chemical modification of cellulose fibres and fibrils...

CHEMICAL MODIFICATION OF CELLULOSE

FIBRES AND FIBRILS FOR DESIGN OF NEW

MATERIALS

Verónica López Durán

Doctoral Thesis

KTH Royal Institute of Technology

School of Engineering Sciences in Chemistry, Biotechnology and Health

Department of Fibre and Polymer Technology

Stockholm, 2018

ii

Supervisors

Professor Lars Wågberg and Dr Per Larsson

Copyright © Verónica López Durán All rights reserved Paper I © 2016, Cellulose Paper II © 2018, Carbohydrate Polymers Paper III © 2018, Manuscript Paper IV © 2017, Cellulose Paper V © 2018, Manuscript

TRITA-CHE Report 2018:1 ISSN 1654-1081 ISBN 978-91-7729-670-6

AKADEMISK AVHANDLING som med tillstånd av Kungliga Tekniska högskolan i Stockholm framlägges till offentlig granskning för avläggande av

teknologie doktorsexamen 23 februari, kl. 10.00 i F3, Lindstedtsvägen 26, KTH, Stockholm Avhandlingen försvaras på engelska. Fakultetsopponent: Professor

Akira Isogai, The University of Tokyo.

iii

To my family

iv

ABSTRACT

Due to the growing interest in biobased materials in today’s society, where the

need for a cyclic economy is obvious, there has been a huge increase in the

interest for using cellulose due to its excellent mechanical and chemical

properties. However, the properties of cellulose have to be modified and

improved in order to satisfy advanced material applications where the cellulose

properties can be tuned to fit the properties of other components in composite

mixtures. This thesis explores the heterogeneous chemical modification of

cellulose for improved material properties of cellulose-based materials and the

use of cellulose fibres and fibrils in novel applications.

In the first part of the work described in this thesis, a fundamental study was

performed to clarify how the chemical composition and the fibre/fibril structure

of the cellulose following chemical modification affect the material properties.

The second part of the work was aimed at exploring the potential for using the

chemically modified fibres/fibrils in novel material applications.

Lignocellulosic fibres with different chemical compositions were modified by

periodate oxidation and borohydride reduction, and it was found that the most

important factor was the amount of holocellulose present in the fibres, since

lignin-rich fibres were less reactive and less responsive to the treatments. Despite

the lower reactivity of lignin-rich fibres, it was however possible to improve

their mechanical properties and to achieve a significant increase in the

compressive strength of papers prepared from modified unbleached kraft fibres.

The chemical modifications were then expanded to nine different molecular

structures and two different degrees of modification. Fibres modified at low

degrees of modification were used to prepare handsheets, followed by

mechanical and physical characterization. Highly modified fibres were also used

to prepare cellulose nanofibrils (CNFs). It was found that the properties of

handsheets and films prepared from modified fibres/fibrils are highly dependent

on the chemical structure of the modified cellulose and, as an example, the

v

ductility was greatly improved by converting secondary alcohols to primary

alcohols. A detailed analysis of the modified fibres and fibrils also showed that,

due to the heterogeneous chemical reaction used, the modified fibrils had a core-

shell structure with a shell of modified cellulose with a lower crystalline order

surrounding a core of crystalline cellulose. The results also showed that the

chemical structure of the modified shell dramatically affects the interaction with

moisture. Materials from fibrils containing covalent crosslinks have shown to be

less sensitive to moisture at the cost of being more brittle.

In a different application, modified CNFs were used as paper strength additives.

Three differently modified CNFs were used: carboxymethylated CNFs,

periodate-oxidised carboxymethylated CNFs and dopamine-grafted

carboxymethylated CNFs. The properties of these CNFs were compared with

that of a microfibrillated cellulose from unbleached kraft fibres. In general, a

great improvement in the dry mechanical properties of handsheets was observed

with the addition of the periodate-oxidised oxidised and dopamine-grafted

modified fibrils, whereas only the periodate-oxidised carboxymethylated CNFs

improved the wet strength.

Finally, it was found that the chemically modified fibres could be used to prepare

a novel low-density material with good mechanical strength, both wet and dry,

and excellent shape recovery capacity in the wet state after mechanical

compression. The fibre networks were produced by solvent exchange from water

to acetone followed by air drying at room temperature. The properties of the

fibre networks could also fairly easily be tuned in terms of porosity, density and

strength.

vi

SAMMANFATTNING

I och med ett växande samhällsintresse för biobaserade material, där behovet av

en så kallad cirkulär ekonomi är uppenbar, har blickarna riktats mot att använda

cellulosa i nya tillämpningsområden. Detta mycket på grund av cellulosans

utmärkta mekaniska och kemiska egenskaper. Egenskaperna måste emellertid

ofta modifieras och anpassas en aning för att uppfylla kraven för dagens

avancerade materialapplikationer. Denna avhandling undersöker heterogen

kemisk modifieringen av cellulosa för att förbättrade materialegenskaper hos

cellulosabaserade material, samt användningen av cellulosafibrer och cellulosa

nanofibriller (CNF:er) i nya applikationer.

Den första delen av denna avhandling sammanfattar en grundläggande studie för

att klargöra hur den kemiska sammansättningen och strukturen hos

cellulosabaserade fibrer och fibriller, som följd av kemisk modifiering, påverkar

materialegenskaperna. Den andra delen av avhandlingen syftar sedan till att visa

potentialen hos de kemiskt modifierade fibrerna/fibrillerna i nya

materialapplikationer.

Först modifierades lignocellulosafibrer av olika kemisk sammansättning genom

periodatoxidation och borhydridreduktion, där det visades att den viktigaste

faktorn för modifiering var holocellulosamängden i fibrerna och således var

ligninrika fibrer mindre reaktiva och mottagliga för denna typ av kemiska

behandlingar. Trots den lägre reaktiviteten hos ligninrika fibrer var det möjligt

att förbättra även deras mekaniska egenskaper, där bland annat en signifikant

ökning i kompressionsstyrka (så kallad SCT) hos papper framställda från

modifierade oblekta kraftfibrerna uppmättes.

De kemiska modifieringarna expanderades sedan till nio olika molekylstrukturer

och två olika grader av modifiering. Fibrer modifierade till låga

modifieringsgrader användes för att tillverka pappersark, följt

materialkarakterisering. Fibrer med hög modifieringsgrad användes för att

framställa CNF:er. Det visade sig att egenskaperna hos pappersark och filmer

vii

framställda från modifierade fibrer/fibriller är starkt beroende av den kemiska

strukturen hos den modifierade cellulosan och som ett exempel förbättrades

töjbarheten avsevärt genom omvandling av sekundära alkoholer till primära

alkoholer. En detaljerad analys av de modifierade fibrerna och fibrillerna visade

vidare att, i och med den använda heterogena kemiska reaktionen, hade de

modifierade fibrillerna en ”core-shell”-struktur med ett skal av amorf modifierad

cellulosa som omger en mer intakt kärna av kristallin cellulosa. Resultaten

visade också att den kemiska strukturen hos det modifierade skalet dramatiskt

påverkar interaktionen med fukt och material från fibriller innehållande

kovalenta tvärbindningar visade sig vara mindre fuktkänsliga, men på bekostnad

av att vara sprödare.

I en annan applikation användes modifierade CNF:er som tillsatsmedel under

papperstillverkning för att förbättra pappersstyrkan. Tre olika modifierade

CNF:er användes: karboxymetylerade CNF:er, periodatoxiderade och

karboxymetylerade CNF:er och dopamin-ympade och karboxymetylerade

CNF:er. Effekterna av tillsats av dessa CNF:er jämfördes också med effekterna

av en mikrofibrillerad cellulosa från oblekta kraftfibrer. I allmänhet observerades

en signifikant förbättring av de mekaniska egenskaperna i torrt tillstånd vid

tillsats av periodatoxiderade och dopaminympade karboxymetylerade CNF:er,

medan periodatoxiderade karboximetylerade CNF även förbättrade våtstyrkan,

speciellt i kombination med en amininnehållande polyelektrolyt.

Slutligen kunde de kemiskt modifierade fibrerna användas för att framställa ett

nytt material med låg densitet och god mekanisk hållfasthet, både i vått och torrt

tillstånd, och utmärkt formåterhämtningskapacitet i det våta tillståndet efter

mekanisk deformation. Fibernätverket framställdes genom lösningsmedelsbyte

från vatten till aceton följt av lufttorkning vid rumstemperatur. Fibernätets

egenskaper kan också ganska lätt anpassas med avseende på porositet, densitet

och styrka.

viii

LIST OF PUBLICATIONS

I. On the relationship between fibre composition and material properties following periodate oxidation and borohydride reduction of lignocellulosic fibres. Verónica López Durán, Per A. Larsson, Lars Wågberg Cellulose (2016), 23, 3495–3510

II. Chemical modification of cellulose-rich fibres to clarify the influence

of the chemical structure on the physical and mechanical properties of cellulose fibres and thereof made sheets. Verónica López Durán, Per A. Larsson, Lars Wågberg Carbohydrate Polymers (2018), 182, 1–7

III. Effect of chemical functionality on the mechanical and barrier

performance of all-cellulose composites. Verónica López Durán, Johannes Hellwig, P. Tomas Larsson, Lars Wågberg, Per A. Larsson Manuscript

IV. Chemically modified cellulose micro- and nanofibrils as paper-strength additives. Rebecca Hollertz, Verónica López Durán, Per A. Larsson, Lars Wågberg Cellulose (2017), 24, 3883–3899

V. Novel cellulose-based light weight, wet resilient materials with

tunable porosity, density and strength. Verónica López Durán, Johan Erlandsson, Lars Wågberg, Per A. Larsson

Manuscript

ix

CONTRIBUTIONS TO THE PAPERS

The author’s contributions to the appended papers are as follows:

I. All the experimental work and preparation of the manuscript

II. All the experimental work and preparation of the manuscript

III. Most of the experimental work and preparation of the manuscript

IV. Part of the experimental work, and part of the manuscript preparation

V. Most of the experimental work and most of the manuscript preparation

Papers not included in this thesis:

a. Macro- and mesoporous nanocellulose beads for use in energy storage devices Johan Erlandsson, Verónica López Durán, Hjalmar Granberg, Mats Sandberg, Per A. Larsson, Lars Wågberg, Applied Materials Today 5 (2016) 246–254

b. Development of Hybrid Coatings to Reduce Flammability of Cellulose Fiber Network via Layer-by-Layer Assembly Oruç Köklükaya, Federico Carosio, Verónica López Durán, Lars Wågberg Manuscript

x

RELATED MATERIAL

In addition to the papers, the work resulted in the following conference

presentations:

a. On the relationship between chemical structure and mechanical

flexibility of cellulose and its derivatives

Verónica López Durán, Per A. Larsson, Lars Wågberg

9th International Paper and Coating Chemistry Symposium, Tokyo,

Japan (2015)

b. Effect of lignin on the material properties of chemically modified fibres

by periodate oxidation and borohydride reduction


American Chemical Society national meeting, San Diego, USA (2016)

c. New insights on the effects of chemical structure on the properties of

cellulose nanofibrils: Characterization, mechanical performance and

barrier properties


American Chemical Society national meeting, San Francisco, USA

(2017)

xi

LIST OF ABBREVIATIONS

DMA Dynamic mechanical analysis

DMAc Dimethylacetamide

DVS Dynamic vapour sorption

EDC 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide

FE-SEM Field emission scanning electron microscope

CNF Cellulose nanofibril

CTMP High-temperature chemothermochemical pulp

MFC Microfibrillated cellulose

PAA Polyacrylic acid

pDADMAC Poly(dimethyldiallylammonium chloride)

PTFE Polytetrafluoroethylene

PVAm Polyvinylamine

RH Relative humidity

RT Room temperature

SCT Short-span compression test

SEM Scanning electron microscopy

TEMPO 4-acetamido-2,2,6,6-tetramethylpiperidine 1-oxyl

WRV Water retention value

xii

TABLE OF CONTENTS

1. INTRODUCTION ....................................................................................... 1

1.1. Context ................................................................................................ 1

1.2. Purpose of the study ............................................................................ 1

2. BACKGROUND ......................................................................................... 3

2.1. Hierarchical structure of cellulose in wood ......................................... 3

2.1. All-cellulose composites: .................................................................... 5

2.2. Oxidative chemical modification of cellulose ..................................... 8

2.2.1. Periodate oxidation ..................................................................... 8

2.2.2. Reduction of aldehydes using sodium borohydride .................... 9

2.2.3. Oxidation of aldehydes using sodium chlorite ............................ 9

2.2.4. Reductive amination of aldehydes ............................................ 10

2.2.5. TEMPO-mediated oxidation ..................................................... 11

2.3. Applications of chemically modified cellulose ................................. 12

2.3.1. Paper-based materials ............................................................... 12

3. EXPERIMENTAL .................................................................................... 13

3.1. Materials ............................................................................................ 13

3.1.1. Fibres and nanofibrils ............................................................... 13

3.1.2. Chemicals ................................................................................. 13

3.2. Methods ............................................................................................. 14

3.2.1. Chemical modifications ............................................................ 14

3.2.2. Characterization of fibres and fibrils ........................................ 17

4. RESULTS AND DISCUSSION ................................................................ 23

xiii

4.1. Influence of fibre composition and chemical structure on the

mechanical properties of chemically modified cellulose (Papers I–III) ......... 23

4.1.1. Chemical modifications by periodate oxidation and borohydride

reduction (Paper I) ..................................................................................... 23

4.1.2. Introduction of functional groups into the cellulose backbone

(Papers II and III) ...................................................................................... 28

a. Relationship chemical structure and mechanical properties of

handsheets made from modified fibres (Paper II) ..................................... 29

b. Effect of chemical functionality on the mechanical and barrier

properties of all-cellulose nanocomposites (Paper III) .............................. 37

4.2. Material applications of modified fibres and fibrils (Papers IV and

V) 45

4.2.1. Chemically modified fibrils as paper strength additives (Paper

IV) 45

4.2.2. Light weight materials from cellulose fibres (Paper V) ............ 48

5. CONCLUSIONS ....................................................................................... 54

6. ACKNOWLEDGEMENTS ...................................................................... 56

7. REFERENCES .......................................................................................... 58

1

1. INTRODUCTION

1.1. Context

There has been a great interest, both from society and from the scientific

community, to replace non-biodegradable and non-recyclable materials. At the

same time, the forest sector is facing the challenge of finding new applications

for forest resources due to a significantly reduced demand from the graphic paper

sector. For the Swedish forest industry in particular, there is also a need to

increase its presence in high growth and high value added markets.

Cellulose from wood is a potential candidate to replace oil-based materials and

to secure the position of the Swedish forest industry in international markets.

However, for many applications, the properties of this biopolymer need to be

improved in order for it to be able to compete with other materials. By using

proper chemical modification, in combination with suitable mechanical

treatments, the properties of cellulose can often be tuned to meet specific

demands.

1.2. Purpose of the study

The overall aim of this work was to combine different chemical modification

routes to modify cellulose fibres and to demonstrate how modified fibres and

fibrils prepared from the fibres can be used to create improved and new material

properties of films and networks prepared from these modified materials.

More specifically the objectives and corresponding activities were:

a. To investigate the influence of different chemical composition on

periodate oxidation and borohydride reduction.

2

b. To create a fundamental understanding of the relationship between the

chemical structure of the modified fibres and the physical and

mechanical properties of handsheets made of these modified fibres.

c. To prepare all-cellulose composites from differently chemically

modified nanofibrils and to compare the mechanical and barrier

properties of films prepared from these nanofibrils.

d. To prepare handsheets with differently modified nanofibrils and to

clarify the link between paper strength properties and fibril properties.

3

2. BACKGROUND

2.1. Hierarchical structure of cellulose in wood

Cellulose is a carbohydrate polymer with β-D-glucopyranose molecules

covalently attached by acetal bonds between the equatorial OH group of C4 and

the C1 carbon forming β-1,4-glycosidic bonds1 (a schematic image of the

hierarchical structure of wood is shown in Figure 1). Cellulose molecules are

linear and have a tendency to form inter- and intramolecular hydrogen bonds2.

These linear cellulose molecules are then associated in the form of fibrils, in

which crystalline regions are interrupted by short less ordered regions. These

fibrils are then associated into fibril aggregates which form cylindrical

shells/layers, intermixed with mainly amorphous lignin and hemicellulose, and

these layers are finally combined to build the cellulose fibre wall3.

Wood is formed as the result of the differentiation of the cells in the cambial

layer. These cells are composed mainly of cellulose, hemicellulose and lignin.

As the cells differentiate, their form and shape change to form the secondary wall

and some variations in the density of cells within the wood also occur. In the

case of softwood (the type of wood used throughout this work) the cells created

during the spring (earlywood) have thinner walls than the cells formed during the

late summer/fall (latewood)4.

The cell walls in wood consist of a primary layer, with cellulose nanofibrils

(CNFs) randomly distributed along the axis of the fibre, and three secondary

layers, S1, S2 and S3 where the fibrils in the thickest layer, S2, are oriented

basically along the fibre axis. The primary layer has the highest content of lignin

while S2 has the highest fraction of cellulose4.

Wood fibres are liberated from wood chips by mechanical or chemical treatment,

or by a combination of the two methods. Pulps produced by different processes

4

have different properties that often make them suitable for specific end products.

For the production of high-temperature chemithermomechanical pulp (CTMP),

wood chips are impregnated with a solution of alkaline sodium sulphite before a

pressurized refining. The sodium sulphite is added to soften the lignin,

facilitating fibre separation in the lignin-rich middle lamella. The energy

consumption has been optimised by preheating at a temperature of ~170 °C5.

Kraft pulping on the other hand is the main process for the production of

chemical pulp. This process involves the cooking of wood chips in a solution of

sodium hydroxide and sodium sulphide at a temperature of about 170 °C for 2 h.

With this treatment, lignin is degraded and largely removed while carbohydrates

undergo only partial degradation and dissolution6. After the kraft process, some

lignin, modified by alkaline condensation reactions, remains in the pulp, giving

these pulps their characteristic brown colour7. Residual lignin is thereafter

removed by a bleaching sequence selected to produce a certain pulp quality for a

certain end-use application. Another chemical process is the acid sulphite

process, by which so-called dissolving pulps are produced. In this process

hemicelluloses are removed to produce a pulp with a cellulose content higher

than 95%6. Dissolving pulp can also be produced by pre-hydrolysis, where wood

chips are pre-treated in a high temperature pre-hydrolysis step in order to remove

hemicelluloses, followed by kraft pulping to remove lignin8.

Wood pulp fibres consist of fibrils which are arranged in aggregates with a width

of 20–50 nm, which corresponds to 4–5 by 4–5 fibrils arranged in a square cross-

section. Each fibril consists is in turn of 20–50 cellulose polymer chains9,10. The

fibrils and their aggregates can be separated, often referred to as fibrillated, into

individual entities by chemical and/or mechanical treatment. This was first

demonstrated by Turbak et al. (1983)11 who produced microfibrillated cellulose

by a homogenization procedure. However, due to the high energy consumption

in this process, different pre-treatment methods have been developed. Examples

of such methods are carboxymethylation12, TEMPO-mediated oxidation13,14 and

cationic modification15,16.These chemical treatments introduce new

5

functionalities into the anhydroglucose units in the cellulose chain. Enzymatic

pre-treatments and mechanical pretreatments are also reported to reduce the

energy needed for fibrillation17,18.

Figure 1. The hierarchical structure of wood: from tree to cellulose. Adapted

from Moon (2008)19

2.1. All-cellulose composites:

All-cellulose composites can be divided into two types: non-derivatized and

derivatized all-cellulose composites. Figure 2 shows a schematic representation

of these two types of composites.

6

Figure 2. Schematic representation of the overall fibril structure for two types of

all-cellulose composites. a) Non-modified cellulose with a partly solubilized

cellulose shell around the crystalline core and b) Amorphous chemically

modified cellulose shell around a crystalline core.

Non-derivatized all-cellulose composites, where both the reinforcing fibres and

the matrix are native cellulose, were first reported by Nishino et al. (2004)20.

These composites were prepared by dissolving kraft fibres in DMAc/LiCl

followed by regeneration in the presence of ramie fibres. An alternative to this

method is to partially dissolve the cellulose on the surface of cellulose fibrils

followed by in-situ regeneration to form a matrix around the non-dissolved

cellulose21. In both cases, no functionalization or modification of the cellulose

was carried out. In the second approach, the cellulose fibrils are chemically

modified to different degrees without dissolution of the cellulose in the

derivatized “shell” surrounding the fibrils. From these modified fibres, different

types of high performance materials can be made, from wet-strong22 films to

thermoplastic films23.

Three of the most recent approaches (besides the methods that have been

described used in this work) for the preparation of all-cellulose composites

using the second procedure are benzylation, oxypropylation and esterification24.

During benzylation, a nucleophilic substitution of an oxide for a halide ion

occurs. Cellulose fibres are swollen in sodium hydroxide followed by addition of

benzyl chloride. The solution is stirred at temperatures above 100 °C for several

a) b)

7

hours. Finally, the thermoplastic material is formed by hot-pressing25,26.

Oxypropylation is performed using a Brønsted base to activate the hydroxyl

groups of the cellulose and create an anionic site to initiate a grafting from of

polypropylene oxide onto the cellulose. The reaction is performed under nitrogen

in an autoclave at temperatures of 130–150 °C27–29. Matsuruma et al. (2000) were

able to esterify cellulose by treatment with a mixture of pyridine, cyclohexane,

hexanoic acid and p-toluene sulfonyl chloride. The reaction takes place on the

surface of individual microfibrils and the less ordered regions of the cellulose

before reaching the interior of the microfibrils and crystalline regions of the

fibrils inside the fibres. Thermoplastic composites were then produced by

compression molding at temperatures between 155 and 170°C30. One major

disadvantage of these reactions is that high temperature and organic solvents are

required.

Zeronian et al. (1964)31 modified cellulose cotton linters following a two-step

procedure. First, periodate oxidation was performed to partially modify the

cellulose to dialdehyde cellulose, and the dialdehydes formed were then reduced

to dialcohols by borohydride reduction. This resulted in highly ductile materials,

but no mechanism to explain the behaviour of the modified cellulose was

proposed. Larsson et al. (2014)32 later performed the same oxidation on cellulose

fibres from wood and described it as a core-shell structure. As in the work by

Zeronian et al., it was shown that the ductility of the fibres increased with

increasing degree of oxidation. Furthermore, by hot pressing the papers made

from the modified fibres it was possible to produce transparent films with a

density as high as 1400 kg/m3, a structure dense enough to exhibit good oxygen

barrier properties23.

8

2.2. Oxidative chemical modification of cellulose

The insoluble character of cellulose in water and in organic solvents is the reason

why most of the cellulose modifications are performed heterogeneously. The

main goal of modification is the improvement, or complete alteration, of the

native properties of cellulose. There are too many approaches currently available

to modify cellulose to describe them all here. The common feature of the

reactions utilised in the present work, i.e. periodate oxidation, borohydride

reduction, chlorite oxidation, reductive amination and TEMPO-mediated

oxidation, is that they all are fairly mild reactions, all carried out in aqueous

media.

2.2.1. Periodate oxidation

Periodate oxidation is a selective method to oxidise a wide range of organic

compounds, primarily those with vicinal functional groups, such as 1,2-diols,

1,2-aminoalcohols, 1,2-hydroxyaldehydes, 1,2-ketone and 1,2-aminoaldehydes.

Due to the mild conditions of the reaction, it has frequently been applied to

carbohydrates, both for the investigation of molecular structure and for analytical

purposes33. It has been suggested that during the oxidation of 1,2-diols a cyclic

ester intermediate is formed between the periodate and the glycol before yielding

the products shown in Figure 3.

Figure 3. Proposed mechanism of periodate oxidation of a diol33.

In the particular case of cellulose, periodate oxidation occurs primarily at the

surface and less in the ordered regions of the cellulose, but during prolonged

9

reactions, crystalline regions of the cellulose are also affected34–36. Kim et al.35

suggested that the oxidation of crystalline cellulose is self-accelerating and that,

when a glucose unit is oxidised and converted to dialdehyde cellulose, the

neighbouring group is more susceptible to oxidation due to the local loss of

crystalline order.

Due to the high cost of periodate and the slow reaction rate at low temperature,

efforts have been made to improve the reaction conditions. The reactivity of

periodate towards cellulose has been reported to be improved by ultrasound

treatment, cellulose pulp micronization, higher temperatures or a combination of

high temperature and the addition of salts37–39. In addition, methods of recycling

the chemical have been reported40–43.

Once the aldehydes are formed, a further conversion into carboxylic acids,

primary alcohols or aminated compounds is possible.

2.2.2. Reduction of aldehydes using sodium borohydride

As shown in Reactions 1 and 2 below, in the ideal case, four moles of aldehydes

or ketone react with one mole of borohydride44. The reaction takes place rapidly

at room temperature and the corresponding alcohols are obtained in good yields.

During the reduction of dialdehyde cellulose, the alkalinity is controlled to avoid

cellulose depolymerisation45.

2.2.3. Oxidation of aldehydes using sodium chlorite

Oxidation of aldehydes to carboxylic acids by sodium chlorite is an inexpensive

method that takes place under acidic conditions. However, hypochlorite ions

must be removed to avoid side reactions since HOCl/Cl- is a stronger oxidant

10

than ClO2-/HOCl and oxidation of ClO2

- to ClO2 can occur44. A way to avoid this

effect is by using hydrogen peroxide can be used as HOCl scavenger as shown in

Reactions 3 to 6.

During the oxidation of dialdehyde cellulose, the cellulose chains are also

depolymerized, and as the degree of oxidation increases (i.e. increasing aldehyde

content) the more pronounced the depolymerisation is45. The product after

chlorite oxidation is known as dicarboxylic acid cellulose.

2.2.4. Reductive amination of aldehydes

Aldehydes and ketones canreact with ammonia and primary or secondary amines

to form primary, secondary or tertiary amines. The reaction involves the

formation of a carbine intermediate which dehydrates to form an imine. Under

weakly acidic to neutral conditions the imine is then protonated and an iminium

ion is formed. Subsequent reduction of this ion produces the correspondent

amine46, as described in Figure 4.

Figure 4. Reductive amination mechanism.

11

2.2.5. TEMPO-mediated oxidation

Oxidation by oxoammonium salts, such as TEMPO, is a mild method for the

oxidation of primary and secondary alcohols to aldehydes and ketones. The

oxidation is typically performed under a catalytic process that involves an in-situ

generation of the oxoammonium cation by one-electron oxidation of a small

quantity of nitroxide with a primary oxidant, such as sodium hypochlorite47, as

shown in Figure 5. The rate of the reaction is highly dependent on the pH and

most of the reactions are carried out under basic conditions. One mole of

carboxylic acid is formed by consuming two moles of NaClO, while one mole of

aldehyde is formed by one mole of NaClO48. Unlike periodate oxidation,

TEMPO oxidation introduces carboxylates and aldehyde groups into the

amorphous regions of cellulose, while maintaining the crystallinity. Furthermore,

only one of the two C6 groups in the repeating unit of cellulose (in a CNF) can

be oxidized, since only one of the C6 hydroxyls is oriented outwards from the

surface of the fibrils10.

Figure 5. Catalyst cycle for the TEMPO/NaOCl oxidation49.

12

2.3. Applications of chemically modified cellulose

Cellulose is a potent candidate for replacing non-renewable and non-

biodegradable materials. Among its advantages, high stiffness, low cost and

biodegradability can be mentioned50–52. However, there is still a need for further

improvement of the properties of cellulose, which in this work has been achieved

by chemical modification. Nowadays a broad range of cellulose applications

include food additives53 and packaging materials54,55, but research efforts are also

being made to develop biomedical scaffolds56–58, energy storage devices59–61 and

composites51,52. The focus of this thesis is, however, on fibre-based materials and

CNFs for barrier films and as paper strength additives.

2.3.1. Paper-based materials

Paper is a network of fibres obtained from cellulosic materials. The network is

able to support loads due to the strong adhesive contact between fibres that are

joined by mechanical interlocking, electrostatic interactions, van der Waals

forces, the interdifusion of less ordered cellulose molecules and hydrogen

bonds62. Paper displays significant rate-dependant mechanical properties, in

particular in response to load, moisture or temperature63. The mechanical

properties of paper are controlled by two main properties on the structural level:

the single fibre and the fibre network. While the properties of individual fibres

are controlled by the plant species, the plant morphology and how the fibre was

liberated from the plant64,65, the fibre network is affected not only by the

properties of the fibres and their interaction, but by the introduction of

chemicals, such as polymers64,66,67.

13

3. EXPERIMENTAL

This section summarizes the most important experimental aspects of the present

work. For more detailed information, the reader is referred to the corresponding

sections in the papers.

3.1. Materials

3.1.1. Fibres and nanofibrils

In Paper I, four different cellulose fibres were used: bleached softwood kraft

pulp, never dried unbleached softwood kraft pulp, peroxide-bleached softwood

CTMP and a dissolving grade sulphite softwood pulp. The first three types of

fibres were supplied by SCA forest Products and the dissolving grade pulp was

obtained from Domsjö Aditya Birla AB, Sweden. The bleached kraft fibres were

also used in Papers II, III and V.

The carboxymethylated CNFs used in Paper IV were provided as 2 wt% gel by

RISE Bioeconomy, Stockholm, Sweden. Prior to the production of CNFs, the

fibres were carboxymethylated according to an earlier described procedure12.

All other CNFs used in this work were prepared at KTH (protocol described

below).

3.1.2. Chemicals

Sodium metaperiodate (99%), was purchased from Alfa Aesar. Dopamine

hydrochloride (98%), 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC),

hydrogen peroxide (30 wt% in water), hydroxylamine hydrochloride (99%),

pDADMAC with a molecular weight of 400–500 kDa, 2-propanol (99.9%),

sodium acetate (99%), sodium borohydride (≥96%), sodium carbonate (≥99.5%),

sodium chlorite (80%), sodium hypochlorite solution (10-15% available

chlorine), sodium phosphate monobasic monohydrate (98%) and 4-acetamido-

14

2,2,6,6-tetramethylpiperidine 1-oxyl (TEMPO) (≥98.0%) were all purchased

from Sigma-Aldrich. Standards solutions of sodium hydroxide and hydrochloric

acid (1 M) were purchased from Merck Millipore. PVAm with a molecular

weight of 340 kDa was obtained from BASF and used as received. Tap water

was used for handsheet preparation but deionised water was used in all other

cases.

3.2. Methods

3.2.1. Chemical modifications

A range of different chemical modifications (Figure 6) were studied. In Paper I,

only periodate oxidation (Structure 2) and borohydride reduction (Structure 3)

were performed. In paper II, the study was expanded to a combination of

different modifications and Structures 2 to 8 were produced. Later, in Paper III,

all the different modifications (Structures 2 to 9) were carried out.

The reactions were stopped by washing the fibres until a conductivity below 5

µS/cm was achieved.

Figure 6. A reaction scheme summarizing the different chemical modifications.

15

Periodate oxidation (Structure 2): In Paper I, a fibre suspension of 4 g/L was

allowed to react with 5.4 g NaIO4/g fibre (four times the amount that is

theoretically needed to oxidize 100% of the diols in pure cellulose) in the

presence of 6.3 wt% 2-propanol at RT for 6 h and 12 h. In addition, for CTMP a

24 h reaction was also performed. In Papers II, III and V, the reaction was

optimised to reduce the chemical consumption and to increase the reaction rate,

and the fibre suspension used had a concentration of 20 g/L, and 1.35 g NaIO4/g

fibre was allowed to react at 50 °C for different times to obtain different

carbonyl contents in the fibres. In Papers II and V, the oxidation was performed

for 30 min, while 2 hours were needed in Paper III to achieve higher degrees of

oxidation. In Paper IV, carboxymethylated CNFs were modified by diluting the

fibrils to 1.5 wt% and using 0.7 g NaIO4/g CNFs. Mixing was carried out by

using an Ultra-turrax particle dispenser for 5 min at 12 000 rpm, after which the

reaction was allowed to proceed for 24 h at RT.

Borohydride reduction (Structure 3): Periodate-oxidised fibres with a

concentration of 4 g/L (Paper I) or 8 g/L (Papers II and III) with 0.01 M

NaH2PO4 were allowed to react with 0.5 g NaBH4/g for 2 h at RT32

Reductive amination (Structure 4): Periodate-oxidised fibres were aminated

with hydroxylamine hydrochloride. Before the reaction between hydroxylamine

and the periodate-oxidised fibres, the pH was set to 4 and thereafter the reactants

were allowed to react for 2 h at RT. The stoichiometric amount of

hydroxylamine added was five times the amount of carbonyl groups. An in-situ

reduction of the imine-formed group was then achieved by adding NaH2PO4 and

0.5 g NaBH4/g fibre. The reaction was then continued for another 2 h at RT.

Chlorite oxidation (Structure 5): Periodate-oxidised fibres were further

oxidised with sodium chlorite. Fibres were dispersed to a concentration of 20

g/L, and NaClO2 and H2O2 were added to a ratio of 2.5 mol/mol aldehyde. The

16

suspension was acidified with 1 M HCl to pH 5 and allowed to react for 48 h at

RT.

TEMPO-mediated oxidation (Structure 6): Non-modified bleached kraft

fibres were oxidised following a previously described procedure68. Fibres were

first dispersed in 0.1 M acetate buffer at pH 4.8. Thereafter, NaClO2, NaClO and

TEMPO were added to the fibre suspension. The reaction was allowed to

proceed for 48 h at 40 °C.

In Paper IV, carboxymethylated CNFs was first prepared, followed by periodate

oxidation or dopamine grafting as shown in Figure 7.

Figure 7. Schematic description of the chemical modification of

carboxymethylated CNFs.

Dopamine-grafted CNFs: Dopamine was grafted to carboxymethylated CNFs

using EDC as a coupling agent. The CNF dispersion was first diluted to 0.7 g/L

using phosphate-buffered saline solution adjusted to pH 5, then mixed with EDC,

followed by the addition of dopamine. The molar ratio of dopamine:COOH was

0.5:1 and of EDC:dopamine 1.67:1. The mixture was allowed to react for 6 h at

RT 69.

The excess of chemicals in the CNF dispersion was removed by dialysis.

17

3.2.2. Characterization of fibres and fibrils

Carbonyl content: In order to determine the content of aldehydes, a

stoichiometric reaction with hydroxylamine was performed where the HCl

released was titrated with NaOH to determine the amount of aldehydes on the

cellulose70

Total charge determination: Conductometric titration was carried out to

determine the total amount of acidic groups71. The fibres were converted to the

H-form and then titrated with NaOH. The titration is characterised by three

different phases. In the first phase, the strong acidic groups are titrated,

decreasing the conductivity due to the lower conductivity of sodium ions than

protons; in the second phase, the immobile weak acidic groups are titrated,

which does not affect the conductivity; and finally the third phase represents

accumulation of NaOH, which leads to a linear increase in conductivity (SCAN-

CM 65:02 Total acidic group content).

Surface charge determinations: Surface charge was determined with a Stabino

polyelectrolyte titrator. This method is based on the determination of the change

in sign of the streaming potential in a dispersion of fibrils using a streaming

current detector72. A cationic polyelectrolyte is stepwise added to the dispersion

and the inflexion point of potential is taken as a point of equivalence and the

charge of the fibril dispersion is determined from the added amount of cationic

charges of the polyelectrolytes.

WRV measurements: The water retention value (WRV) measures the capacity

of fibres to hold water under a centrifugal force field. A test pad of fibres was

formed by dewatering a pulp suspension on a membrane. The fibres were

centrifuged at 30 000 rpm for 30 min, then weighed, dried and weighed again

and the results are presented as g water/g of dry fibres (SCAN 62:00: Water

Retention Value)

18

Handsheet preparation: Handsheets with a target grammage of 100 g/m2 were

prepared in a Rapid Köthen sheet former (Paper Testing Instruments, Austria)

and woven metal wires (The Mesh Company Ltd, Warrington, UK) of 400 mesh

attached to regular sheet-former boards were used to dry the handsheets. The

handsheets were dried for 15 min at 93 °C, under a reduced pressure of 95 kPa.

In paper IV, handsheets were prepared in combination with the chemically

modified CNFs, as shown in Figure 8. Two series of handsheets were prepared,

one with PVAm or pDADMAC with or without retention aid and three different

amounts of CNFs were used: 2 wt%, 5 wt% and 15 wt% with respect to the dry

weight of fibres.

Figure 8. Schematic representation of the preparation of handsheets with/or

without retention aid and CNFs.

CNF preparation: A Microfluidizer M-110 EH (Microfluidics Corp.,

Westwood, MA, USA) was used to prepare CNFs. The microfluidizer had two

chambers connected in series; a coarser chamber combination, 400 µm and 200

µm in diameter, operating at a pressure of 900 bar, followed by a more narrow

19

chamber combination, 200 µm and 100 µm in diameter, operating at a pressure

of 1500 bar. Highly charged fibres were passed one time through the large

chamber and four times through the small chamber combination while the low-

charged fibres were passed three times through the large chamber combination

followed by seven times through the small chamber combination.

Preparation of CNF films: CNFs were first diluted to 0.2 wt% and dispersed

for 10 min at 12 000 rpm using an IKA Ultra Turrax particle dispenser. The

dispersion was vacuum filtered through a 0.45 µm polyvinylidene fluoride

membrane (Durapore, Merck Millipore). After filtration, another membrane of

the same type was placed on top of the film, a paperboard was placed on each

side of the film and the entire sandwich structure was dried for 15 min at 93 °C

under a reduced pressure of 95 kPa, using the dryers of a Rapid Köthen sheet

former.

Nitrogen analysis: This technique consists of incinerating the sample at 1050 °C

in an oxygen-poor environment followed by excitation of the NO formed with O3

to NO2*. The amount of nitrogen was quantified by measuring the light emitted

when NO2* was relaxed to NO2. This measurement was made with an ANTEK

7000 Model 737 (Antek instruments Inc., USA).

X-ray diffraction: The analysis is based on the constructive interference of

monochromatic X-rays in a crystalline sample, in this case cellulose. The

diffraction pattern of the samples was determined in an X’Pert PRO XRD

(PANanalytica, The Netherlands), using copper radiation Kα (1.5418 Å), with a

voltage of 40 kV and a current of 45 mA.

Scanning electron microscopy: A Hitachi S-4800 field-emission scanning

electron microscope was used and the samples were sputtered with a ~10 nm Pt-

Pd coating in a 208 HR Cressinton sputter coater before imaging to suppress

specimen charging.

20

Mechanical testing: Prior to mechanical testing, all the handsheets and films

were conditioned for at least 24 h at 23 °C and 50% RH.

The thickness of the handsheets was evaluated according to SCAN-P 88:01:

Structural thickness and structural density. Thereafter, the handsheets were cut

into 15 mm wide strips and tensile tested using an Instron 5944 with a 500 N

load cell. The strips had a free span of 100 mm between the clamps and were

strained at a constant rate of 100 mm/min. Two handsheets from the same

sample and a total of ten specimens were tested. The results were reported for

those which did not fail immediately in the jaw face. For wet tensile strength

testing, the sheets were soaked in water for 1 h prior to testing according to

SCAN-P 20:95: Wet tensile strength and wet tensile strength retention.

Zero- and short-span73 (0.4 mm) tensile tests were carried out on the handsheets

using a Pulmac Z-span 3000. Eight test pieces were evaluated for each sample.

A short-span compressive test was also performed on chemically modified

unbleached kraft fibres using a L&W Compressive Strength Tester STFI (ABB

Lorentzen and Wettre). The paper strip was placed between two clamps with a

0.7 mm free clamping length, followed by application of a compressive force.

Ten measurements were made per sample.

CNFs films were tested with the same Instron instrument, as previously

described, except that the test pieces had a width of 5 mm, the free span between

the clamps was 15 mm, and the films were strained at a constant rate of 1.5

mm/min. The thickness of the films was measured using a digital Mitutoyo

thickness gauge 547-401 at ten random locations.

Dynamic mechanical analysis: A Perkin-Elmer Instruments DMA7e with a RH

accessory was used to measure the storage modulus as a function of relative

humidity. In the present work the RH was gradually increased to study the

influence of moisture on the mechanical properties of CNFs films.

21

Dynamic vapour sorption: With this technique it is possible to determine how

much and how fast moisture is adsorbed by a sample when the RH is changed

continouosly or in steps. The measurements were performed in a TA instruments

Q5000SA DVS.

Oxygen permeability of films: This technique measures the quantity of oxygen

passing through a plane surface, in this case a CNF film, per unit time. A

MOCON OX-TRAN 2/21 (Mocon Inc, Minneapolis, MN, USA) was used and

the permeability measurements were performed at 23 °C at 50% RH or 80% RH

respectively.

Preparation of low density fibrous networks: The fibres were first subjected to

periodate oxidation to achieve a carbonyl content of 1.3 mmol aldehydes/g fibres

followed by a chlorite oxidation for 20 h. After the fibres had been washed, they

were suspended to a concentration of 10 g/L and mixed with an Ultra-turrax at

10 000 rpm (0 to 30 min). Covalently connected fibre networks were achieved

by first allowing the fibres to react with 2.7 g NaIO4/g fibre for 16 h at RT

without stirring. The structure of the materials was then maintained by a solvent

exchange to acetone for 15 min followed by washing with water until the

conductivity was below 5 µS/cm. Finally, a second solvent exchange to acetone

was carried out. This was repeated twice, with 15 minutes soaking each time,

before allowing the materials to air dry at RT. A schematic description of this

treatment is shown in Figure 9.

22

Figure 9. Preparation of fibre networks.

Compression testing: The mechanical characterization of the fibre networks

was achieved with an Instron 5566 universal testing machine (Norwood, MA,

USA) equipped with a 500 N load cell. The material was characterised in both

the dry and the wet states. Cylindrical samples were placed between two flat

compression plates and compressed at 10%/min until 80% compression was

achieved. Each sample was tested in duplicate. The compressive modulus was

calculated from the initial linear section of the load curve and the yield strength

from the intersection between the tangent to the elastic region and the tangent to

of the plateau region. For wet characterization, the samples were placed in water

to allow them to soften, and were then tested in the same way as the dry samples.

23

4. RESULTS AND DISCUSSION

4.1. Influence of fibre composition and chemical structure on the

mechanical properties of chemically modified cellulose (Papers I–

III)

4.1.1. Chemical modifications by periodate oxidation and

borohydride reduction (Paper I)

In order to determine the influence of chemical composition on the mechanical

properties of the dialdehyde- and dialcohol cellulose modified fibres, four

different lignocellulosic fibres were modified by periodate oxidation and

borohydride reduction, as shown in Figure 8. As shown in Table 1, the fibre

composition had a strong influence on the modification and the different fibres

showed different reactivities to form aldehydes. Bleached fibres had the highest

reactivity, where 1.9 and 2.5 mmol of aldehydes/g of fibres were formed after 6

and 12 h, respectively. Despite the high content of cellulose in the dissolving

grade fibres, the reactivity was lower than that of the bleached kraft fibres, 1.1

and 1.5 mmol/g after 6 and 12 h reaction respectively. During the production of

dissolving fibres, a large amount of hemicelluloses are removed in order to avoid

problems later in the alkalization step74. However, the removal of hemicelluloses

causes increase hornification in the fibre wall75,76.

After periodate oxidation of unbleached kraft fibres, 1.20 and 1.77 mmol of

aldehydes/g of fibres had been formed after 6 and 12 h, respectively. During the

oxidation of CTMP, 1.22 mmol/g and 1.45 mmol/g of aldehydes were formed

after 6 and 12 h, respectively. CTMP fibres were also oxidised for 24 h, with the

aim to achieve the same carbonyl content as bleached kraft fibres in order to

compare the properties obtained after the modification. As shown in Table 1, it

takes almost twice as long for lignin-rich fibres to achieve a carbonyl content

similar to that of bleached fibres. There are naturally several reasons for this

24

difference. Firstly, there is a difference in the concentration and availability of

vicinal diols in lignin-rich fibres. Secondly, it has been reported that periodate

can react with phenolic groups in lignin, such as guaiacol and substituted

guaicols77. During this reaction, periodate reacts with the guaiacol group,

creating a mesomeric radical. In a second oxidation step, the radical may react

with an hydroxyl radical to produce a hemiacetal which spontaneously loses one

mole of methanol and is converted into an ortho-quinone78. This is a side

reaction that competes with the oxidation of cellulose. In this work no efforts

were made to quantify these side reactions, but their existence was qualitatively

observed as a colour change of the unbleached fibres.

Table 1. Aldehyde content of the fibres after periodate oxidation. The numbers

in parentheses show the degree of oxidation assuming pure cellulose.

Fibres 0 h (mmol/g) 6 h (mmol/g) 12 h (mmol/g)

Bleached

chemical 0.03 ± 0.01

1.90 ± 0.10

(16%)

2.50 ± 0.10

(21%)

Dissolving grade 0.03 ± 0.02 1.09 ± 0.04

(9%)

1.48 ± 0.01

(12%)

Unbleached

chemical 0.40 ± 0.10

1.20 ± 0.10

(10%)

1.77 ± 0.02

(15%)

CTMP 0.32 ± 0.03 1.22 ± 0.01

(10%)

1.45 ± 0.01

(12%)

After periodate oxidation, the fibres were further modified with sodium

borohydride to reduce the aldehydes formed to primary alcohols. It was assumed

that the degree of modification was the same after the reduction treatment. The

mechanical properties of papers made from the modified fibres after periodate

oxidation and borohydride reduction were then determined (Figure 10). The

mechanical properties of handsheets are dependent on the density79 and it was

therefore important to determine how the consolidation of the fibres was affected

25

by the modifications. Despite the differences in reactivity of the fibres, only a

slight increase in sheet density was observed after periodate oxidation, related to

the increased fibre-fibre joint strength caused by hemiacetal formation31,64,80. The

greater consolidation could also explain the increase in Young´s modulus. After

borohydride reduction, a significant increase in density was observed,

presumably due to a softening and increased swelling of the fibres after

borohydride reduction32. The density of the fibres increased as the degree of

modification increased. This effect was also observed by Larsson et al. (2016)23.

After borohydride reduction, the tensile strength and the modulus increased to a

greater extent than after periodate oxidation. Larsson et al. (2014)32 suggested

that an amorphous shell of dialcohol cellulose surrounding the crystalline core of

non-modified cellulose is formed after the modification And that this shell

facilitates molecular mobility and provides flexibility. The increase in modulus

of the handsheets prepared after borohydride reduction is probably related to an

increased density and an increase in the number of inter-fibre joints.

In general, the mechanical behaviour of handsheets was dependent on their

chemical composition, the reaction performed and the degree of modification.

Sheets made from CTMP showed the least improvement after chemical

modification presumably due to the high lignin content which affects the strain-

at-break of the fibres. Zhang et al. (2013)81 found that selective removal of lignin

increased the strain-at-break by 20%. This is probably because the removal of

lignin allows a better slippage of the fibrils inside the fibre wall and, besides,

lignin is in itself a stiff polymer. In mechanical pulps, the lignin is still present in

the fibre wall, and this affects the formation of fibre-fibre joints.

26

Bleached Dissolving Unbleached HT-CTMP0

200

400

600

800

1000

1200

Ref IO4(6h) IO

4(6h)+BH

4(1h) IO

4(12h)

IO4(12h)+BH

4(1h) IO

4(24h) IO

4(24h)+BH

4(1h)

Ref IO4(6h) IO

4(6h)+BH

4(1h) IO

4(12h)

IO4(12h)+BH

4(1h) IO

4(24h) IO

4(24h)+BH

4(1h)

Ref IO4(6h) IO

4(6h)+BH

4(1h) IO

4(12h)

IO4(12h)+BH

4(1h) IO

4(24h) IO

4(24h)+BH

4(1h)

Den

sity

(kg

/m3 )

Ref IO4(6h) IO

4(6h)+BH

4(1h) IO

4(12h)

IO4(12h)+BH

4(1h) IO

4(24h) IO

4(24h)+BH

4(1h)

a)


10

20

30

40

50

60

70

80

90

Ten

sile

str

engt

h (M

Pa)

b)


2

4

6

8

10

12

Str

ain-

at-b

reak

(%

)

c)


1

2

3

4

5

6

7

You

ng's

mod

ulus

(G

Pa)

d)

Figure 10. Mechanical properties of handsheets made from modified fibres of

different pulping process.

In order to estimate the properties of individual fibres, zero-span measurements

were also performed. As shown in Figure 11, the modifications affected the

mechanical properties at the fibre level. After periodate oxidation there was a

slight decrease in the zero-span tensile strength of the fibres, which can be

explained by an increase in the stiffness and brittleness of the fibres as a

consequence of the crosslinks. Moreover, crosslinks within individual fibres

prevent slippage between fibrils and thereby reduce the flexibility of single fibril

aggregates needed to distribute the applied stress31,82. In contrast, after periodate

oxidation and borohydride reduction, both the zero- and the short- span tensile

strengths greatly increased, which supports the model that an amorphous shell of

dialcohol cellulose is formed around each fibril32.

27

Bleached Dissolving Unbleached CTMP0

20

40

60

80

100

120

140 Ref IO

4(6h) IO

4(6h)+BH

4(1h) IO

4(12h)

IO4(12h)+BH4(1h) IO4(24h) IO4(24h)+BH4(1h)

Zer

o-sp

an te

nsile

str

engt

h (M

Pa)

Ref IO4(6h) IO4(6h)+BH4(1h) IO4(12h)

IO4(12h)+BH

4(1h) IO

4(24h) IO

4(24h)+BH

4(1h)

a)

Bleached Dissolving Unbleached CTMP0

20

40

60

80

100

120

140

Sho

rt-s

pan

tens

ile s

tren

gth

(MP

a)

b)

Figure 11. Mechanical properties at the fibre level as estimated by zero- and

short-span measurements.

Unbleached kraft fibres are commonly used to produce, for example, board and

liners used to make corrugated boxes. Boxes are piled on top of each other and

stored for long periods of time. Therefore, a high compressive stiffness is usually

desired since it has been shown to reduce the creep and mechanosorptive creep

of these materials. Short-span compression tests (SCTs) were performed to

characterize the compressive properties of these sheets and the results are shown

in Figure 12. There was an initial increase from 27 to 46 kNm/kg when 1.90

mmol of aldehyde/g fibre were introduced. However, rather interestingly, no

further improvement was achieved by prolonged oxidation or borohydride

reduction. The results can probably be explained by a high resistance towards

delamination of fibre layers in the sheet; due to the presence of crosslinks

between the dialdehyde fibres, to the crosslinks within each fibre wall, and due

to the highly consolidated and flexible structure in the dialcohol fibre sheets.

28

20

25

30

35

40

45

50

55

Sho

rt-s

pan

Com

pres

sion

inde

x (k

Nm

/kg)

(1.77 mmol/g)

IO4(12h)+

BH4(1h)

IO4(12h)IO

4(6h)+

BH4(1h)

IO4(6h)Ref

(1.20 mmol/g)

Figure 12. Short-span compression strength of modified unbleached kraft fibres.

4.1.2. Introduction of functional groups into the cellulose

backbone (Papers II and III)

Bleached kraft fibres were modified using a combination of different procedures

(Figure 6). The study was divided into two different parts: chemical

modification of fibres at low degrees of modification and CNFs modified to

higher degrees of modification. Studying CNFs makes it possible to exclude the

influence on the mechanical properties of networks of the structure at the fibre

network level.

29

a. Relationship chemical structure and mechanical properties of

handsheets made from modified fibres (Paper II)

Different methods have been developed to improve the mechanical properties,

i.e. the modulus, and stress and strain at break, of cellulose-fibre-based materials.

However, there is still no clear understanding of the effect that a specific

functional group has on the mechanical performance of handsheets and at which

structural level of the fibre wall or the fibre/fibre joint these chemical

components have their major effect.

Following the scheme shown in Figure 6, the fibres were modified using a

combination of different reactions. First, periodate oxidation was performed to

selectively cleave the C2-C3 bond of the cellulose backbone to produce

dialdehyde cellulose (Structure 2). The formation of aldehydes allows

conversion to other functional groups and different reactions were therefore

performed to produce different functional groups. Borohydride reduction was

used to produce dialcohol cellulose (Structure 3), reductive amination was

carried out to covalently attach hydroxylamine (Structure 4), and chlorite

oxidation was used to produce carboxylic acids (Structure 5). TEMPO oxidation

was also performed to produce carboxylic acids in the C6 position (Structure 6)

of cellulose and this was combined with the preparation of either dialdehyde

cellulose (Structure 7) or dialcohol cellulose (Structure 8).

The degree of modification of the fibres was determined by total aldehyde

content and charge density analysis, as shown in Table 2. A total of 1.36 mmol

of aldehyde/g of fibre was produced after periodate oxidation. The same amount

was introduced when a combination of TEMPO and periodate oxidation was

used (Structure 7), and it was also possible to convert all the aldehydes to

carboxylic acid after chlorite oxidation (Structure 5), i.e. by introducing 1350

µeq/g of charges, compared with 783 µeq/g of carboxylic acids that were

introduced after the TEMPO oxidation (Structure 6).

30

A decrease in charge density was observed after periodate oxidation (Structure 2)

and further reduction with sodium borohydride (Structure 3). It has been reported

that during these reactions low molecular weight compounds are removed32. A

decrease in charge density was also observed when TEMPO and periodate

oxidation (Structure 7), or TEMPO and periodate oxidation followed by

borohydride reduction (Structure 8), were combined. On the other hand, it has

been reported that dialdehyde cellulose can undergo beta-elimination in an

alkaline medium83. From a mechanism point of view, the 5-substituent in the

oxidised anhydroglucose unit (AGU), which is in the beta-position with respect

to the C3-aldehyde, is eliminated. The C2 aldehyde has no beta substituent and

no elimination occurs. Two cellulose fragments are obtained with one aldehyde

group each. This process occurs at the location of an isolated oxidised AGU. In

the presence of adjacent oxidised AGUs, a clustered elimination could also

possibly take place. In this case, a similar process was observed, but the C3

aldehyde gave rise to three fragments; two shortened cellulose chains, and one

low-molecular-weight fragment with two aldehyde groups84.

Table 2. Carbonyl content and charge density of the modified fibres. The

numbers in parentheses show the degree of oxidation.

Sample Carbonyl content

(mmol/g)

Charge density

(µeq/g)

1:Ref 0.03 ± 0.01 60 ± 3

2:IO4 1.36 ± 0.18 (11%) 48 ± 4

3:IO4+BH4 * 42 ± 8

4:IO4+NH2OH/BH4 * 52 ± 1

5:IO4+ClO2 * 1352 ± 74

6:TEMPO * 783 ± 27

7:TEMPO+IO4 1.38 ± 0.02 (11%) 646 ± 7

8:TEMPO+IO4+BH4 * 628 ± 2

*Below the detection limit of the method

31

A decrease in cellulose crystallinity is generally associated with an increase in

the strain-at-break of fibres and materials made from fibres85 and the crystallinity

after the different modifications was therefore measured to study the changes

that occur in the supramolecular structure of the cellulose (Figure 13). After

periodate oxidation (Structure 2) the crystallinity decreased in agreement with

earlier investigations35, but after borohydride reduction (Structure 3), reductive

amination (Structure 4) or chlorite oxidation (Structure 5) an apparent increase in

crystallinity was observed. This apparent increase in crystallinity could possibly

be due to an increase in moisture sorption. It has been reported that a decrease in

crystallinity leads to an increase in moisture sorption86 and larger amounts of

moisture have been reported to induce changes in the XRD spectra87,88.

The amount of moisture absorbed by the different chemical structures was later

evaluated on CNF films by DVS, as shown in Figure 18. The films prepared

from fibrils with different structures showed significant differences.

No change in crystallinity (i.e. no change in diffraction pattern) was observed

after TEMPO oxidation (Structure 6), which agrees well with previous studies89,

and when the introduction of carboxylic acids was combined with a modification

to dialdehyde cellulose (Structure 7) or dialcohol cellulose (Structure 8) the

crystallinity index values obtained were similar to that of the corresponding

chemical treatment without TEMPO.

32

10 15 20 25 30 35

a.u.

2θ (degrees)

(1) Ref (2) IO

4

(3) IO4+ BH

4

(4) IO4+ NH

2OH/BH

4

(5) IO4+ClO

2

(6) TEMPO (7) TEMPO+IO

4

(8) TEMPO+IO4+ BH

4

Figure 13. XRD diffractograms of chemically modified cellulose fibres.

The mechanical properties of fibre networks can be related to the ability of the

fibres to swell in water since this will increase the flexibility of the fibres and

increase the contact area between the fibres in the fibre/fibre joints50. In this

study, the swelling capacity was measured as WRV and the results are shown in

Figure 14. Fibres with a low charge density did not differ significantly from the

reference, whereas highly charged fibres showed a significantly higher WRV. In

the presence of carboxylic acids there is an increase in the osmotic pressure due

to the difference in concentration of mobile ions inside the fibre wall and in the

bulk solution90. Also, as the pH is increased, there was an increase in the WRV

and as the amount of carboxylic groups increased so did the WRV91.

Interestingly, when an increase in concentration of carboxylic acids was

combined with the introduction of either dialdehyde cellulose (Structure 7) or

dialcohol cellulose (Structure 8) an even greater increase in the WRV was

observed than with TEMPO oxidation alone. This trend was most clearly

detected at higher pH even though the charge density of these fibres was

somewhat lower than with TEMPO oxidation alone. This can be explained by a

33

combination of two different effects: a decrease in crystallinity will cause an

increase in the water adsorption and a relatively high charge density will increase

the swelling.

Ref (1) (2) (3) (4) (5) (6) (7) (8)0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

ND pH 2 ND pH 8.5

Sample number

Wat

er r

eten

tion

valu

e (g

wat

er/g

fibre)

0

250

500

750

1000

1250

1500

Cha

rge

dens

ity (

µeq/

g)

(1) Ref(2) IO

4

(3) IO4+BH

4

(4) IO4+NH

2OH/BH

4

(5) IO4+ClO

2

(6) TEMPO(7) TEMPO+ IO

4

(8) TEMPO+IO4+BH

4

Figure 14. WRV (left axis) of chemically modified fibres, measured at pH 2 and

8.5, and the charge density (open circles) of the fibres on the right y-axis.

The modification also changed the morphology of the fibres (Figure 15). After

periodate oxidation (Structure 2) the fibres had a smaller diameter, an effect that

was even more pronounced after reductive amination (Image 4 in Figure 15).

The characteristic densification observed after borohydride reduction (Image 3 in

Figure 15) also starts to be visible, but higher degrees of oxidation were needed

to form a film23,32. After chlorite oxidation (Structure 5), a significant

densification, together with an emerging film formation, was observed,

presumably due to the high level of swelling of these fibres. Densification was

also observed after TEMPO oxidation (Structure 6), and this was even more

34

pronounced after a combination of TEMPO oxidation, periodate oxidation and

borohydride reduction (Structure 8).

Figure 15. FE-SEM images of sheets made of chemically modified fibres.

Figure 16 shows the mechanical properties of sheets made from the modified

fibres. All the modifications improved the properties of the handsheets, but to

different extents. After periodate oxidation (Structure 2), there was an increase in

tensile strength, but this improvement was accompanied by a slight decrease in

strain-at-break, due to the presence of covalent crosslinks in this material. The

presence of crosslinks was further indicated by an increase in the wet strength of

the handsheets (Table 3). After borohydride reduction (Structure 3) the density,

tensile strength and strain-at-break all increased due to a better consolidation

during the pressing and drying for these fibres32. The reductive amination

(Structure 4) did not show the same increase in ductility, which suggests that it is

35

not only the presence of hydroxyls that provides ductility, but that the chemical

environment, i.e. the surface energy and hence the adhesive interaction between

the fibres, is also important. Handsheets prepared from highly charged fibres

showed a large increase in density, which could partly explain the improvement

in their mechanical properties (due to a better fibre-fibre interaction). However,

the results derived from highly charged fibres differ greatly, indicating that the

mechanical properties are controlled not only by density. Handsheets prepared

from dicarboxylic acid cellulose (Structure 5) showed a tensile strength of

75 MPa and a Young´s modulus of 9.6 GPa, values that are comparable to those

of films prepared from nanocellulose13. Nevertheless, the handsheets were shown

to be very stiff, probably due to the emerging film formation observed in Figure

15, which restrains the fibre movement during tensile loading and hence stiffens

the fibre network. Comparing both dialcohol (Structure 3) and dicarboxylic acid

cellulose (Structure 5), the effect of a specific functional group on the properties

of the final material is more evident. The anhydroglucose unit has been opened

for both these types of modification but the final mechanical properties are

completely different. The major difference between these different fibres is the

swelling induced by the high charge of the dicarboxylic acid containing fibres as

shown in Figure 14.

After TEMPO oxidation (Structure 6), there is an increase in both tensile

strength and strain-at-break. The strain-at-break, in this case is similar to that of

sheets made from fibres with a partial derivatization of the cellulose to dialcohol

cellulose, but it should be noted that at higher degrees of modification dialcohol

fibres have shown a strain-at-break of about 40%23. After both TEMPO and

periodate oxidation (Structure 7) a greater increased in strength was observed

than after TEMPO oxidation alone. The handsheets also showed an increase in

density, Young´s modulus and wet tensile strength. On the other hand, after

TEMPO, periodate oxidation and borohydride reduction (Structure 8), an even

greater densification was observed, which agrees well with the WRV and the

SEM images, but the strain-at-break decreased. At the same degree of

36

modification, similar XRD patterns were obtained for the modified fibres.

However, only dialcohol cellulose (Structure 3) showed an increase in strain-at-

break.

(1) Ref (2) IO

4

(3) IO4+BH

4

(4) IO4+NH

2OH/BH

4

(5) IO4+ClO

2

(6) TEMPO (7) TEMPO+IO4

(8) TEMPO+IO4+BH4

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.50

10

20

30

40

50

60

70

80 (8)

(7)

(5)

(6)

(4)

(3)(2)

Ten

sile

str

ess

(MP

a)

Tensile strain (%)

(1)

Figure 16. Representative stress–strain curves of handsheets made of the

differently modified fibres.

37

Table 3. Physical and mechanical properties of the chemically modified fibres.

Sample number Density

(kg/m3)

Young´s

modulus (GPa)

Wet tensile

strength (MPa)

1:Ref 585 2.1 ± 0.1 -*

2:IO4 581 2.2 ± 0.1 8.4 ± 0.1

3:IO4+BH4 720 3.5 ± 0.1 -*

4:IO4+NH2OH/BH4 650 2.7 ± 0.1 2.8 ± 0.1

5:IO4+ClO2 971 9.6 ± 0.3 0.7 ± 0.1

6:TEMPO 767 4.8 ± 0.3 0.4 ± 0.1

7:TEMPO+IO4 830 6.4 ± 0.6 16.0 ± 0.3

8:TEMPO+IO4+BH4 914 6.8 ± 0.6 0.9 ± 0.2

*Too weak to be measured

b. Effect of chemical functionality on the mechanical and barrier

properties of all-cellulose nanocomposites (Paper III)

As a continuation of the study of handsheets made from chemically modified

fibres, the degree of modification was further increased to study in greater detail

the effect that the shell of modified cellulose has on the properties of the

materials. The fibres were first chemically modified and homogenized to CNFs,

and finally films were prepared from these CNFs and their mechanical properties

were evaluated. In films of nanocellulose, the effects of the macroscopic network

structure and density are of little importance.

The degree of modification was determined by the total carbonyl content and

charge density listed in Table 4. After periodate oxidation (Structure 2), 3.2

mmol of aldehydes/g fibre were introduced, which means that about 27% of the

glucose units were cleaved after the treatment. The degree of modification was

kept at the same level when TEMPO and periodate oxidation (Structure 7) were

combined and no aldehydes were found after further modification. The total

38

charge density decreased after periodate oxidation (Structure 2), and after further

reduction (Structure 3) or reductive amination (Structure 4) and, as already

suggested, this was probably related to the oxidation of hemicelluloses that were

lost during the washing procedure. The total charge density after chlorite

oxidation (Structure 5) indicated that all the aldehydes were oxidised to

carboxylic acids. On the other hand, 964 µeq/g of carboxylic acids were

produced after TEMPO oxidation (Structure 6), but the charge decreased after

periodate oxidation (Structure 7) and decreased even further with borohydride

modification (Structure 8).

Once the fibres were modified and the degree of modification had been

determined, the fibres were homogenized. The surface charge density of the

fibrils showed that the low-charge CNFs had some aggregation, while highly

charged fibrils had values similar to the total charge density. Fall et al. (2011)92

observed the same trend where different charge densities were introduced in

cellulose fibres as pre-treatments before homogenization.

39

Table 4. Carbonyl content and total charge density (determined on fibres) and

surface charge density (after homogenization, i.e. on CNFs ) after chemical

modification. The figures in parentheses show the degree of oxidation.

Sample

Carbonyl

content

(mmol/g)

Charge

density

(µeq/g)

Surface charge (after

homogenization)

(µeq/g)

1:Ref 0.03 ± 0.01 60 ± 3 28 ± 1

2:IO4 3.17 ± 0.09

(27%) 45 ± 4 13 ± 1

3:IO4+BH4 0* 36 ± 1 22 ± 2

4:IO4+NH2OH/BH4 0* 47 ± 2 9 ± 1

5:IO4+ClO2 0* 3177±15 3090 ± 45

6:TEMPO 0* 964 ± 59 802 ± 56

7:TEMPO+IO4 3.19 ± 0.08

(27%) 872 ± 34 739 ± 2

8:TEMPO+IO4+BH4 0* 785 ± 21 531 ± 2

9: TEMPO+IO4+ClO2 0* 3556 ± 51 3406 ± 84

*Below the detection limit of the method

As shown in Figure 17 and Table 5 films prepared from non-modified CNFs had

a tensile strength of 267 MPa and a strain-at-break of 7.5%. This result was very

similar to that of the films made of fibrils prepared after TEMPO oxidation

(Structure 6). It should be noted that to be able to homogenize non-modified

fibres, the concentration could not be significantly higher than 0.2 wt%, and that

more fibril aggregates were observed. TEMPO oxidation, on the other hand,

allowed a concentration of 1.8 wt% during homogenization to prepare a

colloidally stable fibril dispersion. After periodate oxidation (Structure 2) a

tensile strength of 168 MPa and a strain-at-break of 3.5% were obtained, i.e. a

weaker and more brittle film than the non-modified film. This is a typical effect

of covalent crosslinks, in this case formed between aldehydes and alcohols22,93.

40

Films produced after borohydride reduction (Structure 3) had a tensile strength

of 126 MPa and a strain-at-break of 13.6%, and were thus significantly more

ductile than the reference, which is presumably because dialcohol cellulose

facilitates slippage between individual fibrils94. The properties of films prepared

after reductive amination with hydroxylamine (Structure 4) had mechanical

properties inferior to those of the rest of the films perhaps because of a lower

degree of fibrillation. Nevertheless, from these results, this pre-treatment does

not seem appropriate for the production of mechanically high-performing CNFs.

When TEMPO oxidation was combined with periodate oxidation (Structure 7),

the tensile strength of the films was significantly higher than that of films

produced with periodate oxidation alone (Structure 2), but the strain-at-break

decreased to an even greater extent.

The combination of TEMPO, periodate oxidation and borohydride reduction

(Sample 8) gave stronger films, but the strain-at-break was less than that with

dialcohol cellulose (Structure 3), i.e. the introduction of charges created less

ductile films. Dialcohol cellulose was very ductile, while dicarboxylic acid

cellulose was very stiff. Hydrolysis leading to a decrease in the DP of the

cellulose might occur at higher degrees of modification when dicarboxylic acids

are introduced. Such a hydrolysis could explain the low strength of di- and

tricarboxylic acid cellulose, and also the decrease in particle size of the fibrils.

By comparing the results obtained from handsheets of modified fibres and CNF

films, it can be seen that CNF films have a higher density, with considerably

higher strain-at-break, which is also due to the higher degree of modification. An

interesting observation was that handsheets prepared from dicarboxylic acid

cellulose (Structure 5), fibres had a Young´s modulus of 9.6 GPa whereas the

films prepared with a fibrils with higher degree of oxidation had a modulus of

only 8.3 GPa, indicating that a lower degree of modification during the

preparation of dicarboxylic acid cellulose (Structure 5) was more beneficial for

the properties of the materials.

41

0 2 4 6 8 10 12 14 160

50

100

150

200

250

(9)

(5)

(6)

(8)

(7)

(4)

(3)

(2)

Ten

sile

str

ess

(MP

a)

Tensile strain (%)

(1) Ref

(2) IO4

(3) IO4+BH

4

(4) IO4+NH2OH/BH4

(5) IO4+ClO

2

(6) TEMPO

(7) TEMPO+IO4

(8) TEMPO+IO4+BH

4

(9) TEMPO+IO4+ClO

2

(1)

Figure 17. Representative stress-strain curve of films prepared from chemically

modified CNFs. Quantitative details are given in Table 5.

Table 5. Mechanical properties of the different CNF films.

Sample Density

(Kg/m3)

Tensile

strength

(MPa)

Strain-

at-break

(%)

Young´s

modulus

(GPa)

1:Reference 1481 267 ± 1 7.5 ± 1.1 9.8 ± 0.9

2:IO4 1479 168 ± 18 3.5 ± 0.8 7.7 ± 0.9

3:IO4+BH4 1482 126 ± 6 13.6 ± 1 7.4 ± 0.2

4:IO4+NH2OH/BH4 1409 85 ± 7 1.6 ± 0.3 6.4 ± 0.8

5:IO4+ClO2 1478 125 ± 10 2.0 ± 0.1 8.3 ± 0.3

6:TEMPO 1478 253 ± 24 9.0 ± 0.2 8.1 ± 0.9

7:TEMPO+IO4 1466 210 ± 12 2.4 ± 0.3 11.3 ± 1.3

8:TEMPO+IO4+BH4 1471 173 ± 13 9.3 + 1.8 8.2 + 0.7

9: TEMPO+IO4+ClO2 1471 159 ± 10 2.9 ± 0.4 8.6 ± 0.4

42

Since cellulose is known to interact readily with moisture, which affects the

mechanical properties of the films, it was interesting to study the effect of

moisture on the properties of the films made from modified CNFs, and RH-

controlled DMA measurements were therefore made. The results are shown in

Figure 18. As can be seen, films made from dialdehyde cellulose (Structure 2)

and combined with carboxylic acids (Structure 8) softened at 40% RH; the latter

showing a more pronounced decrease in modulus than the films based solely on

dialcohol cellulose. This effect could probably, be explained by the presence of

carboxylic acids which are known to swell and soften the material with

increasing RH. Films prepared after reductive amination (Structure 4) showed a

constant storage modulus up to 75% RH, and also showed that, even in presence

of primary alcohols the overall chemical environment influences the behaviour

of the film.

It is well known that aldehydes produced after periodate oxidation are able to

form crosslinks with adjacent hydroxyl groups. These crosslinks increase the wet

strength of handsheets and films and, as observed in Figure 18, they are also able

to limit the change in storage modulus of the films with increasing RH22,93. This

was observed even after TEMPO and periodate oxidation (Structure 7).

However, although a small decrease in modulus was observed in films made of

TEMPO oxidised fibrils at low RH, but significant decrease in modulus was

observed at around 70% RH. In order to obtain a better understanding of the

results from DMA measurements, the moisture sorption in the films was studied

by increasing the RH from 20 to 90% (Figure 18b). After periodate oxidation

(Structure 2) and TEMPO combined with periodate oxidation (Structure 7), both

materials showed a low increase in mass up to high RHs, although TEMPO

combined with periodate oxidation (Structure 7) showed a sudden change in

mass at RHs higher than 90%. After reductive amination (Structure 7) the

increase in mass was low up to a RH of 70%. This behaviour was completely

different from that of dialcohol cellulose (Structure 2), which showed a greater

increase in moisture sorption with increasing RH. Moreover, after TEMPO,

43

periodate oxidation and borohydride reduction (Structure 8) the increase in mass

was even higher, probably due to the presence of carboxylic acids. Films made

of TEMPO-oxidised CNFs (Structure 6) also increased in mass more than

dialcohol cellulose (Structure 2) at high RH. Di- and tri-carboxylic acid

(Structure 5 and 9) showed a change in mass of 18 and 22%, respectively, which

is presumably due to a combination of effects, the presence of carboxylic acids

and the lower crystallinity following periodate oxidation.

50 100 150 200 250

0

2

4

6

8

10

12

14

16

Time (min)

Sto

rage

Mod

ulus

(G

Pa)

a)

0

15

30

45

60

75

90

(1) Ref (2) IO4

(3) IO4 +BH4

(4) IO4 +NH2OH/BH4

(5) IO4 +ClO2

(6) TEMPO (7) TEMPO+IO4

(8) TEMPO+IO4+BH4

(9) TEMPO+IO4 +ClO2 R

elat

ive

Hum

idity

(%

)

50 100 150 200 250

0

5

10

15

20

25

Time (min)

Rel

ativ

e m

oist

ure

upta

ke (

%) b)

0

15

30

45

60

75

90

Rel

ativ

e H

umid

ity (

%)

Figure 18. a) Storage modulus and b) relative increase in moisture content of

CNF films from differently modified fibrils plotted as a function of time, during

which the relative humidity increased stepwise from 30% to 90% RH at a rate of

0.4%-units per minute.

44

Due to the different mechanical properties of the CNF films, and the fact that

cellulose is known to form a good oxygen barrier95, it was also of interest to

measure the oxygen permeability of the different films. At 23 °C and 50% RH,

the reference films had an oxygen permeability of 0.7 mL⋅µm/(m2⋅24h⋅kPa),

which agrees well with previous studies95. The permeability decreased, i.e. a

better barrier was achieved, when functional groups were introduced in the

cellulose backbone. However, at high RH the results varied greatly between the

different films and the permeability tended to increase. Only dialdehyde

cellulose (Structure 2) maintained a low oxygen permeability at 80%, which

agrees well with the low change in storage modulus and mass change detected

with DMA and moisture sorption experiments, respectively. In Table 6 it also is

also evident that a combination of TEMPO and periodate oxidation (Structure 7)

negatively affects the oxygen permeability at high RH. Somewhat surprisingly,

despite their poor mechanical properties, films made from CNFs that had been

subjected to reductive amination had an oxygen permeability that was fairly low

even at high RH.

Table 6. Oxygen permeability of films made of chemically modified CNFs

Sample 23 °C, 50 % RH

(mL ⋅⋅⋅⋅µm/(m2⋅⋅⋅⋅24h⋅⋅⋅⋅kPa))

23 °C, 80 % RH

(mL ⋅⋅⋅⋅µm/(m2⋅⋅⋅⋅24h⋅⋅⋅⋅kPa))

1:Ref 0.7 13.1

2:IO4 0.3 0.3

3:IO4+BH4 0.4 11.4

4:IO4+NH2OH /BH4 0.4 4.3

5:IO4+ClO2 0.5 10.2

6:TEMPO 0.4 11.9

7:TEMPO+IO4 0.4 7.7

8:TEMPO+IO4+BH4 0.2 9.5

9: TEMPO+IO4+ClO2 0.5 14.0

45

4.2. Material applications of modified fibres and fibrils (Papers IV

and V)

4.2.1. Chemically modified fibrils as paper strength additives

(Paper IV)

Different cellulose micro- and nanofibrils were used as paper strength additives.

Carboxymethylated CNFs and microfibrillated cellulose (MFC) from unbleached

kraft fibres were compared with carboxymethylated CNFs that were further

modified, either by periodate oxidation or dopamine grafting. Prior to studying

the mechanical properties of the prepared sheets, the retention of the fibrillated

cellulose was determined. As shown in Figure 19, kraft MFC had a good

retention, probably due to the larger particle size and lower charge density of

these materials, whereas carboxymethylated CNFs showed a poor retention,

which could be explained by their smaller size as well as a greater repulsion

between the fibres and the fibrils. Dopamine-grafted CNFs showed a better

retention than with the non-further modified CNFs (i.e. carboxymethylated

CNFs), but, interestingly, the periodate-oxidised CNFs were retained best by the

fibres; probably due to a greater aggregation of these fibrils. In general, the use

of a retention aid (pDADMAC or PVAm) did not increase the retention of the

fibrils in this study, despite considerable efforts.

46

2 5 150

2

4

6

8

10

12

14

16

2 5 150

2

4

6

8

10

12

14

16

Carboxymethylated CNF

Periodate CNF

Dopamine CNF

Kraft CNF F

ibril

con

tent

(%

)

Added fibrils (%)

a) No retention aid

Carboxymethylated CNF + PVAm

Periodate CNF + PVAm

Dopamine CNF + PVAm

Kraft MFC + PVAm

Periodate CNF + PDADMAC

Fib

ril c

onte

nt (

%)

Added fibrils (%)

b) Retention aid

Figure 19. Retention of the differently modified fibrils during papermaking with

or without the use of retention aids.

The mechanical properties of handsheets are very dependent on their density and

it has been reported that the addition of CNFs increases the density of

handsheets96,97. In order to separate the effect of an increase in density on the

improvement in the strength of the handsheets, the fibres were beaten from 1000

to 10 000 revolutions in a PFI mill. As shown in Figure 20, the strength increase

after the addition of kraft MFC or carboxymethylated MCF it can be concluded

that is probably due to the increased density since the same strength could be

achieved by beating of the fibres to the same density. This is in agreement with

Brodin et al. (2014)98 who concluded, after comparing different studies regarding

CNFs and MFC as strength additives, that the addition of fibrils as additives

would give the same effect as beating. However, the present results show that

after the addition of periodate-oxidised CNFs or dopamine-grafted CNFs, the

strength improvement is considerably higher than that achieved by beating. The

addition of 2 wt% of periodate-oxidised CNFs gave a greater improvement than

beating the fibres for 10 000 revolutions, and at a lower density. Even though the

retention of the fibrils was not affected by the addition of a retention aid, it can

be concluded that the retention aid did indeed have an influence on the

association of the fibrils upon adsorption, since the addition of PVAm and

periodate-oxidised CNFs improved the strength of the handsheets even further

47

than the CNFs alone, an effect that is believed to be due to the formation of

imine groups between the aldehydes and the primary amines of PVAm.

However, in contrast to what was expected from such imine formation, the

combination of pDADMAC and periodate-oxidised CNFs resulted in an equally

good or even greater improvement, probably due to fibre tensile failure which

ultimately becomes the limiting factor for the paper strength.

600 650 700 750 800 850

40

60

80

100

120

140

600 650 700 750 800 850

40

60

80

100

120

140

10000 rev

8000 rev

6000 rev4000 rev

2000 rev

1000 rev

10000 rev8000 rev

6000 rev4000 rev

2000 rev

Ten

sile

inde

x (N

m/g

)

Density (Kg/m3)

Beaten fibres Reference Carboxymethylated NFC Periodate CNF Dopamine CNF Kraft MFC

a) No retention aid

1000 rev

Ten

sile

inde

x (N

m/g

)Density (Kg/m3)

Beaten fibres Reference Reference+ pDADMAC Carboxymethylated CNF Periodate CNF Periodate CNF+ pDADMAC Dopamine CNF Kraft MFC

b) Retention aid

Figure 20. Tensile strength index of handsheets with different CNFs: a) without

and b) with retention aid.

The high wet strength obtained after addition of periodate-oxidised CNFs

(Figure 21) supports the presence of covalent crosslinks. It should also be noted

that, despite the high dry strength achieved after addition of pDADMAC and

periodate-oxidised CNFs, the wet strength of these sheets was lower than that

achieved with a combination of PVAm and periodate oxidised CNF, i.e. a

combination that should be able to form imine bonds. This result thus further

indicates not only that crosslinks form between aldehydes and hydroxyl groups

but also that imine formation is highly likely. Despite the reported ability of

dopamine to give wet adhesion to films69, less improvement in wet strength was

achieved.

48

0 2 4 6 8 10 12 14 160

5

10

15

20

25

30

35

40

45

0 2 4 6 8 10 12 14 160

5

10

15

20

25

30

35

40

45

Reference Reference + pDADMAC Carboxymethylated CNF Periodate CNF Dopamine CNF Kraft MFC Periodate CNF + pDADMAC

Wet

tens

ile in

dex

(Nm

/g)

Fibril content (%)

b) With retention aid Reference Carboxymethylated CNF Periodate CNF Dopamine CNF Kraft CNF

Wet

tens

ile in

dex

(Nm

/g)

Fibril content (%)

a) No retention aid

Figure 21. Wet strength of handsheets after addition of differently modified

CNFs, (a) with and (b) without retention aids.

4.2.2. Light weight materials from cellulose fibres (Paper V)

There is great interest in light weight cellulose materials due the strength, ability

to tolerate harsh chemical conditions and renewability of cellulose. In this work a

new type of 3D structure from chemically modified fibres was developed. The

fibres were first periodate oxidized, followed by chlorite oxidation and a second

periodate oxidation. The later oxidation was performed to allow the fibres to

self-assemble into a shape determined by the vessel in which the fibres were

treated. This self-assembly/shrinking of cellulose in the presence of sodium

periodate was already observed by Jackson et al.99, but the mechanism behind

this self-assembly is still not fully understood.

The fibres were first characterised in terms of their total aldehyde content and

total charge density. As shown in Table 7, the first periodate oxidation was

performed to a relatively low degrees of modification and only 1.35 mmol/g of

aldehyde groups were formed. This was to avoid too severe fibrillation during

the subsequent chlorite oxidation. After the chlorite oxidation, about 0.5 mmol/g

aldehydes remained and after the second periodate oxidation, a total of 4.4

mmol/g of aldehydes were found (i.e. in total about 5.2 mmol glucose units per

gram sample was modified). The fibres were optionally subjected to Ultra-turrax

49

mixing to allow fibrillation, which, as observed in Table 7, did not affect the

carbonyl content to any significant extent. After chlorite oxidation, there was a

large increase in charge density as a result of the introduction of carboxylic acids

(Table 7). However, after the second periodate oxidation, the charge density

decreased from ~870 to ~580 µeq/g.

Table 7. Aldehyde content and total charge density after the different fibre

treatments.

Sample Carbonyl content

(mmol/g)

Total charge

density (µeq/g)

Reference 0.03 ± 0.01 60 ± 3

IO4 (30 min) 1.35 ± 0.24 48 ± 4

IO4 (30 min)+ClO2 (20 h) 0.54 ± 0.03 870 ± 5

IO4 (30 min)+ClO2 (20 h)

+ IO4 (16 h)

0 min mixing 4.4 ± 0.3 578 ± 6

10 min mixing 4.3 ± 0.1 586 ± 10

20 min mixing 4.4 ± 0.1 571 ± 13

30 min mixing 4.4 ± 0.2 594 ± 9

In order to study the morphology of the fibres after chemical modification and

drying to produce light weight structures, SEM images were taken at different

levels of magnification (Figure 22). Despite the mechanical mixing, where some

fibrillation is expected to take place during prolonged mixing, the overall

structure of the fibres was maintained. As can be seen, the fibres are long with a

diameter of the order of a few tens of micrometres (left and middle column of

Figure 22) and, at higher magnification, the subfibrillar structure can be observed

(right column of Figure 22). Figure 22 shows that, the fibre wall did not collapse

and the fibre wall seems very porous. However, somewhat surprisingly, a

specific surface area of only 1.2 m2/g was achieved by BET method, possibly

indicating that the pores seen in the images are closed and not available for gas

adsorption. Also, as the Ultra-Turrax mixing time increased, the fibre networks

50

became denser (Table 9). The fibre networks formed 3D structures without

collapse or significant shrinkage during drying, and it was possible to obtain

light weight fibre-based materials with a porosity between 96 and 94% (i.e. a

density between 54 and 82 kg/cm3) as displayed in Table 8.

Figure 22. SEM images of the modified cellulose fibres forming light weight

materials. a1-a3) No Ultra-turrax mixing; b1-b3) 10 min; c1-c3) 20 min and d1-d3)

30 min of mixing. The left-hand column shows low-magnification images (scale

bar equal to 1000 mm), the middle column images have a magnification of 90

(scale bar equal to 500 µm), and the right-hand column images have a high

magnification (scale bar equal to 10 µm).

51

Table 8. Density and porosity of low-density fibre network materials. Values are

means given with standard deviation based on two measurements.

Mixing time (min) Density (kg/m3) Porosity (%)

0 54 ± 2 96 ± 1

10 65 ± 1 96 ± 0

20 67 ± 1 95 ± 1

30 82 ± 1 95± 1

Compressive stress-strain curves of the different samples are presented in Figure

23. The fibre networks behaved as porous materials with open cells in both the

dry and wet state100 and three different regions were observed. The first region

corresponds to the elastic behaviour of the materials in the low strain region

(≤10%). The second region where 10% ≤ ɛ ≤ 60%, the network continues to

collapse under a constant stress until the opposite sides of the cells meet

(densification strain), and in the third region, when the stress rises steeply101.

Even though the same regions were observed in the wet state, the stress was

much lower, as shown in Figure 23. The decrease in stress suggests that the fibre

network is plasticized by water. A higher mechanical strength was observed with

increasing mixing time, in both the dry and wet states. This effect can be partly

explained by a higher joint strength due to partial fibrillation, and by the increase

in density which increases the contact area between the fibres. Interestingly,

despite the mixing-induced increase in strength, the modulus of the fibre

networks was not affected by the Ultra-turrax mixing. The increase in strength of

the material is due not only to the fibre-fibre joint but also to the properties of the

modified fibres constituting the material. This effect was observed in Paper I and

by Larsson et al. (2014)32, as the improvement in mechanical properties occurs

not only in the fibre network but also in the individual fibres.

52

0 20 40 60 800

250

500

750

1000

1250

1500

1750

0 20 40 60 800

20

40

60

80a)a)

Com

pres

sive

str

ess

(kP

a)

Compressive strain (%)

0 min 10 min 20 min 30 min

a)

Com

pres

sive

str

ess

(kP

a)

Compressive strain (%)

0 min 10 min 20 min 30 min

b)

Figure 23. Effect of mixing time on stress-strain behaviour of fibre networks in:

(a) the dry state and (b) the wet state.

The properties of these materials surpass the properties of other fibre-based light

weight materials. Liu et al.102 reported a compression strength of 14 kPa in the

dry state (no wet strength was reported). Sehaqui et al. (2010)100 prepared foams

from CNFs of different concentrations. Depending on the concentration used,

the density and strength ranged from 7 to 101 kg/m3 and from 1 to 7 MPa,

respectively. Only at high CNF concentrations (> 4.2 wt%) were the foams

stronger than the present fibre networks. However, when low concentrations of

CNFs were used, the fibre networks showed mechanical properties in the same

range. The fibre networks in this work were wet resilient and did not require the

use of freeze-drying to maintain the structure and avoid shrinkage and collapse.

The wet strength, which is due to crosslinks formed between the aldehydes and

hydroxyls groups, also introduced shape recovery properties in water for the

fibre network (Figure 24), i.e. it recovered its original shape after the

compressive load had been removed. Furthermore, it was observed that the

materials, regardless of the mixing time, were able to absorb 13 times their own

weight in water. It was also shown during the characterization of chemically

modified fibres (Paper II) that fibres with a high charge density also had a high

WVR.

53

Figure 24. Shape recovery properties of the fibre networks.

54

5. CONCLUSIONS

The aim of the work described in this thesis was to chemically modify cellulose-

rich fibres and fibrils under heterogeneous and aqueous conditions to understand

how the different functional groups in the cellulose backbone affect the

properties of materials made from such cellulose-rich raw materials. It was also

desired to gain an understanding of the effect that different lignocellulosic

components have on the chemical reactions.

In general it can be concluded that:

a. Fibres with different lignocellulosic compositions reacted differently

towards the same type of chemical modification (in this case periodate

oxidation and borohydride reduction). This can be explained by the

content and availability of vicinal diols and reactions between the

chemicals with units on the lignin. Despite a lower reactivity towards

modification of unbleached kraft fibres showed a great improvement in

the compression strength index.

b. When the surfaces of the cellulose nanofibrils were modified to

introduce different functional groups, to create a shell of modified

amorphous cellulose surrounding a core of crystalline cellulose, the

functional groups largely influence the behaviour and properties of the

fibres, CNFs and materials made thereof. The introduction of aldehydes

led to materials that adsorbed less moisture, but an increased brittleness

was observed. The presence of primary alcohols increased the ductility

of the materials, but an increase in moisture sensitivity was also

observed. The presence of primary alcohols attached to amines did not

however give the same effect. Finally, the introduction of carboxylic

acid led to stiffer materials, but an increase in moisture sensitivity was

also observed in this material.

55

c. The modifications enabled the creation of all-cellulose composites and

created a toolbox that can be used to tune the properties of cellulose-

based materials, or as a platform for further functionalization.

d. Chemically modified CNFs were used as paper strength additives.

Depending on the modification, significant differences in the retention

of the CNFs were observed. It was also clearly shown that the chemical

structure of the modified fibrils was the main factor determining the

improvement in the mechanical properties of the fibril-containing

handsheets, in both the dry and wet state.

e. Light weight materials from chemically modified fibres were

developed. The main feature of this material was that no freeze drying

or supercritical drying was needed to attain a dry structure with

maintained dimensions. The fibre networks showed excellent strength

and stiffness comparable with the properties of materials prepared from

CNFs. Moreover, the materials displayed wet strength, shape recovery

capacity and good water absorbency.

56

6. ACKNOWLEDGEMENTS

I would like to thank my supervisors Lars Wågberg and Per Larsson. Lars, thank

you for giving me the opportunity to work in such a great group, your optimism,

enthusiasm and support makes a great environment for the development of new

ideas. Per, thank you for always being available to answer all kinds of questions

and doubts as well as your enthusiasm for discussion. I truly believe that you

deserve that bottle of champagne (even though I actually won the bet)! I hope

that we can find a way to continue working together in the future. Both of you

have helped me to achieve this goal and grow as a scientist.

BiMaC Innovation is acknowledged for financial support.

Thanks also to Åsa Blademo from RISE Bioeconomy for the help with the

different instruments and availability for discussion.

I would also like to thank my colleagues at FPT for scientific discussions and all

the good times together, special thanks to Maryam, Andrew, Rosana, Ramiro,

Martin, Mia, Daniel, Torbjörn, Carmen, Linda and Emily. I wish to particularly

thank Johan and Rebecca for great collaboration and work on our projects, also

for their friendship and help. Thank you Oruç for the countless times you help

me in the lab, I really appreciate your patience.

To my great friend, Danila, my time at KTH would not have been the same

without you on the other side of the corridor. Thank you so much for your

friendship, thank you for always listening, thank you for being such an authentic

person. Obrigada chica!!

Thanks also to Marcel, Airam and Ziky for your advice and support in spite of

the distance.

57

To Petri, my love, thanks for cheering me up during the bad days and for the

great brain storm sessions. Most important of all thank you for building a home

for the three of us! You have taught me that dreams come true if we have the

courage and the willingness to pursue them. There are not enough words to

describe how much I love you, how lucky I feel to share my life with you and

how immensely happy you make me.

Finalmente quisiera agradecer a mi familia. A mi mamá por apoyar cada uno de

mis sueños, por enseñarme que nada es más importante que ser feliz y que la

felicidad está en las pequeñas cosas de la vida. Tu amor me mantiene calientica

incluso a miles de kilómetros de distancia. A mi papá por enseñarme que el valor

del trabajo y la disciplina son claves para alcanzar el éxito y por todas nuestras

conversaciones terapéuticas. A mi hermana, porque siempre has sido capaz de

sacarme las más grandes carcajadas incluso en mis momentos más difíciles. Los

amo infinitamente y son mi más grande fuente de inspiración.

58

7. REFERENCES

(1) Klemm, D.; Heublein, B.; Fink, H. P.; Bohn, A. Cellulose: Fascinating

Biopolymer and Sustainable Raw Material. Angew. Chemie - Int. Ed.

2005, 44 (22), 3358–3393.

(2) Thomas, S.; Paul, S. A.; Pothan, L. A.; Deepa, B. Natural Fibres:

Structure, Properties and Applications. In Cellulose Fibers: Bio- and

Nano- Polymer Composites; Kalia, S., Kaith, B. S., Kaur, I., Eds.;

Springer: New York, 2011; pp 3–42.

(3) The Structure of Wood. In Wood Chemistry: Fundamentals and

Applications; Sjöström, E., Ed.; Academic Press: New York, 1993; pp 1–

20.

(4) Gibson, L. J. The Hierarchical Structure and Mechanics of Plant

Materials. J R Soc Interface 2012, 9 (76), 2749–2766.

(5) Klinga, N.; Höglund, H.; Sandberg, C. Energy Efficient High Quality

CTMP for Paperboard. J. Pulp Pap. Sci. 2008, 34 (2), 98–106.

(6) Gierer, J. Chemical Aspects of Kraft Pulping. Wood Sci. Technol. 1980,

14, 241–266.

(7) Kirkpatrick, N. Biological Bleaching of Wood Pulps - A Viable

Chlorine-Free Bleaching Technology? Water Sci. Technol. 1991, 24 (3–

4), 75–79.

(8) Sixta, H.; Potthanst, A.; Krotschek, A. W. Chemical Pulping Processes.

In Handbook of Pulp; Sixta, H., Ed.; Wiley-VCH Verlag GmbH & Co.:

Weinheim, 2006; pp 325–365.

(9) Wickholm, K.; Larsson, P. T.; Iversen, T. Assignment of Non-

Crystalline Forms in Cellulose I by CP/MAS 13C NMR Spectroscopy.

Carbohydr. Res. 1998, 312 (3), 123–129.

(10) Isogai, A.; Saito, T.; Fukuzumi, H. TEMPO-Oxidized Cellulose

Nanofibers. Nanoscale 2011, 3 (1), 71–85.

(11) Turbak, A. F.; Snyder, F. W.; Sandberg, K. R. Microfibrillated Cellulose,

59

a New Cellulose Product: Properties, Uses, and Commercial Potential. J.

Appl. Polym. Sci. Appl. Polym. Symp. 1983, 37, 815–827.

(12) Wågberg, L.; Decher, G.; Norgren, M.; Lindström, T.; Ankerfors, M.;

Axnäs, K. The Build-up of Polyelectrolyte Multilayers of

Microfibrillated Cellulose and Cationic Polyelectrolytes. Langmuir 2008,

24 (3), 784–795.

(13) Saito, T.; Hirota, M.; Tamura, N.; Kimura, S.; Fukuzumi, H.; Heux, L.;

Isogai, A. Individualization of Nano-Sized Plant Cellulose Fibrils by

Direct Surface Carboxylation Using TEMPO Catalyst under Neutral

Conditions. Biomacromolecules 2009, 10 (7), 1992–1996.

(14) Saito, T.; Nishiyama, Y.; Putaux, J. L.; Vignon, M.; Isogai, A.

Homogeneous Suspensions of Individualized Microfibrils from TEMPO-

Catalyzed Oxidation of Native Cellulose. Biomacromolecules 2006, 7

(6), 1687–1691.

(15) Aulin, C.; Ström, G. Multilayered Alkyd Resin/nanocellulose Coatings

for Use in Renewable Packaging Solutions with a High Level of

Moisture Resistance. Ind. Eng. Chem. Res. 2013, 52 (7), 2582–2589.

(16) Olszewska, A.; Eronen, P.; Johansson, L. S.; Malho, J. M.; Ankerfors,

M.; Lindström, T.; Ruokolainen, J.; Laine, J.; Österberg, M. The

Behaviour of Cationic Nanofibrillar Cellulose in Aqueous Media.

Cellulose 2011, 18 (5), 1213–1226.

(17) Henriksson, M.; Henriksson, G.; Berglund, L. A.; Lindström, T. An

Environmentally Friendly Method for Enzyme-Assisted Preparation of

Microfibrillated Cellulose (MFC) Nanofibers. Eur. Polym. J. 2007, 43

(8), 3434–3441.

(18) Pääkko, M.; Ankerfors, M.; Kosonen, H.; Nykänen, A.; Ahola, S.;

Österberg, M.; Ruokolainen, J.; Laine, J.; Larsson, P. T.; Ikkala, O.;

Lindström, T. Enzymatic Hydrolysis Combined with Mechanical

Shearing and High-Pressure Homogenization for Nanoscale Cellulose

Fibrils and Strong Gels. Biomacromolecules 2007, 8 (6), 1934–1941.

60

(19) Moon, R. J. McGraw-Hill Yearbook in Science and Technology;

McGraw-Hill, 2008; pp 225–228.

(20) Nishino, T.; Matsuda, I.; Hirao, K. All-Cellulose Composite.

Macromolecules 2004, 37 (20), 7683–7687.

(21) Gindl, W.; Keckes, J. All-Cellulose Nanocomposite. Polymer (Guildf).

2005, 46 (23), 10221–10225.

(22) Larsson, P. A.; Kochumalayil, J. J.; Wågberg, L. Oxygen and Water

Vapour Barrier Films with Low Moisture Sensitivity Fabricated from

Self-Cross-Linking Fibrillated Cellulose. In The pulp and paper research

fundamental research society, 15 th fundamental research symposium,

Lancashire, UK; 2013; pp 851–866.

(23) Larsson, P. A.; Wågberg, L. Towards Natural-Fibre-Based

Thermoplastic Films Produced by Conventional Papermaking. Green

Chem. 2016, 18, 3324–3333.

(24) Huber, T.; Müssig, J.; Curnow, O.; Pang, S.; Bickerton, S.; Staiger, M. P.

A Critical Review of All-Cellulose Composites. J. Mater. Sci. 2012, 47

(3), 1171–1186.

(25) Lu, X.; Zhang, M. Q.; Rong, M. Z.; Shi, G.; Yang, G. C. All-Plant Fiber

Composites. I: Unidirectional Sisal Fiber Reinforced Benzylated Wood.

Polym. Compos. 2002, 23 (4), 624–633.

(26) Lu, X.; Zhang, M. Q.; Rong, M. Z.; Shi, G.; Yang, G. C. Self-Reinforced

Melt Processable Composites of Sisal. Compos. Sci. Technol. 2003, 63

(2), 177–186.

(27) Gandini, A.; Da Silva Curvelo, A. A.; Pasquini, D.; De Menezes, A. J.

Direct Transformation of Cellulose Fibres into Self-Reinforced

Composites by Partial Oxypropylation. Polymer (Guildf). 2005, 46 (24),

10611–10613.

(28) de Menezes, A. J.; Pasquini, D.; Curvelo, A. A. da S.; Gandini, A. Self-

Reinforced Composites Obtained by the Partial Oxypropylation of

Cellulose Fibers. 2. Effect of Catalyst on the Mechanical and Dynamic

61

Mechanical Properties. Cellulose 2009, 16 (2), 239–246.

(29) de Menezes, A. J.; Pasquini, D.; Curvelo, A. A. da S.; Gandini, A. Self-

Reinforced Composites Obtained by the Partial Oxypropylation of

Cellulose Fibers. 1. Characterization of the Materials Obtained with

Different Types of Fibers. Carbohydr. Polym. 2009, 76 (3), 437–442.

(30) Matsumura, H.; Sugiyama, J.; Glasser, W. G. Cellulosic

Nanocomposites. I. Thermally Deformable Cellulose Hexanoates from

Heterogeneous Reaction. J. Appl. Polym. Sci. 2000, 78 (13), 2242–2253.

(31) Zeronian, S. H.; Hudson, F. L.; Peters, R. H. The Mechanical Properties

of Paper Made from Periodate Oxycellulose Pulp and from the Same

Pulp after Reduction with Borohydride. Tappi 1964, 47 (9), 557–564.

(32) Larsson, P. A.; Berglund, L. A.; Wågberg, L. Highly Ductile Fibres and

Sheets by Core-Shell Structuring of the Cellulose Nanofibrils. Cellulose

2014, 21, 323–333.

(33) Mechanism of Periodate Oxidation. In Periodate Oxidation of Diol and

Other Functional Groups; Dryhurst, G., Ed.; Pergamon Press: Oxford,

1970; pp 24–54.

(34) Potthast, A.; Kostic, M.; Schiehser, S.; Kosma, P.; Rosenau, T. Studies

on Oxidative Modifications of Cellulose in the Periodate System:

Molecular Weight Distribution and Carbonyl Group Profiles.

Holzforschung 2007, 61 (6), 662–667.

(35) Kim, U. J.; Kuga, S.; Wada, M.; Okano, T.; Kondo, T. Periodate

Oxidation of Crystalline Cellulose. Biomacromolecules 2000, 1 (3), 488–

492.

(36) Varma, A. J.; Chavan, V. B. A Study of Crystallinity Changes in

Oxidised Celluloses. Polym. Degrad. Stab. 1995, 49 (2), 245–250.

(37) Aimin, T.; Hongwei, Z.; Gang, C.; Guohui, X.; Wenzhi, L. Influence of

Ultrasound Treatment on Accessibility and Regioselective Oxidation

Reactivity of Cellulose. Ultrason. Sonochem. 2005, 12 (6), 467–472.

(38) Sirviö, J.; Hyvakko, U.; Liimatainen, H.; Niinimaki, J.; Hormi, O.

62

Periodate Oxidation of Cellulose at Elevated Temperatures Using Metal

Salts as Cellulose Activators. Carbohydr. Polym. 2011, 83 (3), 1293–

1297.

(39) Sirviö, J.; Liimatainen, H.; Niinimäki, J.; Hormi, O. Dialdehyde

Cellulose Microfibers Generated from Wood Pulp by Milling-Induced

Periodate Oxidation. Carbohydr. Polym. 2011, 86 (1), 260–265.

(40) Mehltretter, C. L. Electrochemical Production of Periodate

Oxypolysaccharides. US2713553 A, 1955.

(41) Lister, M. W.; Rosenblum, P. The Oxidation of Nitrite and Iodate Ions

by Hypochlorite Ons. Can. J. Chem. 1961, 39, 1645–1651.

(42) Liimatainen, H.; Sirviö, J.; Pajari, H.; Hormi, O.; Niinimäki, J.

Regeneration and Recycling of Aqueous Periodate Solution in

Dialdehyde Cellulose Production. J. Wood Chem. Technol. 2013, 33,

258–266.

(43) Koprivica, S.; Siller, M.; Hosoya, T.; Roggenstein, W.; Rosenau, T.;

Potthast, A. Regeneration of Aqueous Periodate Solutions by Ozone

Treatment: A Sustainable Approach for Dialdehyde Cellulose

Production. ChemSusChem 2016, 9 (8), 825–833.

(44) Chaikin, S. W.; Brown, W. G. Reduction of Aldehydes, Ketones and

Acid Chlorides by Sodium Borohydride. J. Am. Chem. Soc. 1949, 71 (1),

122–125.

(45) Sihtola, H. Chemical Properties of Modified Celluloses. Die Makromol.

Chemie 1959, 35 (1), 250–265.

(46) Abdel-Magid, A. F.; Carson, K. G.; Harris, B. D.; Maryanoff, C. A.;

Shah, R. D. Reductive Amination of Aldehydes and Ketones with

Sodium Triacetoxyborohydride. Studies on Direct and Indirect Reductive

Amination Procedures. J. Org. Chem. 1996, 61 (11), 3849–3862.

(47) Vishtal, A.; Retulainen, E. Boosting the Extensibility Potential of Fibre

Networks: A Review. BioResources 2014, 9 (4), 7933–7983.

(48) Saito, T.; Okita, Y.; Nge, T. T.; Sugiyama, J.; Isogai, A. TEMPO-

63

Mediated Oxidation of Native Cellulose: Microscopic Analysis of

Fibrous Fractions in the Oxidized Products. Carbohydr. Polym. 2006, 65

(4), 435–440.

(49) Zhao, M.; Li, J.; Mano, E.; Song, Z.; Tschaen, D. M.; Grabowski, E. J.

J.; Reider, P. J. Oxidation of Primary Alcohols to Carboxylic Acids with

Sodium Chlorite Catalyzed by TEMPO and Bleach. J. Org. Chem. 1999,

64 (7), 2564–2566.

(50) Wågberg, L.; Annergren, G. Physicochemical Characterization of

Papermaking Fibres. In Advances in Paper Science and Technology -

11th Fundamental Research Symposium; Baker, C., Ed.; Cambridge,

1997; pp 1–82.

(51) Bledzki, A. K.; Gassan, J. Composites Reinforced with Cellulose Based

Fibres. Prog. Polym. Sci. 1999, 24 (2), 221–274.

(52) Kalia, S.; Dufresne, A.; Cherian, B. M.; Kaith, B. S.; Avérous, L.;

Njuguna, J.; Nassiopoulos, E. Cellulose-Based Bio- and

Nanocomposites: A Review. Int. J. Polym. Sci. 2011, 2011.

(53) Coffey, D. G.; Bell, D. A.; Henderson, A. Food Polysaccharides and

Their Applications. In Food Polysaccharides and Their Applications;

Stephen, A. M., Phillips, G. O., Williams, P., Eds.; Taylor & Francis

Group: Boca Raton, 2006; pp 147–174.

(54) Piergiovanni, L.; Limbo, S. Cellulosic Packaging Materials. In Food

packaging materials; Springer: New York, 2016; pp 23–31.

(55) Li, F.; Mascheroni, E.; Piergiovanni, L. The Potential of Nanocellulose

in the Packaging Field: A Review. Packag. Technol. Sci. 2015, 28, 475–

508.

(56) Novotna, K.; Havelka, P.; Sopuch, T.; Kolarova, K.; Vosmanska, V.;

Lisa, V.; Svorcik, V.; Bacakova, L. Cellulose-Based Materials as

Scaffolds for Tissue Engineering. Cellulose 2013, 20 (5), 2263–2278.

(57) Petersen, N.; Gatenholm, P. Bacterial Cellulose-Based Materials and

Medical Devices: Current State and Perspectives. Appl. Microbiol.

64

Biotechnol. 2011, 91 (5), 1277–1286.

(58) Liang, D.; Hsiao, B. S.; Chu, B. Functional Electrospun Nanofibrous

Scaffolds for Biomedical Applications. Adv. Drug Deliv. Rev. 2007, 59

(14), 1392–1412.

(59) Jabbour, L.; Bongiovanni, R.; Chaussy, D.; Gerbaldi, C.; Beneventi, D.

Cellulose-Based Li-Ion Batteries: A Review. Cellulose 2013, 20 (4),

1523–1545.

(60) Pushparaj, V. L.; Shaijumon, M. M.; Kumar, A.; Murugesan, S.; Ci, L.;

Vajtai, R.; Linhardt, R. J.; Nalamasu, O.; Ajayan, P. M. Flexible Energy

Storage Devices Based on Nanocomposite Paper. Proc. Natl. Acad. Sci.

U. S. A. 2007, 104 (34), 13574–13577.

(61) Nyholm, L.; Nyström, G.; Mihranyan, A.; Strømme, M. Toward Flexible

Polymer and Paper-Based Energy Storage Devices. Adv. Mater. 2011, 23

(33), 3751–3769.

(62) Schmied, F. J.; Teichert, C.; Kappel, L.; Hirn, U.; Bauer, W.; Schennach,

R. What Holds Paper Together: Nanometre Scale Exploration of

Bonding between Paper Fibres. Sci. Rep. 2013, 3, 1–6.

(63) Haslach, H. W. Moisture and Rate-Dependent Mechanical Properties of

Paper: A Review. Mech. Time-Dependent Mater. 2000, 4 (3), 169–210.

(64) Lindström, T.; Wågberg, L.; Larsson, T. On the Nature of Joint Strength

in Paper: A Review of Dry and Wet Strength Resins Used in Paper

Manufacturing. In Advances in Paper Science and Technology: 13th

Fundamental Research Symposium, Cambridge; 2005; pp 457–562.

(65) Page, D. H.; El-Hosseiny, F. The Mechanical Properties of Single Wood

Pulp Fibres. Part VI. Fibril Angle and the Shape of the Stress-Strain

Curve. Fibre Sci. Technol. 1975, 8 (1), 21–31.

(66) Page, D. H. A Theory for the Tensile Strength of Paper. Tappi 1969, 52

(4), 674–681.

(67) Torgnysdotter, A.; Wågberg, L. Tailoring of Fibre/fibre Joints in Order

to Avoid the Negative Impacts of Drying on Paper Properties. Nord.

65

Pulp Pap. Res. J. 2006, 21 (3), 411–418.

(68) Tanaka, R.; Saito, T.; Isogai, A. Cellulose Nanofibrils Prepared from

Softwood Cellulose by TEMPO/NaClO/NaClO2 Systems in Water at pH

4.8 or 6.8. Int. J. Biol. Macromol. 2012, 51 (3), 228–234.

(69) Karabulut, E.; Pettersson, T.; Ankerfors, M.; Wågberg, L. Adhesive

Layer-by-Layer Films of Carboxymethylated Cellulose Nanofibril-

Dopamine Covalent Bioconjugates Inspired by Marine Mussel Threads.

ACS Nano 2012, 6 (6), 4731–4739.

(70) Zhao, H.; Heindel, N. D. Determination of Degree of Substitution of

Formyl Groups in Polyaldehyde Dextran by the Hydroxylamine

Hydrochloride Method. Pharm. Res. 1991, 8 (3), 400–402.

(71) Scalla, A. M.; Katz, S.; Argyropoulos, D. S. Conductometric Titration of

Cellulosic Fibres. In Cellulose and Wood Chemistry and Technology;

Schuerch, Ed.; John Wiley: New York, 1989; p 1457.

(72) Wachernig, H. Comprehensive Ionic Charge Sensing for

Macromolecules and Particles. In Poster presented at: ACS Colloid and

Surface Science Symposium; New York, 2009.

(73) Batchelor, W. J.; Westerlind, B. S. Measurement of Short Span Stress-

Strain Curves of Paper. Nord. Pulp Pap. Res. J. 2003, 18 (1), 44–50.

(74) Christov, L. P.; Prior, B. A. Xylan Removal from Dissolving Pulp Using

Enzymes of Aureobasidium Pullulans. Biotechnol. Lett. 1993, 15 (12),

1269–1274.

(75) Duchesne, I.; Hult, E. L.; Molin, U.; Daniel, G.; Iversen, T.; Lennholm,

H. The Influence of Hemicellulose on Fibril Aggregation of Kraft Pulp

Fibres as Revealed by FE-SEM and CP/MAS 13C-NMR. Cellulose

2001, 8 (2), 103–111.

(76) Palme, A.; Theliander, H.; Brelid, H. Acid Hydrolysis of Cellulosic

Fibres: Comparison of Bleached Kraft Pulp, Dissolving Pulps and Cotton

Textile Cellulose. Carbohydr. Polym. 2016, 136, 1281–1287.

(77) Adler, E.; Hernestam, S. Estimation of Phenolic Hydroxyl Groups in

66

Lignin. I. Periodate Oxidation of Guaiacol Compounds. Acta Chem.

Scand. 1955, 9, 319–334.

(78) Lai, Y.-Z.; Guo, X.-P.; Situ, W. Estimation of Phenolic Hydroxyl Groups

in Wood by a Periodate Oxidation Method. J. Wood Chem. Technol.

1990, 10 (3), 365–377.

(79) Seth, R. S. Fibre Quality Factors in Papermaking-I. The Importance of

Fibre Lenght and Strength. Mat Res Soc Symp Proc, Mater. Res. Soc.

1990, 197, 125–141.

(80) Larsson, P. A.; Gimåker, M.; Wågberg, L. The Influence of Periodate

Oxidation on the Moisture Sorptivity and Dimensional Stability of Paper.

Cellulose 2008, 15 (6), 837–847.

(81) Zhang, S.; Fei, B.; Yu, Y.; Cheng, H.; Wang, C. Effect of the Amount of

Lignin on Tensile Properties of Single Wood Fibers. For. Sci. Pract.

2013, 15 (1), 56–60.

(82) Hosseinpourpia, R.; Adamopoulos, S.; Mai, C. Tensile Strength of

Handsheets from Recovered Fibers Treated with N-Methylol Melamine

and 1,3-Dimethylol-4,5-Dihydroxyethyleneurea. J. Appl. Polym. Sci.

2015, 132 (3), 41280.

(83) Veelaert, S.; De Wit, D.; Gotlieb, K. F.; Verhé, R. Chemical and

Physical Transitions of Periodate Oxidized Potato Starch in Water.

Carbohydr. Polym. 1997, 33 (97), 153–162.

(84) Potthast, A.; Schiehser, S.; Rosenau, T.; Kostic, M. Oxidative

Modifications of Cellulose in the Periodate System - Reduction and

Beta-Elimination Reactions: 2nd ICC 2007, Tokyo, Japan, October 25-

29, 2007. Holzforschung 2009, 63 (1), 12–17.

(85) Ward, K. Crystallinity of Cellulose and Its Significance for the Fiber

Properties. Text. Res. J. 1950, 20 (6), 363–372.

(86) Mihranyan, A.; Llagostera, A. P.; Karmhag, R.; Strømme, M.; Ek, R.

Moisture Sorption by Cellulose Powders of Varying Crystallinity. Int. J.

Pharm. 2004, 269 (2), 433–442.

67

(87) Manjunath, B. R.; Venkataraman, A.; Stephen, T. The Effect of Moisture

Present in Polymers on Their X-Ray Diffraction Patterns. J. Appl. Polym.

Sci. 1973, 17, 1091–1099.

(88) Agarwal, U. P.; Ralph, S. A.; Baez, C.; Reiner, R. S.; Verrill, S. P. Effect

of Sample Moisture Content on XRD-Estimated Cellulose Crystallinity

Index and Crystallite Size. Cellulose 2017, 24 (5), 1971–1984.

(89) Saito, T.; Hirota, M.; Tamura, N.; Kimura, S.; Fukuzumi, H.; Heux, L.;

Isogai, A. Individualization of Nano-Sized Plant Cellulose Fibrils

Achieved by Direct Surface Carboxylation Using TEMPO Catalyst.

Biomacromolecules 2009, 10, 1992–1996.

(90) Grignon, J.; Scallan, a. M. Effect of pH and Neutral Salts upon the

Swelling of Cellulose Gels. J. Appl. Polym. Sci. 1980, 25, 2829–2843.

(91) Lindström, T.; Carlsson, G. The Effect of Carboxyl Groups and Their

Ionic Form during Drying on the Hornification of Cellulose Fibers. Sven.

papperstidning 1982, 85 (15), R146–R151.

(92) Fall, A. B.; Lindström, S. B.; Sundman, O.; Ödberg, L.; Wågberg, L.

Colloidal Stability of Aqueous Nanofibrillated Cellulose Dispersions.

Langmuir 2011, 27 (18), 11332–11338.

(93) Gimåker, M.; Olsson, A.-M.; Salmén, L.; Wågberg, L. On the

Mechanisms of Mechano-Sorptive Creep Reduction by Chemical Cross-

Linking. In Advances in pulp and paper research,14th fundamental

research symposium, Oxford, UK; 2009; pp 1001–1017.

(94) Larsson, P. A.; Berglund, L. A.; Wågberg, L. Ductile All-Cellulose

Nanocomposite Films Fabricated from Core − Shell Structured Cellulose

Nanofibrils. Biomacromolecules 2014, 15, 2218–2223.

(95) Yang, Q.; Fukuzumi, H.; Saito, T.; Isogai, A.; Zhang, L. Transparent

Cellulose Films with High Gas Barrier Properties Fabricated from

Aqueous Alkali/urea Solutions. Biomacromolecules 2011, 12 (7), 2766–

2771.

(96) Eriksen, Ø.; Syverud, K.; Gregersen, Ø. The Use of Microfibrillated

68

Cellulose Produced from Kraft Pulp as Strength Enhancer in TMP Paper.

Nord. Pulp Pap. Res. J. 2008, 23 (3), 299–304.

(97) Sehaqui, H.; Zhou, Q.; Berglund, L. A. Nanofibrillated Cellulose for

Enhancement of Strength in High-Density Paper Structures. Nord. Pulp

Pap. Res. J. 2013, 28 (2), 182–189.

(98) Brodin, F. W.; Gregersen, Ø. W.; Syverud, K. Cellulose Nanofibrils:

Challenges and Possibilities as a Paper Additive or Coating Material - A

Review. Nord. Pulp Pap. Res. J. 2014, 29 (1), 156–166.

(99) Jackson, E. L.; Hudson, C. S. Application of the Cleavage Type of

Oxidation by Periodic Acid to Starch and Cellulose. J. Am. Chem. Soc.

1937, 59 (10), 2049–2050.

(100) Sehaqui, H.; Salajková, M.; Zhou, Q.; Berglund, L. A. Mechanical

Performance Tailoring of Tough Ultra-High Porosity Foams Prepared

from Cellulose I Nanofiber Suspensions. Soft Matter 2010, 6 (8), 1824.

(101) Ashby, M. F. The Properties of Foams and Lattices. Philos. Trans. R.

Soc. A Math. Phys. Eng. Sci. 2006, 364, 15–30.

(102) Liu, Y.; Lu, P.; Xiao, H.; Heydarifard, S.; Wang, S. Novel Aqueous

Spongy Foams Made of Three-Dimensionally Dispersed Wood-Fiber:

Entrapment and Stabilization with NFC/MFC within Capillary Foams.

Cellulose 2017, 24 (1), 241–251.

chemical modification of cellulose fibres and fibrils...

Documents