction (%)
S
N
C
H
S
N
C
H
Ref.
2.5
0.1
64.9
5.5
1.8
0.1
64.7
5.9
>10
1.7
0.1
62.3
5.1
1.6
0.1
62.1
5.9
>5
1.8
0.1
64.9
5.3
1.6
0.1
63.5
5.8
5-10
2.0
0.1
62.6
5.4
1.7
0.1
65.1
6.0
1-5
7.2
0.1
62.1
5.4
3.0
0.1
68.5
6.4
0-5
23.0
0.1
51.6
4.8
4.0
0.1
66.1
6.3
0-1
27.1
0.1
48.9
4.2
4.8
0.1
63.4
5.9
Ref.(Paper-I)
1.7
0.1
64.9
6.3
0-1 (Paper-I)
8.6
0.1
61.3
5.9
0-1*
25.1
0.1
49.6
4.8
*Extra sample precipitated with CO2 and washed with water/hydrochloric acid (pH ~2).
 Yield for <1000 Da cut-off precipitated lignin
The amount of lignin obtained after precipitation and acidic washing was reported
for the unfractionated (i.e., reference) and the 0-1 kDa solid samples (Paper I). To
calculate lignin yield, the following analyses were performed on the precipitated
lignin sample: Klason lignin (TAPPI 222 om-02), acid-soluble lignin (according
to Dence 1992), and precipitated material in the acid-soluble filtrate. For more
details, see Paper I.
Table 3 outlines the amount of lignin precipitated at pH 9, the material in the
filtrate precipitated at pH 2, and the sum of both. However, by extracting the
sulphur content from the precipitated lignin values in Table 3, it can be
determined that low molecular weight lignin precipitated to pH 9 constitutes
approximately 30% of the sample (i.e., dividing 6.8 g/L and 21.3 g/L after
subtracting the amount of suphur in each sample). It is well known that the
amount of low molecular weight lignin depends on the cooking conditions and
23
Results and Discussion
that prolonged cooking at the end of the final phase of degradation can increase
the amount of low molecular weight lignin (Dong and Fricke 1995).
Table 3. Lignin yield in reference and fractionated liquid samples.
Sample
(cut-off kDa)
Ref. (Paper-I)
0-1 (Paper-I)
Lignin precipitated
to pH 9
(excl. filtrate)
g/L
21.3
6.8
Material in filtrate from pH
9 precipitated to pH 2
(filtrate)
g/L
5.8
6.0
Total
(incl. filtrate)
g/L
~27
~13
 Molecular weight data of precipitated lignin
Size exclusion chromatography (SEC) was performed on the unfractionated (i.e.,
reference) and the fractionated solid samples to determine the effect of membrane
separation for the fractionated samples. All samples were acetylated before
analysis in the SEC-system with Tertrahydrofuran (THF) as eluent, and
polystyrene standards were used for calibration. For more details see Papers I-II.
As seen in Table 4, the weight-average (Mw) and number-average (Mn) molecular
weights are in fairly good agreement with the membrane cut-offs. The samples in
Paper I were acetylated according to Gellerstedt (1992) with ethanol, pyridine and
acetic anhydride, while the samples in Paper II were acetylated by
acetobromination according to Guerra et al. (2006), which could be an
explanation for the differences between the results from Paper I and Paper II, but
the trend still shows that more homogeneous materials with narrower molecular
weight distributions are obtained with cross-flow fractionation. High-average
molecular weight, Mz, for 0-1 kDa sample (Paper I) shows a large decrease in Mz
compared to its reference (i.e., unfractionated sample).
24
Results and Discussion
Table 4. SEC data for acetylated reference and fractionated lignin samples, including
high-average (Mz), weight-average (Mw) and number-average (Mn) molecular weights as
well as molecular-weight dispersity (ĐM).
THF-SEC
Sample
(cut-off kDa)
Mz
Mw
Mn
ĐM
Ref.
-
20200
5000
4.1
>10
-
33500
9500
3.5
>5
-
28200
8000
3.5
5-10
-
4900
2300
2.2
1-5
-
4700
2000
2.3
0-5
-
4100
1700
2.4
0-1
-
2700
1200
2.1
Ref. (Paper-I)
13109
3525
847
4.2
0-1 (Paper-I)
2590
1096
508
2.2
 Chemical characteristics and thermal properties of precipitated lignin
Knowledge of the homogeneity in chemical characteristics and thermal properties
of the unfractionated (i.e., reference) and the fractionated solid lignin samples
could be an advantage for the development of high value application products.
Therefore, aliphatic hydroxyl and aromatic hydroxyl groups were analysed by
phosphorous-31 NMR, the glass transition temperature (Tg) was analysed by
differential scanning calorimetry (DSC), and the decomposition temperature (Td)
was analysed by thermogravimetric analysis (TGA) (see Table 5). For more
details, see Paper II.
Prior to phosphorous-31 NMR analysis each sample was functionalised by
phosphitylation as described in Paper II. From the
31
P NMR data, it can be seen
that phenolic content increases with decreased molecular weight and vice versa
for aliphatic hydroxyl content, in agreement with previous studies (Wada et al.
25
Results and Discussion
1962, Zhang and Gellerstedt 2001). Due to the cleavage of β-O-4 linkages during
kraft cooking, the low molecular weight lignin fractions will have a higher
phenolic content. An advantage of higher phenolic hydroxyl content is that these
groups have a relatively high reactivity (Weber and Weber 2010), which might be
beneficial for future applications.
For the use of lignin in thermal processes, it could be advantageous to use kraft
lignin fractions with different glass transition temperatures and decomposition
temperatures. To determine the thermal stability of kraft lignin samples, Td was
determined by TGA, where the temperature of decomposition is defined as the
temperature at which 5% weight loss has occurred (Figure 4). In Table 5, it can be
seen that Td differs between the lignin fractions. In general, Td increases with
molecular weight except for the 5-10 kDa sample, which showed the highest Td.
An explanation for this might be the low molecular-weight dispersity (ĐM) for the
5-10 kDa sample (Table 4).
In the DSC trials, Tg was determined from the second heat scan, while the first
heating scan was performed to eliminate thermal history stored within the
polymer. However, it has been reported that cross-linking occurs during the first
heating scan, which could increase the observed Tg (Cui et al. 2013). The Tg for
the unfractionated kraft lignin sample (i.e., reference) was 144°C, but after
fractionation, Tgs between 70 and 170°C were obtained. In higher molecular
weight lignin, the movement of the chain mobility of lignin is reduced, resulting
in a higher Tg. The dependence of Tg on molecular weight was reported earlier for
lignin samples obtained via solvent fractionation (Yoshida et al. 1990). Although
a trend can be seen in Tg for fractionated lignin, the exact Tg values are uncertain
because Tg is affected by low molecular weight contaminants and irreversible
condensation reactions (Guido et al. 2009, Cui et al. 2013).
26
Results and Discussion
Figure 4. Thermogravimetric analysis (TGA) of reference and fractionated lignin
samples. Decomposition temperature was estimated at 5% of weight loss.
Table 5. Functional groups, Tg and Td in the reference and fractionated lignin samples.
31
P NMR
Sample
(cut-off kDa)
Tg by DSC
Td by TGA
Ph-OH
mmol/g
Aliph-OH
mmol/g
°C
°C
Ref.
3.5
2.2
144
215
> 10
2.8
2.6
170
217
>5
2.9
2.3
159
190
5-10
3.9
1.9
140
240
1-5
3.7
1.4
94
193
0-5
3.6
1.4
82
175
0-1
4.1
1.3
70
163
Dynamic mechanical analysis was used to investigate the viscoelastic response of
dry lignin fractions using parallel-plate dynamic torsional rheometry. In Figure 5,
the storage (G′) and loss (G″) moduli response is presented for four samples:
27
Results and Discussion
unfractionated kraft lignin (reference), 0-5 kDa, 1-5 kDa, and >5 kDa. For the
unfractionated (i.e., reference) and > 5 kDa samples, the degree of softening (i.e.,
the change in G′ from the glassy to the rubbery phase) results in a drop of one to
two decades in G′, while the low molecular weight samples — 0-5 kDa and 1-5
kDa — show up to five decades of drop in G′. Differences in the molecular
weight clearly affected the degree of softening, and 31P NMR data (Table 5) show
that the higher molecular weight fractions had higher proportions of aliphatic-OH
groups than the lower molecular weight fractions. Aliphatic-OH groups are more
capable of intermolecular hydrogen bonding, resulting in greater mobility
restriction (Dallmeyer et al. 2013, Uraki et al. 2012). Except for cross-linking that
can occur during heating (Guido et al. 2009, Cui et al. 2013), higher molecular
weight kraft lignin is more cross-linked (i.e., condensed) by nature, which might
contribute to the lower degree of softening compared to lower molecular weight
samples. When G″ > G′, the material is considered to be liquid-like, and for low
molecular weight lignin (i.e., 0-5 kDa and 1-5 kDa), the rubbery plateau is
reached. A clear rubber plateau is not seen for the high molecular weight samples
(i.e., reference and >5 kDa), within chosen temperature interval.
From Figure 5, it seems that the low molecular weight fractions (i.e., 0-5 kDa and
1-5 kDa, have better high-temperature flow-like behaviour compared to higher
molecular weight fractions. Hence, the low molecular weight fractions could be
useful for applications where flow is important, e.g., melt spinning fibres. In a
recently published article 15 kDa cut-off permeate (i.e., <15 kDa) softwood kraft
lignin produced a smooth carbon fibre (Nordström et al. 2013). Dynamic rheology
could be an efficient tool to evaluate the behaviour of lignin samples during the
various thermal processing reactions. For example, dynamic rheology could be
used to evaluate the spinnability of a lignin sample and to predict the spinning
28
Results and Discussion
temperature, as described elsewhere (Dr. Sudip Chowdhury and Prof. John Kadla
personal communications, Chowdhury et al. 2012).
Figure 5. Dynamic rheological response measured at the first heat responses for
unfractionated (Ref.) and fractionated lignin samples.
 Cross-flow filtration of dissolved LignoBoost lignin
As described in the introduction, the LignoBoost process is a commercial process
with regard to lignin extraction from the kraft process. To facilitate the use of
extracted kraft lignin for high value applications, it could be interesting for future
LignoBoost lignin users to fractionate the lignin according to the desired
molecular weight to enhance the properties of kraft lignin. Therefore, the aim of
this work was to separate already extracted kraft lignin by CFF and to study the
results of varying the following parameters: lignin concentration, pH of lignin
solution, and ionic strength of the lignin solution.
29
Results and Discussion
For fractionation, a ceramic membrane of 5 kDa was used with a transmembrane
pressure (TMP) of 3 bar and a velocity of 1 m/s generated by a gear pump. Each
dissolved LignoBoost lignin sample was run in the filtration system for ~24 hours
with permeate recirculation to the feed tank. Dry contents of the permeate
samples were determined according to SCAN-N 22:96. For more details, see
Paper III.
Table 6 summarises the seven different lignin solutions that were prepared for
CFF trials. Three different concentrations, three different pH and two different
ionic strengths were chosen as parameters. Because the sample with 20% lignin
was dissolved with the highest sodium hydroxide concentration, this sample
obtained the highest ionic strength (1.56). Therefore, all other samples were
adjusted to an ionic strength of 1.56 with Na2SO4, with one exception: one sample
with a lignin concentration of 10% and pH 12.5 was prepared at an even higher
ionic strength by adding 30% extra Na2SO4 (marked as >I)*. Each lignin sample
was dissolved for at least four days before CFF trials were performed.
Table 6. Experimental conditions where X denotes the different lignin solutions prepared.
pH
>13.5
13.5
13.0
12.5
12.5
Lignin concentration (w/w %)
20%
X
-
15%
X
X
-
10%
X
X
X
X >I*
The effect of varying lignin concentration on permeate flux can be seen in Figure
6; higher lignin concentrations decrease the permeate flux. This trend could be
explained by the phenomena of a polarisation layer, where a higher concentration
30
Results and Discussion
lowers the permeate flux (Berg and Smolders 1990, Bowen and Jenner 1995).
That the 20% and 15% samples are relatively close to each other and a bit far
from the 10% sample might at least partly be related to the water-lignin ratio.
These ratios are 9, 5.67 and 4 kg H2O/kg lignin, where an increase in the amount
of water present might affect the polarisation layer, reducing the layer’s
resistance. However, other effects cannot be excluded based on this study, and
further studies could yield more understanding about this behaviour.
Flux [L/h,m2]
15
10
10% lignin pH 13.5 (9 kg
H2O/kg lignin)
5
15% lignin pH 13.5 (5.67
kg H2O/kg lignin)
0
20% lignin pH 13.5 (4 kg
H2O/kg lignin)
0
10
20
30
40
Time [h]
Figure 6. Effect of lignin concentration for lignin dissolved at pH 13.5.
Figure 7 presents the effects of different pH on permeate flux, indicating that
higher permeate fluxes are achieved at higher pH. An explanation might be that
lignin becomes more deprotonated at higher pH, resulting in a greater charge on
the lignin molecules. Through a higher repulsion between lignin molecules, the
polarisation layer might become more permeable. It has been reported that the
pKa value of coniferyl alcohol at room temperature is approximately 10.2 but
increases with increasing molecular weight of the lignin molecule (Norgren and
Lindström 2000).
31
Results and Discussion
Flux [L/h,m2]
15
10% lignin pH 13.5
10
10% lignin pH 13.0
5
10% lignin pH 12.5
15% lignin pH > 13.5
0
0
10
20
30
40
15% lignin pH 13.5
Time [h]
Figure 7. Effect of pH for lignin dissolved at 10 and 15% concentrations.
Figure 8 presents the effect of increased ionic strength on permeate flux, showing
that higher permeate fluxes are achieved at higher ionic strengths. Norgren and
Lindström (2000) have shown that an increase in ionic strength lowers the
apparent pKa of lignin. This could be an explanation for the increased flux at a
higher ionic strength, because lignin with a lower pKa would be more charged.
Furthermore, as in the case of the pH effect, this increased charge could generate
higher repulsions between molecules and hence reduce the resistance of the
polarisation layer. This suggests that pKa had a larger effect on the results than
shielding of charges did, which is the other effect of increasing ionic strength.
This result might be explained by the Debye length, which shows smaller changes
at higher ionic strengths.
32
Results and Discussion
Flux [L/h,m2]
15
10
10% lignin pH 12.5, > I
5
10% lignin pH 12.5
0
0
10
20
30
40
Time [h]
Figure 8. Effect of ionic strength for lignin dissolved at pH 12.5.
Table 7 and Figure 9 shows SEC data for the permeate sample, which has been
precipitated, washed and acetylated for THF-SEC analysis. Unfortunately, the
sample at 20% lignin concentration most likely condensed during the
precipitation trials and was therefore not included. There are no large differences
in the SEC data or clear trends between the different samples, indicating that
separation of lignin molecules is performed by the membrane. Published
molecular weight data (either Mw or Mn) on acetylated LignoBoost lignin report
molecular weights of 5500 Da (Moosavifar et al. 2006). This molecular weight
data for the starting material indicate that fractionation with the 5 kDa membrane
was successful, as the permeate fractions have lower molecular weight values.
Table 7. THF-SEC data for permeate samples, weight-average (Mw), number-average
(Mn), and molecular-weight dispersity (ĐM).
Permeate-Sample
15% pH >13.5
15% pH 13.5
10% pH 13.5
10% pH 13.0
10% pH 12.5
10% pH 12.5, > I
Mw
1600
1700
1700
1600
1700
1400
Mn
1000
900
1000
1000
1000
900
33
ĐM
1.54
1.83
1.68
1.67
1.80
1.60
Results and Discussion
Figure 9. THF-SEC analysis for permeate samples with UV detection at 280 nm.
Table 8 shows rejection coefficients for all samples after CFF. The rejection
coefficient, R = 1-(cp/cf), is calculated from the concentration in the permeate (cp)
and the concentration in the feed (cf). All trials, except for the experiment with
20% lignin content, have similar rejection coefficients. The fact that the
experiments with lignin contents of 10% and 15% have similar rejection
coefficient and very similar molecular size distributions in the permeates suggests
that it is the properties of the membrane itself that serve as the separating agent
for lignin fractionation. The reason why the 20% experiment has a lower rejection
coefficient is not fully understood. It should, however, be remembered that the
rejection coefficients are based on concentrations and that the permeate flux of the
20% experiment was very low.
34
Results and Discussion
Table 8. Rejection coefficients for permeate samples.
Permeate Sample
~0 h
Rejection coefficient
~9 h
~22 h
(%)
39
35
20% pH 13.5
17
15% pH > 13.5
53
54
40
15% pH 13.5
54
58
44
10% pH 13.5
48
46
32
10% pH 13.0
55
49
46
10% pH 12.5
55
58
50
10% pH 12.5, > I
56
54
47
As shown in Table 8, there is a difference in the rejection coefficients of the 20%
lignin and samples with lower lignin concentrations, indicating that more lignin,
relative to the feed concentration, is passed through the membrane for 20% lignin.
Additionally, it was shown that the permeate flux is different for different
samples. By combining the rejection coefficient and permeate flux, the actual
amount of low molecular weight lignin being produced in ~kg lignin/h,m2 can be
calculated (Figure 10 and Figure 11).
Figure 10 shows the mass flux of lignin in the permeate. However, this value does
not account for feed concentration, and therefore, the actual relative production is
unknown. For the actual value for low molecular weight production, the rejection
coefficient needs to be included in the calculations. Figure 11 shows the
production in the permeate flux relative to the rejection coefficient by multiplying
the value in Figure 10 with 1-the rejection coefficient (i.e., cp/cf; where cp is the
permeate concentration and cf is the feed concentration). By comparing Figure 10
and Figure 11 it is evident that the calculated production of lignin in the permeate
flux changes when the rejection coefficient is included in the calculation. The
35
Results and Discussion
smallest change is observed for the sample with the high lignin concentration of
20%, which is the sample with the lowest rejection coefficient (Table 8). For the
other samples, there is a larger change when the rejection coefficient is included
in the calculations, which reduces the calculated amount of fractionated lignin
produced. This result demonstrates that, when optimising the cross-flow
separation, it is important to not only investigate the permeate flux but also to
investigate the rejection coefficient before concluding appropriate system
parameters. In this study, it is shown that production of large amounts of low
molecular weight kraft lignin can be accomplished from solutions with a lower
lignin concentration (i.e., 10%). Furthermore, higher pH and ionic strength is
beneficial to achieve a higher lignin production of low molecular weight lignin.
To perform cost estimation other parameters also needs to be considered. These
kinds of production costs have been reported for membrane filtration of black
cp*(cp/cf) * Flux [~kg lignin/day,m2]
liquor (Wallberg et al. 2005, Jönsson and Wallberg 2007).
cp * Flux [~kg lignin/day,m2]
20
15
10
5
0
0
Time [h]
20
10. Amount of lignin in the
Figure 11.
permeates. Based on permeate
concentration, cp in wt %.
20
20% pH 13.5
15% pH > 13.5
15
15% pH 13.5
10% pH 13.5
10
10% pH 13.0
10% pH 12.5
5
10% pH 12.5, > I
0
0
Time [h]
20
Figure 11.
10. Amount of lignin in the
permeates. Based on cp (wt%) and
1-R.
36
Results and Discussion
4.2 FRACTIONATION OF NANOCELLULOSE (PAPER IV)
In the search for new energy efficient methods to produce homogenous
nanocellulose, a novel process called nanopulping has been developed and
combined with fractionation. In nanopulping cellulose nanoparticles/fibres are
manufactured by a disintegration method integrated with a pulping-type process
(i.e., pressurised reactor with a valve). In this study, kraft pulp was used as a raw
material and pulp pretreatments were performed before nanopulping (i.e.
enzymatic pretreatment and TEMPO oxidation). The product from the
nanopulping step was then fractionated with CFF to obtain more homogenous
nanocellulose.
 Energy demand for nanopulping
The energy demand of the nanopulping process has been estimated, and is ~450
kWh/tonne (~1.6 GJ/tonne ). This energy demand is much lower than that
reported for nanocellulose production, which can be between 20000 and 30000
kWh/tonne (Siró and Plackett 2010), and can also be compared to that reported by
Ankerfors (2012), who reported an energy demand of 500-2300 kWh/tonne for
nanocellulose production.
The energy calculation is based on the experimental conditions for nanopulp
production (i.e., 1 g dry pulp/L water and 160°C) combined with how the process
might be implemented in industry. A pulp consistency of 1 weight percent, which
is heated from 90-160°C from the pulp mill’s system, corresponds to an energy
demand of ~760 kWh/tonne (~2.7 GJ/tonne). However, by using the existing
environment, it is possible to recover energy by flashing from 160-130°C,
resulting in a total energy demand for the nanopulping process of ~450
kWh/tonne (~1.6 GJ/tonne).
37
Results and Discussion
 Effect of different pretreatment methods
As a starting raw material, oxygen delignified softwood kraft pulp with original
kappa number 10.0 was used in the nanopulping process. Different pretreatments
of the kraft pulp were performed before the nanopulping, enzyme pretreatment as
well as TEMPO oxidation. After nanopulping, the obtained samples were size
fractionated with CFF, resulting in permeate fractions with different
concentrations.
Table 9. Results for non-pretreated kraft pulp (Reference), endoglucanase and TEMPO
pretreated samples: after nanopulping (NP) and after filtration (i.e., in the permeate,
Perm.).
Endoglucanase
Pretreatment
Reference
Sample
TEMPO Pretreatment
1
Conc.
after NP
(g/l)
1.01
Conc.
of Perm.
(g/l)
0.22
Conc.
after NP
(g/l)
1.02
Conc.
of Perm.
(g/l)
0.26
Conc.
after NP
(g/l)
1.08
Conc.
of Perm.
(g/l)
0.14
2
1.02
0.29
0.99
0.28
1.03
0.17
3
1.03
0.24
0.98
0.27
1.02
0.17
Average
1.02
0.23
1.00
0.27
1.05
0.16
STDEV
0.008
0.02
0.02
0.01
0.03
0.02
Reject.coeff. (%)
78
73
85
In Table 9, results from endoglucanase and TEMPO pretreated samples are
presented.
Enzymatic
pretreatment
with
monocomponent
endoglucanase
decreased the rejection coefficient from 78% to 73%, which indicates an increase
of nanopulped material in the permeate. Engström et al. (2006) reported that
monocomponent endoglucanase can be used to increase the reactivity and
accessibility of dissolving pulp. Furthermore, it was demonstrated that pulp
pretreatment with monocomponent endoglucanase enhance the mechanical
38
Results and Discussion
separation of nanofibrils (Henriksson et al. 2007). These effects from
endoglucanase could be a reason for the improved efficiency of nanopulping,
indicated by the higher content of material in the permeate (Table 9). The high
rejection coefficient for the TEMPO pretreated sample is, however, not in
agreement with previous results reported by Tsuguyuki et al. (2009), where
TEMPO-oxidised pulp treated in a homogeniser showed a large increase in
fibrillation.
 Fractionation by cross-flow filtration
After nanopulping, the nanopulped material was fractionated via tubular CFF with
ceramic membranes of two different cut-offs: 0.8 µm and 0.45 µm. The
nanocellulose in the permeate was thereafter characterised with AFM, where the
sample image’s scan size is 5 μm x 5 μm.
Before filtration:
a) Nanopulped sample
After filtration:
b) Higher cut-off
c) Lower cut-off
Figure 12. AFM images of material after nanopulping (a) and after filtration (b and c),
scan size 5 μm x 5 μm.
In Figure 12, image a), it can be observed that the nanopulped material, contains
both long and short nanocellulose with dimensions of approximately 10-20 nm
wide and ~0.5 μm long. In images b) and c), the nanocellulose after CFF (i.e., in
39
Results and Discussion
the permeate) is presented. It is apparent that nanocellulose can be separated into
different fractions with more homogeneous size distribution by CFF.
4.3 FRACTIONATION OF HOT WATER EXTRACT FROM
SOFTWOOD (PAPER V)
The purpose of this investigation was to study the combination of two
downstream separation methods, CFF and chromatography, for fractionation and
purification of wood components. Hot water extract of softwood sawdust
produced in the lab was used as a model liquid to simulate galactoglucomannan
rich process liquids, such as thermomechanical pulping liquid. The target of the
separation was to obtain three fractions: galactoglucomannan (GGM), lignincarbohydrate complexes (LCCs), and lignin. The isolation method consisted of
CFF, where high and low molecular species were removed, followed by two-step
adsorption chromatography to separate the wood species by affinity.
 Yield for membrane fractionated hot water extract
Pressurised hot water extraction of softwood sawdust was performed at 163°C for
60 minutes. The total yield of extracted material, after removing sawdust and
fibres, in the filtrate was ~14.4%, based on the initial amount of saw dust. This
extract was thereafter fractionated by CFF with 5 kDa and 1 kDa membrane cutoffs, to obtain a purified fraction between 1 and 5 kDa The amount of dissolved
material in the 1-5 kDa fraction was 34% (see Table 10). The amount of extracted
material in the 1-5 kDa sample was almost three times as much as that in the >5
kDa sample. This result was in accordance with an experimental condition chosen
from Song et al. (2008) to produce hot water extracted material within this
40
Results and Discussion
molecular weight region. Hence, the 1-5 kDa sample was used for the next step of
separation by chromatography.
Table 10. Material yield in each filtered fraction (based on dissolved material 14.4%).
Samples
Yield in CFF fractions wt %
Cross-flow filtration:
>5 kDa
15
1-5 kDa
34
0-1 kDa
43
 Composition of membrane fractionated hot water extract
The chemical composition of the unfractionated hot water extract (i.e., reference)
and CFF fractionated samples was determined by Klason lignin analysis (TAPPI
222 om-02), sugar analysis (SCAN-CM 71:09) and acid soluble lignin (according
to Dence 1992) on a freeze dried aliquot of each liquid sample.
The data in Figure 13 show that the >5 kDa and the 1-5 kDa fractions contain
greater amounts of carbohydrates and smaller amounts of “other” material
compared to the unfractionated reference sample and 0-1 kDa permeate. Klason
lignin content was highest for the 0-1 kDa sample. This result might be explained
by the low molecular weights of the small amounts of lignin that are extracted.
This result is consistent with a previous report by Örså et al. (1997), wherein most
of the aromatic species in the hot water extract states to be of low molecular
weight.
41
Results and Discussion
100.0
80.0
60.0
40.0
20.0
0.0
Reference
Carbohydrates
>5 kDa
1-5 kDa
Klason lignin
0-1 kDa
Acid soluble lignin
Other
Figure 13. Chemical composition before and after CFF (wt%). Acid soluble lignin was
present in very low amounts and is therefore not notable in this table.
From carbohydrate analysis in Figure 14 it is obvious that GGM and
arabinoglucuronoxylan were extracted and constituted a major part of the extracts.
The analysis suggests that this GGM is of higher molecular weight because it is
mainly found in the >5 kDa and 1-5 kDa fractions, while arabinose, mainly the
side chain of xylan, is cleaved off xylan and enriched in the 0-1 kDa fraction.
Reference
>5 kDa
1-5 kDa
0-1 kDa
Ara %
Gal %
Glc %
Xyl %
Man %
Other
Klasonlignin
Figure 14. Distribution of carbohydrates before and after CFF (wt%).
42
Results and Discussion
 Molecular weight distribution for fractionated hot water extract
To analyse the molecular weight distributions of the initial hot water extract and
CFF fractionated samples an aliquot of freeze dried samples was dissolved in 0.1
M NaOH and analysed with a water (alkali) SEC system and pullulan standards.
The molecular weight distribution curves are presented in Figure 15 and Figure
16.
Figure 15 details the detection of all components using a refractive index (RI)
detector. Most of the RI signal will reflect the concentration of carbohydrates,
because the amount of lignin is fairly small in the samples (Figure 13). The graph
shows two effects. First, as expected, the effect of CFF is shown by a shift
towards longer retention times for samples with lower molecular weight. Second,
for the 0-1 kDa sample, there is a higher signal at retention times of 21-24
minutes, which corresponds to a molecular weight of 2 kDa and lower, whereas
the >5 kDa and 1-5 kDa samples show a decrease in RI detection signal for the
same retention interval. For hot water extracts performed in similar conditions
Song et al. (2008) reported weight-average molecular weight of ~5 kDa, and for
TMP process liquids Persson et al. (2010) reported a weight-average molecular
weight of ~10 kDa.
Figure 16 details the absorbance spectra using a UV detector at a wavelength of
280 nm, which corresponds to aromatic species (i.e., lignin). From the Klason
lignin analyses in Figure 13, there is a small amount of lignin in all of the
samples, but in the 0-1 kDa sample, the amount of lignin is somewhat increased.
When analysing the data from the pullulan standard, the UV peaks had an
approximately 0.2 minute shorter retention time than the peaks from the RI
detector. Comparing the peaks from the UV and RI graphs, some peaks are
overlapping, which might suggest the presence of LCCs.
43
Results and Discussion
Figure 15. Water (alkali) SEC analysis with refractive index (RI) detector.
Figure 16. Water (alkali) SEC analysis with UV detector at 280 nm.
 Separation by chromatography of fractionated hot water extract
From the first separation by CFF, the 1-5 kDa liquid fraction with a relative high
yield and purity was selected as a sample for the second separation by
chromatography.
Adsorption chromatography was performed in two steps. First, the 1-5 kDa hot
water extract was pumped into a hydrophobic sorption column, resulting in two
samples/streams: permeate and material adsorbed on the column. Second, the
permeate from the first step was pumped into a second column with a higher
44
Results and Discussion
selectivity for aromatic structures, resulting in another two samples/streams:
permeate and material absorbed on the column.
The first hydrophobic column aimed to adsorb relatively pure lignin moieties
(here named column 1), while the second column aimed to separate LCCs from
GGM due to its very high selectivity for aromatic compounds (here named
column 2).
Table 11 shows that the isolated LCC fraction contained ~10% aromatics, the
upgraded GGM fraction ~1.5% aromatics, and the lignin fraction ~56%
aromatics. Based on the detected sugars, the calculated amount of GGM in the
upgraded GGM fraction was 79%. The calculated degree of detection, that is the
detectable sugars and lignin, was 80-85% in all fractions, and the shortfall was
most likely due to a combination of measuring errors. Excluding losses in
handling, the obtained products from CFF hot water extract are as follows, based
on internal relative values: lignin fraction (adsorbed to column 1) 7%, LCC
fraction (adsorbed to column 2) 5%, and GGM (permeate from column 2) 88%.
Table 11. Yield and amount of aromatics and galactoglucomannan (GGM).
Samples
Dissolved starting material
Cross-flow filtration:
1-5 kDa
Chromatography:
Adsorbed to column 1
Adsorbed to column 2
Permeate from column 2
45
Yield based on
initial amount of
saw dust
Aromatics
mg/g
144
wt%
-
49
5.5
1.8
1.4
23.7
55.7
10.2
1.5
Concluding remarks
5. CONCLUDING REMARKS
Process streams from the primary separation of wood components are obtained as
complex mixtures of components, which need to be separated and purified. Raw
materials with higher homogeneity can be obtained by cross-flow filtration (CFF).
For fractionation of kraft lignin, three topics have been examined: characteristics
of lignin fractions, extraction of low molecular weight lignin, and the behaviour
of re-dissolved lignin during CFF.
By filtering weak black liquor with membranes of different cut-offs, more
homogenous fractions of lignin could be obtained regarding size, characteristics,
and thermal properties. Through fractionation, it is possible to tailor lignin’s
molecular weight and consequently the amount of carbohydrates, amount of
phenolic hydroxyl groups, and thermal properties. Through dynamic mechanical
analysis (DMA), the degree of softening and flow can be monitored, and this
analysis might be a useful tool for predicting visco-elastic behavior during
thermal processing.
To improve the extraction of low molecular weight lignin (<1000 Da cut-off)
from black liquor, precipitation needs to be performed at lower temperatures than
with other fractions due to the formation of rigid, fairly large lignin
46
Concluding remarks
particles/agglomerates. However, lowering the pH at ambient temperature, results
in the formation of fine particles of precipitated lignin. As shown by differential
scanning calorimetry (DSC), the Tg of low molecular weight lignin could be as
low as 70°C or lower if irreversible cross-linking occurred during the first heat
scan. This could be an explanation for the low molecular weight behaviour during
precipitation at higher temperatures.
Due to the recent introduction of the LignoBoost process, which produces kraft
lignin of relatively high purity, there might be an interest to fractionate lignin
extracted by the LignoBoost process to obtain lignin fractions with more specific
properties for the production of high value application products. In this
investigation, the possibilities for improved production of low molecular weight
lignin were investigated. These results show (during the conditions of this set-up)
that lower concentrations of lignin in the feed result in a higher yield of low
molecular lignin in the permeate, relative to the feed concentration, and that
permeate flux could be increased by a higher pH and ionic strength.
For kraft pulp, a novel method for preparing nanocellulose was developed called
nanopulping, which appears to be energy efficient. The method is based on a
combination of the acceleration of chemical pulp fibres and steam explosion,
where the rapid pressure drop causes acceleration. The efficiency of the method
might be increased if the pulp is pretreated. To produce a more homogeneous size
distribution CFF, was used.
Hot water extract was used to demonstrate that a biorefinery process that
combines the steric separation of filtration with the selective separation of
adsorption chromatography can be used to isolate dissolved wood components in
47
Concluding remarks
upgraded fractionations, a process in which several products are obtained,
including polymeric lignin, LCCs and GGM.
The final separation of wood components using CFF has been successfully used
in this thesis to separate lignin fractions according to molar mass or particle size.
Furthermore, there is a possibility to combine CFF and adsorption
chromatography to obtain hemicellulose fractions rich in lignin or almost
completely free of lignin.
48
Notations
6. NOTATIONS
CFF
cf
cp
Da
ÐM
DMA
DSC
GGM
LCC
MFC
Mw
Mn
NMR
NP
ΔP
Q
R
SEC
Td
Tg
TEMPO
TGA
TMP
UV
v
Vif
Vp
Vr
VR
VRF
wt%
Cross flow filtration
Concentration in feed
Concentration in permeate
Dalton
Molecular-weight dispersity
Dynamic mechanical analysis
Differential scanning calorimetry
Galactoglucomannan
Lignin carbohydrate complex
Microfibrillated cellulose
Weight-average molecular weight
Number-average molecular weight
Nuclear magnetic resonance
Nanopulping
Transmembrane pressure (symbol)
Feed flow
Rejection coefficient
Size exclusion chromatography
Decomposition temperature
Glas transition temperature
2,2,6,6-tetramethylpiperidine-1-oxyl
Thermogravimetric analysis
Transmembrane pressure (abbreviation)
Ultraviolet
Cross-flow velocity
Initial feed volume
Permeate volume
Retentate volume
Volume reduction
Volume reduction factor
Weight percent
49
Acknowledgements
7. ACKNOWLEDGEMENTS
First of all, I want to express my deepest gratitude to Prof. Mikael E. Lindström for
accepting me as your PhD student. I have learned a lot, and it has been a wonderful time.
Thank you for your scientific support and guidance. You are always so positive, kind and
encouraging, and that has been a great help along the way and meant a lot to me.
To all my co-supervisors. I am very happy to have had so many carrying and kind cosupervisors who have supported me. Prof. Hans Theliander, I am sincerely grateful for
your ambitious supervision and great thoughtfulness. Prof. Gunnar Henriksson, I am very
grateful for your enthusiastic supervision and generous support. Ass. Prof. Martin
Lawoko, not only my supervisor but also a friend; thank you so much for the great support
and all the nice times running together. Dr. Liming Zhang, thank you for always having
the door open for nice discussions.
To Assoc. Prof. Ulrica Edlund, thank you for taking your time and the valuable inputs for
the quality check.
The Wallenberg Wood Science Center is gratefully acknowledged for financial support.
To all colleges and friends at FPT, you have been many during the years, thank you for a
great time, helpfulness and kindness. Sara and Anna, you have been my “rocks” and
“lights” along the way. Emma and Nicholas, thank you for being so dear and for all lunchlaughs. Linn and Michaela thank you for other memorable activates outside work. Shoaib,
Dimitri, Ramiro, Petri, and, Kasinee thank you for always being so helpful and caring. To
all my lovely colleagues, collaborates and friends during the years: Olena, Helena, Göran,
Susanne, Louise, Andreas, Farhan, Daniel, Dina, Yan, Ran, Emil, Malin, Jacob, Gisela,
Oruc, Lars W, Carl, Adam, Felix, Elisabeth, Monica, Lars Ö, Lars B, Caroline, Anders,
Stefan, Joby, Ngesa, Torbjörn, Per, Pontus, Rebecca, Linda, Jiebing, Dongfang, Yadong,
Yujia, Raquel and Myriam. Jan-Erik, Mia, Mona, Therese, Ewa, Inga, Kjell, Karin, Bosse
and Fredrick thank you for all your help; you are valuable.
To my dear colleagues in Gothenburg, you are all amazing: Tuve, Niklas, Susanne,
Kerstin and Hampus. Till mina väninnor, tack för att ni är så fina, förstående och härliga.
Till min familj. Jag är så tacksam och lycklig över att ha en sådan kärleksfull och
hjälpsam familj med underbara föräldrar, en fantastisk syster med familj, mina fina
morföräldrar och aunts & unlces med familjer, farföräldar och gammelmormor i hjärtat,
och pojkvän av 24k guld <3 Ni betyder allt för mig.
50
References
8.REFERENCES
Abe, K., Iwamoto, S., and Yano, H., (2007), Obtaining cellulose nanofibers with a uniform
width of 15 nm from wood. Biomacromolecules, 8, 3276-3278.
Alén, R., Patja, P., and Sjöström, E., (1979), Carbon dioxide precipitation of lignin from
pine kraft black liquor, Tappi Journal, 62, 108-110.
Alén, R., Sjöström, E., and Vaskikari, P., (1985), Carbon dioxide precipitation of lignin
from alkaline pulping liquors, Cellulose Chemistry and Technology 19, 537-541.
Alén, R., Sjöström, E., and Vaskikari, P., (1986), Ultrafiltration studies on alkaline pulping
liquors, Cellulose Chemistry and Technology, 20, 417-420.
Andersson, A., Persson, T., Zacchi, G., Stalbrand, H., and Jönsson, A.-S., (2007),
Comparison of diafiltration and size-exclusion chromatography to recover hemicelluloses from
process water from thermomechanical pulping of spruce, Applied Biochemistry and
Biotechnology, 137, 971-983.
Ankerfors, M., (2012), Microfibrillated cellulose: energy-efficient preparation techniques,
Licentiate Thesis, Royal Institute of Technology, KTH, Stockholm, Sweden.
Antonsson, S., Henriksson, G., Johansson, M., and Lindström, M. E., (2008), Low Mwlignin fractions together with vegetable oils as available oligomers for novel paper-coating
applications as hydrophobic barrier, Industrial Crops and Products, 27, 98-103.
Balakshin, M., Capenema, E., Gracz, H., Chang, H. M., and Jameel, H., (2011),
Quantification of lignin-carbohydrate linkages with high-resolution NMR spectroscopy
Planta, 233, 1097-1110.
Berg, G. B., and Smolders, C. A., (1990), Flux decline in ultrafiltration processes,
Desalination, 77, 101-133.
Bowen, W. R., and Jenner, F., (1995), Theoretical descriptions of membrane filtration of
colloids and fine particles: an assessment and review, Advances in Colloid and Interface
Science, 56, 141-200.
51
References
Brasch, D. J., and Free, K. W., (1965), Prehydrolysis-kraft pulping of pinus radiata grown in
New Zealand, Tappi, 48, 245-248.
Braun, J. L., Holtman, K. M., and Kadla, J. F., (2005), Lignin-based carbon fibers:
Oxidative thermostabilization of kraft lignin, Carbon, 43, 385-394.
Brodin, I., Sjöholm, E., and Gellerstedt, G., (2009), Kraft lignin as feedstock for chemical
products: The effects of membrane filtration, Holzforschung 63, 290-297.
Chakraborty, A., Sain, M., and Kortschot, M., (2005), Cellulose microfibrils: A novel
method of preparation using high shear refining and cryocrushing, Holzforschung, 59, 102107.
Chen, W., Yu, H., Lui, Y., Chen, P., Zhang, M., and Hai, Y., (2011), Individualization of
cellulose nanofibres from wood using high-intensity ultrasonication combined with chemical
pretreatments, Carbohydrate Polymers, 83, 1804-1811.
Chowdhury, S., Gao, G., and Kadla, J.F., (2012), Investigating the viscoelasticity of ligninpolyacrylonitrile blends using dynamic rheology, Proceedings, 243rd ACS meeting, March 2529, San Diego, California, USA, Abstract: CELL 119.
Cui, C., Sadeghifar, H., Sen, S., and Argyropoulos, D.S., (2013), Toward thermoplastic
lignin polymers; Part II, BioResources, 8, 864-886.
Curtolo, F., and Eksteen, N. C., (2006), Method and apparatus for manufacturing
microfibrillated cellulose fiber, US Patent 0006189.
Dallmeyer I, Chowdhury S, and Kadla J.F., (2013), Preparation and characterization of kraft
lignin-based moisture-responsive films with reversible shape-change capability,
Biomacromolecules, 14, 2354-2363.
Dence, C. W., (1992), The determination of lignin, Methods in Lignin Chemistry, S. Y. Lin
and C. Dence (Eds.), Springer-Verlag, Heidelberg, Germany, 33-61.
Dong, D., and Fricke, A. L., (1995), Effects of multiple pulping variables on the molecular
weight and molecular weight distribution of kraft lignin, Journal of Wood Chemistry and
Technology 15, 369-393.
Drouin, M. P., and Desrochers, M. J., (1988), Isolation of lignin from spent kraft liquor by
hyper and ultrafiltration, Chemical engineering technology in forest products processing,
AIChE forest products division, 2, 58-66.
Eichhorn, S. J., Dufresne, A., Aranguren, M., Marcovich, N. E., Capadona, J. R, Rowan,
S. J., Weder, C., Thielemans, W., Roman, M., Reneckar, S., Gindl, W., Veigel, S., Keckes,
J., Yano, H., Abe, K., Nogi, M., Nakagaito, A. N., Mangalam, A., Simonsen, J., Benight,
A. S., Bismarck, A., Berglund, L. A., and Peijs, T., (2010), Review: current international
research in cellulose nanofibres and nanocomposites, Journal of Material Science, 45, 1-33.
52
References
Elegir, G., Bussini, D., Antonsson, S., Lindström, M. E., and Zoia, L., (2007), Laccaseinitiated cross-linking of lignocellulose fibres using a ultra-filtered lignin isolated from kraft
black liquor, Applied Microbiology Biotechnology 77, 809-817.
Engström, A.-C., Ek, M., and Henriksson G., (2006), Improved accessibility and reactivity
of dissolving pulp for the viscose process – pre-treatment with monocomponent
endoglucanase, Biomacromolecules, 7, 2027-2031.
Forss, K., and Fuhrmann, A., (1976), Karatex: The lignin-based adhesive for plywood,
particle board and fibre board, Paperi ja Puu, 58, 817-824.
Gellerstedt, G., (1992), Gel permeation chromatography, Methods in Lignin Chemistry, S. Y.
Lin and C. Dence (Eds.), Springer-Verlag, Heidelberg, Germany, 487-497.
Gierer, J., (1980), Chemical aspects of kraft pulping, Wood Science Technology, 14, 241-266.
Grace, T. M., (1989), Black liquor evaporation, Pulp and paper manufacture - Alkaline
pulping, Joint textbook committee of the Paper Industry, 477-530.
Granata, A., and Argyropolous, D.S.J., (1995), 1, 2-Chloro-4,4,5,5-tetramethyl-1,3,2dioxaphospholane, a reagent for the accurate determination of the uncondensed and condensed
phenolic moieties in lignins, Journal of Agricultural and Food Chemistry, 43, 1538-1544.
Griggs, B. F., Gratzl, J. S., Chen, C.-L., and Venica, A., (1985), Chemical characterization
of kraft lignin and kraft lignin products, International Symposium on Wood and Pulping
Chemistry, August 26-30, Vancouver, Canada, 5-6.
Guerra, A., Filpponen, I., Lucia, L. A., and Argyropolous, D.S.J., (2006), Comparative
evaluation of three lignin isolation protocols for various wood species, Journal of Agricultural
and Food Chemistry, 54, 9696-9705.
Guido, N., Mija, A., Vincent, L., and Sbirrazzuoli, N., (2009), Molecular mobility and
relaxation process of isolated lignin studied by multifrequency calorimetric experiments,
Physical Chemistry Chemical Physics, 11, 1227-1236.
Hartman, J., (2006), Hemicellulose as a barrier material, Licentiate Thesis, Royal Institute of
Technology, KTH, Stockholm, Sweden.
Hartman, J., Albertsson, A. C., Söderqvist Lindblad, M., and Sjöberg, J., (2006), Oxygen
barrier materials from renewable sources: Material properties of softwood hemicellulose-based
films, Journal of Applied Polymer Science, 100, 2985-2991.
Henriksson, M., Henriksson, G., Berglund, L. A., and Lindström, T., (2007), An
environmentally friendly method for enzyme-assisted preparation of microfibrillated cellulose
(MFC) nanofibers, European Polymer Journal, 43, 3434-3441.
53
References
Herrick, F. W., Casedier, R. L., Hamilton, J. K., and Sandberg, K. R., (1983),
Microfibrillated cellulose: morphology and accessibility. Journal of Applied Polymer Science:
Applied Polymer Symposium, 37, 797-813.
Hough, W. J., (1910), Process of treating spent alkaline pulping liquors, United States Patent
949,324.
Ishikawa, H., Ide, S., (1993), Method of producing finely divided fibrous cellulose particles.
US Patent 5,269,470.
Iwamoto, S., Nakagaito, A. N., Yano, H., and Nogi, M., (2005), Optically transparent
composites reinforced with plant fiber-based nanofibers, Applied Physics A, 81, 1109-112.
Jönsson, A.-S., Wallberg, O., (2009), Cost estimates of kraft lignin recovery by ultrafiltration,
Desalination, 237, 254-267.
Kadla, J.F., Kubo, S., Venditti, R.A., Gilbert, R.D., Compere, A.L., and Griffith, W.
(2002), Lignin-based carbon fibers for composite fiber applications, Carbon, 40, 2913-2920.
Keyoumu, A., Sjödahl, R., Henriksson, G., Ek, M., Gellerstedt, G., and Lindström, M. E.,
(2004), Continuous nano- and ultra-filtration of kraft pulping black liquor with ceramic filters,
A method for lowering the load on the recovery boiler while generating valuable side-products,
Industrial Crops and Products, 20, 143-150.
Klemm, D., Kramer, F., Moritz, S., Lindström, T., Ankerfors, M., Gray, D., and Dorris,
A., (2011), Nanocelluloses: A new family of nature-based materials, AngewanteChemie, 50,
5438-5466.
Kubo, S., and Kadla, J. F., (2005), Kraft Lignin/poly(ethylene oxide) blends: Effect of lignin
structure on miscibility and hydrogen bonding, Applied Polymer Science, 98, 1437-1444.
Lake, M. A., and Blackburn, J. C., (2011), Process for recovering lignin, US Patent
Application US20110294991 A1.
Lawoko, M., Henriksson, G., and Gellerstedt, G., (2005), Structural differences between the
lignin-carbohydrate complexes present in wood and in chemical pulps, Biomacromolecules, 6,
3467-3473.
Lin, S. Y., (1992), Commercial spent pulping liquors, Methods in Lignin Chemistry, S. Y. Lin
and C. Dence (Eds.), Springer-Verlag, Heidelberg, Germany, 75-80.
Lin, S. Y., and Detroit, W., (1981), In chemical heterogeneity of technical lignins: Its
significance in lignin utilization, Proceedings of the 1st International Symposium on Wood and
Pulping Chemistry, Stockholm, Sweden, June 9-12, 1981, 44-52, SPCI (Svenska Cellulosaoch Pappers Ingenjörsföreningen), Stockholm.
Lindberg, J. J., Tylli, H., and Majani, C., (1964), Notes on the molecular weight and the
fraction of lignins with organic solvents, Paperi ja Puu, 46, 521-526.
54
References
Lindblad, M. S., Ranucci, E., and Albertsson, A. C., (2001), Biodegradable polymers from
renewable sources. New hemicellulose-based hydrogels, Macromolecular Rapid
Communications, 22, 962-967.
Madani, A., Kiiskinen, H., Olson, J. A., and Martinez, D. M., (2011), Fractionation of
microfibrillated cellulose and its effects on tensile index and elongation of paper, Nordic Pulp
and Paper Research Journal, 26, 306-311.
Merewether, J.W.T., (1962), The precipitation of lignin from eucalyptus kraft black liquors,
Tappi Journal 45, 159-163.
Moosavifar, A., Sedin, P., and Theliander, H., (2006), Viscosity and boiling point elevation
of black liquor: Consequences when lignin is extracted from the black liquor, Nordic Pulp and
Paper Research Journal, 21, 180-187.
Mörck, R., Kringstad, K., and Månsson, P., (1982), Structural heterogeneity in softwood
kraft lignin as studied by GPC, 13C-NMR, 1H-NMR, Canadian Wood Chemistry Symposium,
September 13-15, Niagara Falls, Ontario, Canada, 161-168.
Mörck, R., Yoshida, H., and Kringstad, K. P., (1986), Fractionation of kraft lignin by
successive extraction with organic solvents, functional groups 13C-NMR-spectra and
molecular weight distributions, Holzforschung, 40(supplement issue), 51-60.
Nagy, M., Kosa, M., Theliander, H., and Ragauskas, A. J., (2010), Characterization of CO2
precipitated kraft lignin to promote its utilization, Green Chemistry, 12, 31-34.
Nakagaito, A. N., Yano, H., (2004), The effect of morphological changes from pulp fiber
towards nano-scale fibrillated cellulose on the mechanical properties of high-strength plant
fiber based composites, Applied Physics, A78, 547-552.
Niemi, H., Lahti, J., Hatakka, H., Kärki, S., Rovio, S., Kallioinen, M., Mänttäri, M., and
Louhi-Kultanen, M., (2011), Fractionation of organic and inorganic compounds from black
liquor by combining membrane separation and crystallization, Chemical Engineering
Technology 34, 593-598.
Norberg, I., (2012), Carbon fibres from kraft lignin, Ph.D. Thesis, Royal Institute of
Technology, KTH, Stockholm, Sweden.
Nordin, A-K., (2008), Process design for ultrafiltration of complex process streams, Ph.D.
Thesis, Lund University, Sweden.
Nordström, Y., Norberg, I., Sjöholm, E., and Drougge, R. (2013), A new softening agent
for melt spinning of softwood kraft lignin, Journal of Applied Polymer Science, 129, 12741279.
Norgren, M., and Lindström, B., (2000), Dissociation of phenolic groups in kraft lignin at
elevated temperatures, Holzforschung, 54, 519-527.
55
References
Passinen, K., (1968), Chemical composition of spent liquors, Proceedings of IUPAC
EUCEPA Symposium: Recovery of Pulping Chemicals, Helsinki, Finland, 183-209.
Persson, T., Krawczyk, H., Nordin, A.-K., and Jönsson, A.-S., (2010), Fractionation of
process water in thermomechanical pulp mills, Bioresource Technology, 101, 3884-3892.
Pollak, A. Keilen, J. J., Jr., and Drum, L. F., (1949), Method of producing lignin from black
liquor, United States Patent 2,464,828.
Pouteau, C., Dole, P., Cathala, B., Averous, L., and Boquillon, N., (2003), Antioxidant
properties of lignin in polypropylene, Polymer Degradation and Stability, 81, 9-18.
Pääkkö, M., Ankerfors, M., Kosonen, H., Nykänen, A., Ahola, S., Österberg, M.,
Ruokolainen, J., Laine, J., Larsson, P. T., Ikkala, O., and Lindström, T., (2007),
Enzymatic hydrolysis combined with mechanical shearing and high-pressure homogenization
for nanoscale cellulose fibrils and strong gels, Biomacromolecules, 8, 1934-1941.
Ragauskas, A. J., Williams, C. K., Davison, B. H., Britovsek, G., Cairney, J., Eckert, C.
A., Frederick Jr., W. J., Hallett, J. P., Leak, D. J., Liotta, C. L., Mielenz, J. R., Murphy,
R., Templer, and R.,Tschaplinski, T., (2006), The path forward for biofuels and
biomaterials, Science, 311, 484-489.
Ragnar, M., Lindgren, C. T., and Nilvebrant, N. O., (2000), pKa-values of guaiacyl and
syringyl phenols related to lignin, Journal of Wood Chemistry and Technology, 20, 277-305.
Reboulet, H. J., (1941), Molding material and method of producing the same, United States
Patent 2,228,976.
Rinman, E. L., (1911), Process of treating waste liquors from soda pulp mills, United States
Patent 1,005,882.
Ripperger, S., Gösele, W., and Alt, C., (2009), Filtration, 1. Fundamentals, Ullmann's
Encyclopedia of Industrial Chemistry, Wiley-VCH Verlag, Weinheim.
Rojas, O. J., Song, J., Argyropoulos, D. S., and Bullón, J., (2006), Lignin separation from
kraft black liquors by tangential ultrafiltration, La Chimica e I´Industria, 88-95.
Saito, T., Nishiyama, Y., Putaux, J-L, Vignon, M., and Isogai, A., (2006), Homogeneous
suspensions of individualized microfibrils from TEMPO-catalyzed oxidation of native
cellulose, Biomacromolecules, 7, 1687-1691.
Sannigrahi, P., Ho Kim, D., Jung, S., and Ragauskas, A., (2011), Pseudo-lignin and
pretreatment chemistry, Energy and Environmental Science, 4, 1306-1310.
Scott, C. W., (1940), Process of producing molding composition from lignin and
formaldehyde, United States Patent 2,201,797.
56
References
Siró, I., and Plackett, D., (2010), Microfibrillated cellulose and new nanocomposite
materials: a review, Cellulose, 17, 459-494.
Skogsindustrin, (2012), En faktasamling 2012 års branschstatistik, Skogsindustrierna,
Stockholm, Sweden.
Song, T., Pranovich, A., Sumerskiy, I., and Holmbom, B., (2008), Extraction of
galactoglucomannan from spruce wood with pressurised hot water, Holzforschung, 62, 659666.
Stewart, D., (2008), Lignin as a base material for materials applications: Chemistry,
application and economics, Industrial Crops and Products, 27, 202-207.
Sundin, J., (2000), Precipitation of kraft lignin under alkaline conditions, Ph.D. Thesis, Royal
Institute of Technology, KTH, Stockholm, Sweden.
Taniguchi, T., and Okamura, K., (1998), New films produced from microfibrillated natural
fibres, Polymer International, 47, 291-294.
Teleman, A., (2009), Hemicelluloses and pectins, Pulp and paper chemistry, Wood chemistry
and wood biotechnology 1, Ek, M., Gellerstedt, G., Henriksson, G., (Eds.), Walter de Gryter
GmbH & Co, Berlin.
Tessié du Motay, C. M., (1872), Improvement in recovering waste alkalies used in treating
paper-pulp, United States Patent 132,452.
Theliander, H., (2009), Recover of cooking chemicals: the treatment and burning of black
liquor, Pulp and paper chemistry, Pulping chemistry and technology 2, 1 st edition, Ek, M.,
Gellerstedt, G., Henriksson, G., (Eds.), Walter de Gryter GmbH & Co, Berlin.
Thornton, J., Ekman, R., Holmbom, B., and Örså, F., (1994), Polysaccharides dissolved
from Norway spruce in thermomechanical pulping and peroxide bleaching, Journal of Wood
Chemistry and Technology, 14, 159-175.
Toledano, A., García, A., Mondragon, I., and Labidi, L., (2010), Lignin separation and
fractionation by ultrafiltration, Separation and Purification Technology, 71, 38-43.
Tomani, P., (2013), Update on LignoBoost lignin and applications, SPCI Convention
(Svenska Cellulosa- och Pappers Ingenjörsföreningen), September 25-26, Stockholm, Sweden.
Tomlinson, G. H., and Tomlinson Jr., G. H., (1946), Method of treating lignocellulosic
material, United States Patent 2,406,867.
Tomlinson, G. H., and Tomlinson Jr., G. H., (1948), Method of treating lignocellulosic
material, Canadian Patent CA 448,476.
Tsuguyuki, S., Masayuki, H., Naoyuki, T., Satoshi, K., Hayaka, F., Laurent, H., and
Akira, I., (2009), Individualization of nano-sized plant cellulose fibrils by direct surface
57
References
carboxylation using TEMPO catalyst under neutral conditions, Biomacromolecules, 7, 19921996.
Turbak, A. F., Snyder, F.W., and Sandberg, K. R., (1982), Food products containing
microfibrillated cellulose, U.S. Patent 4,341,807.
Turbak, A. F.; Snyder, F. W., and Sandberg, K. R., (1983), Microfibrillated cellulose: A
new cellulose product: properties, uses, and commercial potential, Journal of Applied Polymer
Science: Applied Polymer Symposium, 37, 815-827.
Turbak, A. F., Snyder, F.W., and Sandberg, K. R., (1983), Microfibrillated cellulose, U.S.
Patent 4,374,702.
Turbak, A. F., Snyder, F. W., and Sandberg, K. R., (1983), Suspensions containing
microfibrillated cellulose, U.S. Patent 4,378,381.
Uloth, V. C., and Wearing, J. T., (1989), Kraft lignin recovery: Acid precipitation versus
ultrafiltration, Part I: Laboratory test results, Pulp and Paper Canada, 90, 67-71.
Uraki Y, Sugiyama Y, Koda K, Kubo S, Kishimoto T, and Kadla J. F., (2012), Thermal
mobility of β-O-4-type artificial lignin, Biomacromolecules, 13, 867-872.
Vishtal, A., Kraslawski, A., (2011), Challenges in industrial application of technical lignins,
BioResources, 6, 3547-3568.
Wada, S., Iwamida, T., Iizima, R., and Yabe, K., (1962), Fractional precipitation of kraft
black liquor from spruce wood, Chemistry of High Polymers (Kabunshi Kagaku), 19, 699-703.
Wallberg, O. and Jönsson, A.-S., (2003), Influence of the membrane cut-off during
ultrafiltration of kraft black liquor with ceramic membranes, Chemical Engineering Research
and Design, 81(A), 1379-1384.
Wallberg, O., Jönsson, A.-S., and Wimmerstedt, R., (2003), Fractionation and concentration
of kraft black liquor lignin with ultrafiltration, Desalination, 154, 187-199.
Wallberg, O., Jönsson, A.-S., and Wimmerstedt, R., (2003), Ultrafiltration of kraft black
liquor with a ceramic membranes, Desalination, 156, 145-153.
Wallberg, O., Holmqvist, A., and Jönsson, A.-S., (2005), Ultrafiltration of kraft cooking
liquors from a continuous cooking process, Desalination, 180, 109-118.
Wallmo, H., and Theliander, H., (2009), The influence of hemicelluloses during the
precipitation of lignin in kraft black liquor, Nordic Pulp and Paper Research Journal, 24, 165171.
Weber, M., and Weber, M., (2010), Phenolic resins: A century of progress, Louis, P. (Ed.),
Springer-Verlag Berlin Heidelberg, 9-23.
58
References
Willför, S., Sjöholm, R., Laine, C., Roslund, M., Hemming, J., and Holmbom, B, (2003),
Characterisation of water-soluble galactoglucomannans from Norway spruce wood and
thermomechanical pulp, Carbohydrate Polymers, 52, 175-187.
Willför, S., Rehn, P., Sundberg, A., Sundberg, K., and Holmbom, B., (2003), Recovery of
water-soluble acetylgalactoglucomannans from mechanical pulp of spruce, Tappi Journal, 2,
27-32.
Willför, S., Sundberg, K., Tenkanen, M., and Holmbom, B., (2008), Spruce-derived
mannans - A potential raw material for hydrocolloids and novel advanced natural materials,
Carbohydrate Polymers, 72, 197-210.
Woerner, D. L., and McCarthy, J. L., (1984), Ultrafiltration of kraft black liquor, American
Institute of Chemical Engineers, AIChE Symposium Series 80, 25-33.
Woerner, D. L., and McCarthy, J. L., (1987), Ultrafiltration of pulp mill liquors, Tappi
Journal, 70, 126-129.
Yoshida, H., Mörck, R., Kringstad, K. P., and Hatakeyama, H., (1987), Fractionation of
kraft lignin by successive extraction with organic solvents, II Thermal properties of kraft lignin
fractions, Holzforschung, 41, 171-176.
Yoshida, H., Mörck, R., Kringstad, K.P., and Hatakeyama, H. J., (1990), Kraft lignin in
polyurethanes. II. Effects of the molecular weight of kraft lignin on the properties of
polyurethanes from a kraft lignin–polyether triol –polymeric MDI system, Journal of Applied
Polymer Science, 40, 1819-1832.
Zhang, L., and Gellerstedt, G., (2001), Lignin fractionation and quantitative structural
analysis of the lignin fractions with 13C and HSQC NMR techniques, Proceedings, Postsymposium Workshop of the 11th International Symposium on Wood and Pulping Chemistry,
3:595, Grenoble, France,
Zhao, H. P., Feng, X. Q., and Gao, H., (2007), Ultrasonic technique for extracting nanofibres
from nature materials, Applied Physics Letters, 90, 073112-1.
Zimmermann, T., Pöhler, E., Geiger, T., (2004), Cellulose fibrils for polymer reinforcement,
Advanced Engineering Materials, 6, 754-761.
Öhman, F., and Theiander, H., (2006), Washing lignin precipitated from kraft black liquor,
Paperi ja Puu 88, 287-292.
Öhman, F., Wallmo, H., and Theliander, H., (2007), A novel method for washing lignin
precipitated from kraft black liquor - Laboratory trials, Nordic Pulp and Paper Research
Journal 22, 9-16.
Örså, F., Holmbom, B., and Thornton, J., (1997), Dissolution and dispersion of spruce wood
components into hot water, Wood Science and Technology, 31, 279-290.
59